1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988-2016 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{} 1988-2016 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) 1988-2016 Free Software Foundation, Inc.
125 This edition of the GDB manual is dedicated to the memory of Fred
126 Fish. Fred was a long-standing contributor to GDB and to Free
127 software in general. We will miss him.
130 * Summary:: Summary of @value{GDBN}
131 * Sample Session:: A sample @value{GDBN} session
133 * Invocation:: Getting in and out of @value{GDBN}
134 * Commands:: @value{GDBN} commands
135 * Running:: Running programs under @value{GDBN}
136 * Stopping:: Stopping and continuing
137 * Reverse Execution:: Running programs backward
138 * Process Record and Replay:: Recording inferior's execution and replaying it
139 * Stack:: Examining the stack
140 * Source:: Examining source files
141 * Data:: Examining data
142 * Optimized Code:: Debugging optimized code
143 * Macros:: Preprocessor Macros
144 * Tracepoints:: Debugging remote targets non-intrusively
145 * Overlays:: Debugging programs that use overlays
147 * Languages:: Using @value{GDBN} with different languages
149 * Symbols:: Examining the symbol table
150 * Altering:: Altering execution
151 * GDB Files:: @value{GDBN} files
152 * Targets:: Specifying a debugging target
153 * Remote Debugging:: Debugging remote programs
154 * Configurations:: Configuration-specific information
155 * Controlling GDB:: Controlling @value{GDBN}
156 * Extending GDB:: Extending @value{GDBN}
157 * Interpreters:: Command Interpreters
158 * TUI:: @value{GDBN} Text User Interface
159 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
160 * GDB/MI:: @value{GDBN}'s Machine Interface.
161 * Annotations:: @value{GDBN}'s annotation interface.
162 * JIT Interface:: Using the JIT debugging interface.
163 * In-Process Agent:: In-Process Agent
165 * GDB Bugs:: Reporting bugs in @value{GDBN}
167 @ifset SYSTEM_READLINE
168 * Command Line Editing: (rluserman). Command Line Editing
169 * Using History Interactively: (history). Using History Interactively
171 @ifclear SYSTEM_READLINE
172 * Command Line Editing:: Command Line Editing
173 * Using History Interactively:: Using History Interactively
175 * In Memoriam:: In Memoriam
176 * Formatting Documentation:: How to format and print @value{GDBN} documentation
177 * Installing GDB:: Installing GDB
178 * Maintenance Commands:: Maintenance Commands
179 * Remote Protocol:: GDB Remote Serial Protocol
180 * Agent Expressions:: The GDB Agent Expression Mechanism
181 * Target Descriptions:: How targets can describe themselves to
183 * Operating System Information:: Getting additional information from
185 * Trace File Format:: GDB trace file format
186 * Index Section Format:: .gdb_index section format
187 * Man Pages:: Manual pages
188 * Copying:: GNU General Public License says
189 how you can copy and share GDB
190 * GNU Free Documentation License:: The license for this documentation
191 * Concept Index:: Index of @value{GDBN} concepts
192 * Command and Variable Index:: Index of @value{GDBN} commands, variables,
193 functions, and Python data types
201 @unnumbered Summary of @value{GDBN}
203 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
204 going on ``inside'' another program while it executes---or what another
205 program was doing at the moment it crashed.
207 @value{GDBN} can do four main kinds of things (plus other things in support of
208 these) to help you catch bugs in the act:
212 Start your program, specifying anything that might affect its behavior.
215 Make your program stop on specified conditions.
218 Examine what has happened, when your program has stopped.
221 Change things in your program, so you can experiment with correcting the
222 effects of one bug and go on to learn about another.
225 You can use @value{GDBN} to debug programs written in C and C@t{++}.
226 For more information, see @ref{Supported Languages,,Supported Languages}.
227 For more information, see @ref{C,,C and C++}.
229 Support for D is partial. For information on D, see
233 Support for Modula-2 is partial. For information on Modula-2, see
234 @ref{Modula-2,,Modula-2}.
236 Support for OpenCL C is partial. For information on OpenCL C, see
237 @ref{OpenCL C,,OpenCL C}.
240 Debugging Pascal programs which use sets, subranges, file variables, or
241 nested functions does not currently work. @value{GDBN} does not support
242 entering expressions, printing values, or similar features using Pascal
246 @value{GDBN} can be used to debug programs written in Fortran, although
247 it may be necessary to refer to some variables with a trailing
250 @value{GDBN} can be used to debug programs written in Objective-C,
251 using either the Apple/NeXT or the GNU Objective-C runtime.
254 * Free Software:: Freely redistributable software
255 * Free Documentation:: Free Software Needs Free Documentation
256 * Contributors:: Contributors to GDB
260 @unnumberedsec Free Software
262 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
263 General Public License
264 (GPL). The GPL gives you the freedom to copy or adapt a licensed
265 program---but every person getting a copy also gets with it the
266 freedom to modify that copy (which means that they must get access to
267 the source code), and the freedom to distribute further copies.
268 Typical software companies use copyrights to limit your freedoms; the
269 Free Software Foundation uses the GPL to preserve these freedoms.
271 Fundamentally, the General Public License is a license which says that
272 you have these freedoms and that you cannot take these freedoms away
275 @node Free Documentation
276 @unnumberedsec Free Software Needs Free Documentation
278 The biggest deficiency in the free software community today is not in
279 the software---it is the lack of good free documentation that we can
280 include with the free software. Many of our most important
281 programs do not come with free reference manuals and free introductory
282 texts. Documentation is an essential part of any software package;
283 when an important free software package does not come with a free
284 manual and a free tutorial, that is a major gap. We have many such
287 Consider Perl, for instance. The tutorial manuals that people
288 normally use are non-free. How did this come about? Because the
289 authors of those manuals published them with restrictive terms---no
290 copying, no modification, source files not available---which exclude
291 them from the free software world.
293 That wasn't the first time this sort of thing happened, and it was far
294 from the last. Many times we have heard a GNU user eagerly describe a
295 manual that he is writing, his intended contribution to the community,
296 only to learn that he had ruined everything by signing a publication
297 contract to make it non-free.
299 Free documentation, like free software, is a matter of freedom, not
300 price. The problem with the non-free manual is not that publishers
301 charge a price for printed copies---that in itself is fine. (The Free
302 Software Foundation sells printed copies of manuals, too.) The
303 problem is the restrictions on the use of the manual. Free manuals
304 are available in source code form, and give you permission to copy and
305 modify. Non-free manuals do not allow this.
307 The criteria of freedom for a free manual are roughly the same as for
308 free software. Redistribution (including the normal kinds of
309 commercial redistribution) must be permitted, so that the manual can
310 accompany every copy of the program, both on-line and on paper.
312 Permission for modification of the technical content is crucial too.
313 When people modify the software, adding or changing features, if they
314 are conscientious they will change the manual too---so they can
315 provide accurate and clear documentation for the modified program. A
316 manual that leaves you no choice but to write a new manual to document
317 a changed version of the program is not really available to our
320 Some kinds of limits on the way modification is handled are
321 acceptable. For example, requirements to preserve the original
322 author's copyright notice, the distribution terms, or the list of
323 authors, are ok. It is also no problem to require modified versions
324 to include notice that they were modified. Even entire sections that
325 may not be deleted or changed are acceptable, as long as they deal
326 with nontechnical topics (like this one). These kinds of restrictions
327 are acceptable because they don't obstruct the community's normal use
330 However, it must be possible to modify all the @emph{technical}
331 content of the manual, and then distribute the result in all the usual
332 media, through all the usual channels. Otherwise, the restrictions
333 obstruct the use of the manual, it is not free, and we need another
334 manual to replace it.
336 Please spread the word about this issue. Our community continues to
337 lose manuals to proprietary publishing. If we spread the word that
338 free software needs free reference manuals and free tutorials, perhaps
339 the next person who wants to contribute by writing documentation will
340 realize, before it is too late, that only free manuals contribute to
341 the free software community.
343 If you are writing documentation, please insist on publishing it under
344 the GNU Free Documentation License or another free documentation
345 license. Remember that this decision requires your approval---you
346 don't have to let the publisher decide. Some commercial publishers
347 will use a free license if you insist, but they will not propose the
348 option; it is up to you to raise the issue and say firmly that this is
349 what you want. If the publisher you are dealing with refuses, please
350 try other publishers. If you're not sure whether a proposed license
351 is free, write to @email{licensing@@gnu.org}.
353 You can encourage commercial publishers to sell more free, copylefted
354 manuals and tutorials by buying them, and particularly by buying
355 copies from the publishers that paid for their writing or for major
356 improvements. Meanwhile, try to avoid buying non-free documentation
357 at all. Check the distribution terms of a manual before you buy it,
358 and insist that whoever seeks your business must respect your freedom.
359 Check the history of the book, and try to reward the publishers that
360 have paid or pay the authors to work on it.
362 The Free Software Foundation maintains a list of free documentation
363 published by other publishers, at
364 @url{http://www.fsf.org/doc/other-free-books.html}.
367 @unnumberedsec Contributors to @value{GDBN}
369 Richard Stallman was the original author of @value{GDBN}, and of many
370 other @sc{gnu} programs. Many others have contributed to its
371 development. This section attempts to credit major contributors. One
372 of the virtues of free software is that everyone is free to contribute
373 to it; with regret, we cannot actually acknowledge everyone here. The
374 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
375 blow-by-blow account.
377 Changes much prior to version 2.0 are lost in the mists of time.
380 @emph{Plea:} Additions to this section are particularly welcome. If you
381 or your friends (or enemies, to be evenhanded) have been unfairly
382 omitted from this list, we would like to add your names!
385 So that they may not regard their many labors as thankless, we
386 particularly thank those who shepherded @value{GDBN} through major
388 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
389 Jim Blandy (release 4.18);
390 Jason Molenda (release 4.17);
391 Stan Shebs (release 4.14);
392 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
393 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
394 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
395 Jim Kingdon (releases 3.5, 3.4, and 3.3);
396 and Randy Smith (releases 3.2, 3.1, and 3.0).
398 Richard Stallman, assisted at various times by Peter TerMaat, Chris
399 Hanson, and Richard Mlynarik, handled releases through 2.8.
401 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
402 in @value{GDBN}, with significant additional contributions from Per
403 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
404 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
405 much general update work leading to release 3.0).
407 @value{GDBN} uses the BFD subroutine library to examine multiple
408 object-file formats; BFD was a joint project of David V.
409 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
411 David Johnson wrote the original COFF support; Pace Willison did
412 the original support for encapsulated COFF.
414 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
416 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
417 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
419 Jean-Daniel Fekete contributed Sun 386i support.
420 Chris Hanson improved the HP9000 support.
421 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
422 David Johnson contributed Encore Umax support.
423 Jyrki Kuoppala contributed Altos 3068 support.
424 Jeff Law contributed HP PA and SOM support.
425 Keith Packard contributed NS32K support.
426 Doug Rabson contributed Acorn Risc Machine support.
427 Bob Rusk contributed Harris Nighthawk CX-UX support.
428 Chris Smith contributed Convex support (and Fortran debugging).
429 Jonathan Stone contributed Pyramid support.
430 Michael Tiemann contributed SPARC support.
431 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
432 Pace Willison contributed Intel 386 support.
433 Jay Vosburgh contributed Symmetry support.
434 Marko Mlinar contributed OpenRISC 1000 support.
436 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
438 Rich Schaefer and Peter Schauer helped with support of SunOS shared
441 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
442 about several machine instruction sets.
444 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
445 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
446 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
447 and RDI targets, respectively.
449 Brian Fox is the author of the readline libraries providing
450 command-line editing and command history.
452 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
453 Modula-2 support, and contributed the Languages chapter of this manual.
455 Fred Fish wrote most of the support for Unix System Vr4.
456 He also enhanced the command-completion support to cover C@t{++} overloaded
459 Hitachi America (now Renesas America), Ltd. sponsored the support for
460 H8/300, H8/500, and Super-H processors.
462 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
464 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
467 Toshiba sponsored the support for the TX39 Mips processor.
469 Matsushita sponsored the support for the MN10200 and MN10300 processors.
471 Fujitsu sponsored the support for SPARClite and FR30 processors.
473 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
476 Michael Snyder added support for tracepoints.
478 Stu Grossman wrote gdbserver.
480 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
481 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
483 The following people at the Hewlett-Packard Company contributed
484 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
485 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
486 compiler, and the Text User Interface (nee Terminal User Interface):
487 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
488 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
489 provided HP-specific information in this manual.
491 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
492 Robert Hoehne made significant contributions to the DJGPP port.
494 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
495 development since 1991. Cygnus engineers who have worked on @value{GDBN}
496 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
497 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
498 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
499 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
500 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
501 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
502 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
503 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
504 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
505 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
506 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
507 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
508 Zuhn have made contributions both large and small.
510 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
511 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
513 Jim Blandy added support for preprocessor macros, while working for Red
516 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
517 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
518 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
519 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
520 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
521 with the migration of old architectures to this new framework.
523 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
524 unwinder framework, this consisting of a fresh new design featuring
525 frame IDs, independent frame sniffers, and the sentinel frame. Mark
526 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
527 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
528 trad unwinders. The architecture-specific changes, each involving a
529 complete rewrite of the architecture's frame code, were carried out by
530 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
531 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
532 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
533 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
536 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
537 Tensilica, Inc.@: contributed support for Xtensa processors. Others
538 who have worked on the Xtensa port of @value{GDBN} in the past include
539 Steve Tjiang, John Newlin, and Scott Foehner.
541 Michael Eager and staff of Xilinx, Inc., contributed support for the
542 Xilinx MicroBlaze architecture.
545 @chapter A Sample @value{GDBN} Session
547 You can use this manual at your leisure to read all about @value{GDBN}.
548 However, a handful of commands are enough to get started using the
549 debugger. This chapter illustrates those commands.
552 In this sample session, we emphasize user input like this: @b{input},
553 to make it easier to pick out from the surrounding output.
556 @c FIXME: this example may not be appropriate for some configs, where
557 @c FIXME...primary interest is in remote use.
559 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
560 processor) exhibits the following bug: sometimes, when we change its
561 quote strings from the default, the commands used to capture one macro
562 definition within another stop working. In the following short @code{m4}
563 session, we define a macro @code{foo} which expands to @code{0000}; we
564 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
565 same thing. However, when we change the open quote string to
566 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
567 procedure fails to define a new synonym @code{baz}:
576 @b{define(bar,defn(`foo'))}
580 @b{changequote(<QUOTE>,<UNQUOTE>)}
582 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
585 m4: End of input: 0: fatal error: EOF in string
589 Let us use @value{GDBN} to try to see what is going on.
592 $ @b{@value{GDBP} m4}
593 @c FIXME: this falsifies the exact text played out, to permit smallbook
594 @c FIXME... format to come out better.
595 @value{GDBN} is free software and you are welcome to distribute copies
596 of it under certain conditions; type "show copying" to see
598 There is absolutely no warranty for @value{GDBN}; type "show warranty"
601 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
606 @value{GDBN} reads only enough symbol data to know where to find the
607 rest when needed; as a result, the first prompt comes up very quickly.
608 We now tell @value{GDBN} to use a narrower display width than usual, so
609 that examples fit in this manual.
612 (@value{GDBP}) @b{set width 70}
616 We need to see how the @code{m4} built-in @code{changequote} works.
617 Having looked at the source, we know the relevant subroutine is
618 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
619 @code{break} command.
622 (@value{GDBP}) @b{break m4_changequote}
623 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
627 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
628 control; as long as control does not reach the @code{m4_changequote}
629 subroutine, the program runs as usual:
632 (@value{GDBP}) @b{run}
633 Starting program: /work/Editorial/gdb/gnu/m4/m4
641 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
642 suspends execution of @code{m4}, displaying information about the
643 context where it stops.
646 @b{changequote(<QUOTE>,<UNQUOTE>)}
648 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
650 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
654 Now we use the command @code{n} (@code{next}) to advance execution to
655 the next line of the current function.
659 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
664 @code{set_quotes} looks like a promising subroutine. We can go into it
665 by using the command @code{s} (@code{step}) instead of @code{next}.
666 @code{step} goes to the next line to be executed in @emph{any}
667 subroutine, so it steps into @code{set_quotes}.
671 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
673 530 if (lquote != def_lquote)
677 The display that shows the subroutine where @code{m4} is now
678 suspended (and its arguments) is called a stack frame display. It
679 shows a summary of the stack. We can use the @code{backtrace}
680 command (which can also be spelled @code{bt}), to see where we are
681 in the stack as a whole: the @code{backtrace} command displays a
682 stack frame for each active subroutine.
685 (@value{GDBP}) @b{bt}
686 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
688 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
690 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
691 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
693 #4 0x79dc in expand_input () at macro.c:40
694 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
698 We step through a few more lines to see what happens. The first two
699 times, we can use @samp{s}; the next two times we use @code{n} to avoid
700 falling into the @code{xstrdup} subroutine.
704 0x3b5c 532 if (rquote != def_rquote)
706 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
707 def_lquote : xstrdup(lq);
709 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
712 538 len_lquote = strlen(rquote);
716 The last line displayed looks a little odd; we can examine the variables
717 @code{lquote} and @code{rquote} to see if they are in fact the new left
718 and right quotes we specified. We use the command @code{p}
719 (@code{print}) to see their values.
722 (@value{GDBP}) @b{p lquote}
723 $1 = 0x35d40 "<QUOTE>"
724 (@value{GDBP}) @b{p rquote}
725 $2 = 0x35d50 "<UNQUOTE>"
729 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
730 To look at some context, we can display ten lines of source
731 surrounding the current line with the @code{l} (@code{list}) command.
737 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
739 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
742 538 len_lquote = strlen(rquote);
743 539 len_rquote = strlen(lquote);
750 Let us step past the two lines that set @code{len_lquote} and
751 @code{len_rquote}, and then examine the values of those variables.
755 539 len_rquote = strlen(lquote);
758 (@value{GDBP}) @b{p len_lquote}
760 (@value{GDBP}) @b{p len_rquote}
765 That certainly looks wrong, assuming @code{len_lquote} and
766 @code{len_rquote} are meant to be the lengths of @code{lquote} and
767 @code{rquote} respectively. We can set them to better values using
768 the @code{p} command, since it can print the value of
769 any expression---and that expression can include subroutine calls and
773 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
775 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
780 Is that enough to fix the problem of using the new quotes with the
781 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
782 executing with the @code{c} (@code{continue}) command, and then try the
783 example that caused trouble initially:
789 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
796 Success! The new quotes now work just as well as the default ones. The
797 problem seems to have been just the two typos defining the wrong
798 lengths. We allow @code{m4} exit by giving it an EOF as input:
802 Program exited normally.
806 The message @samp{Program exited normally.} is from @value{GDBN}; it
807 indicates @code{m4} has finished executing. We can end our @value{GDBN}
808 session with the @value{GDBN} @code{quit} command.
811 (@value{GDBP}) @b{quit}
815 @chapter Getting In and Out of @value{GDBN}
817 This chapter discusses how to start @value{GDBN}, and how to get out of it.
821 type @samp{@value{GDBP}} to start @value{GDBN}.
823 type @kbd{quit} or @kbd{Ctrl-d} to exit.
827 * Invoking GDB:: How to start @value{GDBN}
828 * Quitting GDB:: How to quit @value{GDBN}
829 * Shell Commands:: How to use shell commands inside @value{GDBN}
830 * Logging Output:: How to log @value{GDBN}'s output to a file
834 @section Invoking @value{GDBN}
836 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
837 @value{GDBN} reads commands from the terminal until you tell it to exit.
839 You can also run @code{@value{GDBP}} with a variety of arguments and options,
840 to specify more of your debugging environment at the outset.
842 The command-line options described here are designed
843 to cover a variety of situations; in some environments, some of these
844 options may effectively be unavailable.
846 The most usual way to start @value{GDBN} is with one argument,
847 specifying an executable program:
850 @value{GDBP} @var{program}
854 You can also start with both an executable program and a core file
858 @value{GDBP} @var{program} @var{core}
861 You can, instead, specify a process ID as a second argument, if you want
862 to debug a running process:
865 @value{GDBP} @var{program} 1234
869 would attach @value{GDBN} to process @code{1234} (unless you also have a file
870 named @file{1234}; @value{GDBN} does check for a core file first).
872 Taking advantage of the second command-line argument requires a fairly
873 complete operating system; when you use @value{GDBN} as a remote
874 debugger attached to a bare board, there may not be any notion of
875 ``process'', and there is often no way to get a core dump. @value{GDBN}
876 will warn you if it is unable to attach or to read core dumps.
878 You can optionally have @code{@value{GDBP}} pass any arguments after the
879 executable file to the inferior using @code{--args}. This option stops
882 @value{GDBP} --args gcc -O2 -c foo.c
884 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
885 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
887 You can run @code{@value{GDBP}} without printing the front material, which describes
888 @value{GDBN}'s non-warranty, by specifying @code{--silent}
889 (or @code{-q}/@code{--quiet}):
892 @value{GDBP} --silent
896 You can further control how @value{GDBN} starts up by using command-line
897 options. @value{GDBN} itself can remind you of the options available.
907 to display all available options and briefly describe their use
908 (@samp{@value{GDBP} -h} is a shorter equivalent).
910 All options and command line arguments you give are processed
911 in sequential order. The order makes a difference when the
912 @samp{-x} option is used.
916 * File Options:: Choosing files
917 * Mode Options:: Choosing modes
918 * Startup:: What @value{GDBN} does during startup
922 @subsection Choosing Files
924 When @value{GDBN} starts, it reads any arguments other than options as
925 specifying an executable file and core file (or process ID). This is
926 the same as if the arguments were specified by the @samp{-se} and
927 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
928 first argument that does not have an associated option flag as
929 equivalent to the @samp{-se} option followed by that argument; and the
930 second argument that does not have an associated option flag, if any, as
931 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
932 If the second argument begins with a decimal digit, @value{GDBN} will
933 first attempt to attach to it as a process, and if that fails, attempt
934 to open it as a corefile. If you have a corefile whose name begins with
935 a digit, you can prevent @value{GDBN} from treating it as a pid by
936 prefixing it with @file{./}, e.g.@: @file{./12345}.
938 If @value{GDBN} has not been configured to included core file support,
939 such as for most embedded targets, then it will complain about a second
940 argument and ignore it.
942 Many options have both long and short forms; both are shown in the
943 following list. @value{GDBN} also recognizes the long forms if you truncate
944 them, so long as enough of the option is present to be unambiguous.
945 (If you prefer, you can flag option arguments with @samp{--} rather
946 than @samp{-}, though we illustrate the more usual convention.)
948 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
949 @c way, both those who look for -foo and --foo in the index, will find
953 @item -symbols @var{file}
955 @cindex @code{--symbols}
957 Read symbol table from file @var{file}.
959 @item -exec @var{file}
961 @cindex @code{--exec}
963 Use file @var{file} as the executable file to execute when appropriate,
964 and for examining pure data in conjunction with a core dump.
968 Read symbol table from file @var{file} and use it as the executable
971 @item -core @var{file}
973 @cindex @code{--core}
975 Use file @var{file} as a core dump to examine.
977 @item -pid @var{number}
978 @itemx -p @var{number}
981 Connect to process ID @var{number}, as with the @code{attach} command.
983 @item -command @var{file}
985 @cindex @code{--command}
987 Execute commands from file @var{file}. The contents of this file is
988 evaluated exactly as the @code{source} command would.
989 @xref{Command Files,, Command files}.
991 @item -eval-command @var{command}
992 @itemx -ex @var{command}
993 @cindex @code{--eval-command}
995 Execute a single @value{GDBN} command.
997 This option may be used multiple times to call multiple commands. It may
998 also be interleaved with @samp{-command} as required.
1001 @value{GDBP} -ex 'target sim' -ex 'load' \
1002 -x setbreakpoints -ex 'run' a.out
1005 @item -init-command @var{file}
1006 @itemx -ix @var{file}
1007 @cindex @code{--init-command}
1009 Execute commands from file @var{file} before loading the inferior (but
1010 after loading gdbinit files).
1013 @item -init-eval-command @var{command}
1014 @itemx -iex @var{command}
1015 @cindex @code{--init-eval-command}
1017 Execute a single @value{GDBN} command before loading the inferior (but
1018 after loading gdbinit files).
1021 @item -directory @var{directory}
1022 @itemx -d @var{directory}
1023 @cindex @code{--directory}
1025 Add @var{directory} to the path to search for source and script files.
1029 @cindex @code{--readnow}
1031 Read each symbol file's entire symbol table immediately, rather than
1032 the default, which is to read it incrementally as it is needed.
1033 This makes startup slower, but makes future operations faster.
1038 @subsection Choosing Modes
1040 You can run @value{GDBN} in various alternative modes---for example, in
1041 batch mode or quiet mode.
1049 Do not execute commands found in any initialization file.
1050 There are three init files, loaded in the following order:
1053 @item @file{system.gdbinit}
1054 This is the system-wide init file.
1055 Its location is specified with the @code{--with-system-gdbinit}
1056 configure option (@pxref{System-wide configuration}).
1057 It is loaded first when @value{GDBN} starts, before command line options
1058 have been processed.
1059 @item @file{~/.gdbinit}
1060 This is the init file in your home directory.
1061 It is loaded next, after @file{system.gdbinit}, and before
1062 command options have been processed.
1063 @item @file{./.gdbinit}
1064 This is the init file in the current directory.
1065 It is loaded last, after command line options other than @code{-x} and
1066 @code{-ex} have been processed. Command line options @code{-x} and
1067 @code{-ex} are processed last, after @file{./.gdbinit} has been loaded.
1070 For further documentation on startup processing, @xref{Startup}.
1071 For documentation on how to write command files,
1072 @xref{Command Files,,Command Files}.
1077 Do not execute commands found in @file{~/.gdbinit}, the init file
1078 in your home directory.
1084 @cindex @code{--quiet}
1085 @cindex @code{--silent}
1087 ``Quiet''. Do not print the introductory and copyright messages. These
1088 messages are also suppressed in batch mode.
1091 @cindex @code{--batch}
1092 Run in batch mode. Exit with status @code{0} after processing all the
1093 command files specified with @samp{-x} (and all commands from
1094 initialization files, if not inhibited with @samp{-n}). Exit with
1095 nonzero status if an error occurs in executing the @value{GDBN} commands
1096 in the command files. Batch mode also disables pagination, sets unlimited
1097 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1098 off} were in effect (@pxref{Messages/Warnings}).
1100 Batch mode may be useful for running @value{GDBN} as a filter, for
1101 example to download and run a program on another computer; in order to
1102 make this more useful, the message
1105 Program exited normally.
1109 (which is ordinarily issued whenever a program running under
1110 @value{GDBN} control terminates) is not issued when running in batch
1114 @cindex @code{--batch-silent}
1115 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1116 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1117 unaffected). This is much quieter than @samp{-silent} and would be useless
1118 for an interactive session.
1120 This is particularly useful when using targets that give @samp{Loading section}
1121 messages, for example.
1123 Note that targets that give their output via @value{GDBN}, as opposed to
1124 writing directly to @code{stdout}, will also be made silent.
1126 @item -return-child-result
1127 @cindex @code{--return-child-result}
1128 The return code from @value{GDBN} will be the return code from the child
1129 process (the process being debugged), with the following exceptions:
1133 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1134 internal error. In this case the exit code is the same as it would have been
1135 without @samp{-return-child-result}.
1137 The user quits with an explicit value. E.g., @samp{quit 1}.
1139 The child process never runs, or is not allowed to terminate, in which case
1140 the exit code will be -1.
1143 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1144 when @value{GDBN} is being used as a remote program loader or simulator
1149 @cindex @code{--nowindows}
1151 ``No windows''. If @value{GDBN} comes with a graphical user interface
1152 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1153 interface. If no GUI is available, this option has no effect.
1157 @cindex @code{--windows}
1159 If @value{GDBN} includes a GUI, then this option requires it to be
1162 @item -cd @var{directory}
1164 Run @value{GDBN} using @var{directory} as its working directory,
1165 instead of the current directory.
1167 @item -data-directory @var{directory}
1168 @itemx -D @var{directory}
1169 @cindex @code{--data-directory}
1171 Run @value{GDBN} using @var{directory} as its data directory.
1172 The data directory is where @value{GDBN} searches for its
1173 auxiliary files. @xref{Data Files}.
1177 @cindex @code{--fullname}
1179 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1180 subprocess. It tells @value{GDBN} to output the full file name and line
1181 number in a standard, recognizable fashion each time a stack frame is
1182 displayed (which includes each time your program stops). This
1183 recognizable format looks like two @samp{\032} characters, followed by
1184 the file name, line number and character position separated by colons,
1185 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1186 @samp{\032} characters as a signal to display the source code for the
1189 @item -annotate @var{level}
1190 @cindex @code{--annotate}
1191 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1192 effect is identical to using @samp{set annotate @var{level}}
1193 (@pxref{Annotations}). The annotation @var{level} controls how much
1194 information @value{GDBN} prints together with its prompt, values of
1195 expressions, source lines, and other types of output. Level 0 is the
1196 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1197 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1198 that control @value{GDBN}, and level 2 has been deprecated.
1200 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1204 @cindex @code{--args}
1205 Change interpretation of command line so that arguments following the
1206 executable file are passed as command line arguments to the inferior.
1207 This option stops option processing.
1209 @item -baud @var{bps}
1211 @cindex @code{--baud}
1213 Set the line speed (baud rate or bits per second) of any serial
1214 interface used by @value{GDBN} for remote debugging.
1216 @item -l @var{timeout}
1218 Set the timeout (in seconds) of any communication used by @value{GDBN}
1219 for remote debugging.
1221 @item -tty @var{device}
1222 @itemx -t @var{device}
1223 @cindex @code{--tty}
1225 Run using @var{device} for your program's standard input and output.
1226 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1228 @c resolve the situation of these eventually
1230 @cindex @code{--tui}
1231 Activate the @dfn{Text User Interface} when starting. The Text User
1232 Interface manages several text windows on the terminal, showing
1233 source, assembly, registers and @value{GDBN} command outputs
1234 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1235 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1236 Using @value{GDBN} under @sc{gnu} Emacs}).
1238 @item -interpreter @var{interp}
1239 @cindex @code{--interpreter}
1240 Use the interpreter @var{interp} for interface with the controlling
1241 program or device. This option is meant to be set by programs which
1242 communicate with @value{GDBN} using it as a back end.
1243 @xref{Interpreters, , Command Interpreters}.
1245 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1246 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1247 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1248 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1249 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1250 @sc{gdb/mi} interfaces are no longer supported.
1253 @cindex @code{--write}
1254 Open the executable and core files for both reading and writing. This
1255 is equivalent to the @samp{set write on} command inside @value{GDBN}
1259 @cindex @code{--statistics}
1260 This option causes @value{GDBN} to print statistics about time and
1261 memory usage after it completes each command and returns to the prompt.
1264 @cindex @code{--version}
1265 This option causes @value{GDBN} to print its version number and
1266 no-warranty blurb, and exit.
1268 @item -configuration
1269 @cindex @code{--configuration}
1270 This option causes @value{GDBN} to print details about its build-time
1271 configuration parameters, and then exit. These details can be
1272 important when reporting @value{GDBN} bugs (@pxref{GDB Bugs}).
1277 @subsection What @value{GDBN} Does During Startup
1278 @cindex @value{GDBN} startup
1280 Here's the description of what @value{GDBN} does during session startup:
1284 Sets up the command interpreter as specified by the command line
1285 (@pxref{Mode Options, interpreter}).
1289 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1290 used when building @value{GDBN}; @pxref{System-wide configuration,
1291 ,System-wide configuration and settings}) and executes all the commands in
1294 @anchor{Home Directory Init File}
1296 Reads the init file (if any) in your home directory@footnote{On
1297 DOS/Windows systems, the home directory is the one pointed to by the
1298 @code{HOME} environment variable.} and executes all the commands in
1301 @anchor{Option -init-eval-command}
1303 Executes commands and command files specified by the @samp{-iex} and
1304 @samp{-ix} options in their specified order. Usually you should use the
1305 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1306 settings before @value{GDBN} init files get executed and before inferior
1310 Processes command line options and operands.
1312 @anchor{Init File in the Current Directory during Startup}
1314 Reads and executes the commands from init file (if any) in the current
1315 working directory as long as @samp{set auto-load local-gdbinit} is set to
1316 @samp{on} (@pxref{Init File in the Current Directory}).
1317 This is only done if the current directory is
1318 different from your home directory. Thus, you can have more than one
1319 init file, one generic in your home directory, and another, specific
1320 to the program you are debugging, in the directory where you invoke
1324 If the command line specified a program to debug, or a process to
1325 attach to, or a core file, @value{GDBN} loads any auto-loaded
1326 scripts provided for the program or for its loaded shared libraries.
1327 @xref{Auto-loading}.
1329 If you wish to disable the auto-loading during startup,
1330 you must do something like the following:
1333 $ gdb -iex "set auto-load python-scripts off" myprogram
1336 Option @samp{-ex} does not work because the auto-loading is then turned
1340 Executes commands and command files specified by the @samp{-ex} and
1341 @samp{-x} options in their specified order. @xref{Command Files}, for
1342 more details about @value{GDBN} command files.
1345 Reads the command history recorded in the @dfn{history file}.
1346 @xref{Command History}, for more details about the command history and the
1347 files where @value{GDBN} records it.
1350 Init files use the same syntax as @dfn{command files} (@pxref{Command
1351 Files}) and are processed by @value{GDBN} in the same way. The init
1352 file in your home directory can set options (such as @samp{set
1353 complaints}) that affect subsequent processing of command line options
1354 and operands. Init files are not executed if you use the @samp{-nx}
1355 option (@pxref{Mode Options, ,Choosing Modes}).
1357 To display the list of init files loaded by gdb at startup, you
1358 can use @kbd{gdb --help}.
1360 @cindex init file name
1361 @cindex @file{.gdbinit}
1362 @cindex @file{gdb.ini}
1363 The @value{GDBN} init files are normally called @file{.gdbinit}.
1364 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1365 the limitations of file names imposed by DOS filesystems. The Windows
1366 port of @value{GDBN} uses the standard name, but if it finds a
1367 @file{gdb.ini} file in your home directory, it warns you about that
1368 and suggests to rename the file to the standard name.
1372 @section Quitting @value{GDBN}
1373 @cindex exiting @value{GDBN}
1374 @cindex leaving @value{GDBN}
1377 @kindex quit @r{[}@var{expression}@r{]}
1378 @kindex q @r{(@code{quit})}
1379 @item quit @r{[}@var{expression}@r{]}
1381 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1382 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1383 do not supply @var{expression}, @value{GDBN} will terminate normally;
1384 otherwise it will terminate using the result of @var{expression} as the
1389 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1390 terminates the action of any @value{GDBN} command that is in progress and
1391 returns to @value{GDBN} command level. It is safe to type the interrupt
1392 character at any time because @value{GDBN} does not allow it to take effect
1393 until a time when it is safe.
1395 If you have been using @value{GDBN} to control an attached process or
1396 device, you can release it with the @code{detach} command
1397 (@pxref{Attach, ,Debugging an Already-running Process}).
1399 @node Shell Commands
1400 @section Shell Commands
1402 If you need to execute occasional shell commands during your
1403 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1404 just use the @code{shell} command.
1409 @cindex shell escape
1410 @item shell @var{command-string}
1411 @itemx !@var{command-string}
1412 Invoke a standard shell to execute @var{command-string}.
1413 Note that no space is needed between @code{!} and @var{command-string}.
1414 If it exists, the environment variable @code{SHELL} determines which
1415 shell to run. Otherwise @value{GDBN} uses the default shell
1416 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1419 The utility @code{make} is often needed in development environments.
1420 You do not have to use the @code{shell} command for this purpose in
1425 @cindex calling make
1426 @item make @var{make-args}
1427 Execute the @code{make} program with the specified
1428 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1431 @node Logging Output
1432 @section Logging Output
1433 @cindex logging @value{GDBN} output
1434 @cindex save @value{GDBN} output to a file
1436 You may want to save the output of @value{GDBN} commands to a file.
1437 There are several commands to control @value{GDBN}'s logging.
1441 @item set logging on
1443 @item set logging off
1445 @cindex logging file name
1446 @item set logging file @var{file}
1447 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1448 @item set logging overwrite [on|off]
1449 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1450 you want @code{set logging on} to overwrite the logfile instead.
1451 @item set logging redirect [on|off]
1452 By default, @value{GDBN} output will go to both the terminal and the logfile.
1453 Set @code{redirect} if you want output to go only to the log file.
1454 @kindex show logging
1456 Show the current values of the logging settings.
1460 @chapter @value{GDBN} Commands
1462 You can abbreviate a @value{GDBN} command to the first few letters of the command
1463 name, if that abbreviation is unambiguous; and you can repeat certain
1464 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1465 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1466 show you the alternatives available, if there is more than one possibility).
1469 * Command Syntax:: How to give commands to @value{GDBN}
1470 * Completion:: Command completion
1471 * Help:: How to ask @value{GDBN} for help
1474 @node Command Syntax
1475 @section Command Syntax
1477 A @value{GDBN} command is a single line of input. There is no limit on
1478 how long it can be. It starts with a command name, which is followed by
1479 arguments whose meaning depends on the command name. For example, the
1480 command @code{step} accepts an argument which is the number of times to
1481 step, as in @samp{step 5}. You can also use the @code{step} command
1482 with no arguments. Some commands do not allow any arguments.
1484 @cindex abbreviation
1485 @value{GDBN} command names may always be truncated if that abbreviation is
1486 unambiguous. Other possible command abbreviations are listed in the
1487 documentation for individual commands. In some cases, even ambiguous
1488 abbreviations are allowed; for example, @code{s} is specially defined as
1489 equivalent to @code{step} even though there are other commands whose
1490 names start with @code{s}. You can test abbreviations by using them as
1491 arguments to the @code{help} command.
1493 @cindex repeating commands
1494 @kindex RET @r{(repeat last command)}
1495 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1496 repeat the previous command. Certain commands (for example, @code{run})
1497 will not repeat this way; these are commands whose unintentional
1498 repetition might cause trouble and which you are unlikely to want to
1499 repeat. User-defined commands can disable this feature; see
1500 @ref{Define, dont-repeat}.
1502 The @code{list} and @code{x} commands, when you repeat them with
1503 @key{RET}, construct new arguments rather than repeating
1504 exactly as typed. This permits easy scanning of source or memory.
1506 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1507 output, in a way similar to the common utility @code{more}
1508 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1509 @key{RET} too many in this situation, @value{GDBN} disables command
1510 repetition after any command that generates this sort of display.
1512 @kindex # @r{(a comment)}
1514 Any text from a @kbd{#} to the end of the line is a comment; it does
1515 nothing. This is useful mainly in command files (@pxref{Command
1516 Files,,Command Files}).
1518 @cindex repeating command sequences
1519 @kindex Ctrl-o @r{(operate-and-get-next)}
1520 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1521 commands. This command accepts the current line, like @key{RET}, and
1522 then fetches the next line relative to the current line from the history
1526 @section Command Completion
1529 @cindex word completion
1530 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1531 only one possibility; it can also show you what the valid possibilities
1532 are for the next word in a command, at any time. This works for @value{GDBN}
1533 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1535 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1536 of a word. If there is only one possibility, @value{GDBN} fills in the
1537 word, and waits for you to finish the command (or press @key{RET} to
1538 enter it). For example, if you type
1540 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1541 @c complete accuracy in these examples; space introduced for clarity.
1542 @c If texinfo enhancements make it unnecessary, it would be nice to
1543 @c replace " @key" by "@key" in the following...
1545 (@value{GDBP}) info bre @key{TAB}
1549 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1550 the only @code{info} subcommand beginning with @samp{bre}:
1553 (@value{GDBP}) info breakpoints
1557 You can either press @key{RET} at this point, to run the @code{info
1558 breakpoints} command, or backspace and enter something else, if
1559 @samp{breakpoints} does not look like the command you expected. (If you
1560 were sure you wanted @code{info breakpoints} in the first place, you
1561 might as well just type @key{RET} immediately after @samp{info bre},
1562 to exploit command abbreviations rather than command completion).
1564 If there is more than one possibility for the next word when you press
1565 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1566 characters and try again, or just press @key{TAB} a second time;
1567 @value{GDBN} displays all the possible completions for that word. For
1568 example, you might want to set a breakpoint on a subroutine whose name
1569 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1570 just sounds the bell. Typing @key{TAB} again displays all the
1571 function names in your program that begin with those characters, for
1575 (@value{GDBP}) b make_ @key{TAB}
1576 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1577 make_a_section_from_file make_environ
1578 make_abs_section make_function_type
1579 make_blockvector make_pointer_type
1580 make_cleanup make_reference_type
1581 make_command make_symbol_completion_list
1582 (@value{GDBP}) b make_
1586 After displaying the available possibilities, @value{GDBN} copies your
1587 partial input (@samp{b make_} in the example) so you can finish the
1590 If you just want to see the list of alternatives in the first place, you
1591 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1592 means @kbd{@key{META} ?}. You can type this either by holding down a
1593 key designated as the @key{META} shift on your keyboard (if there is
1594 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1596 If the number of possible completions is large, @value{GDBN} will
1597 print as much of the list as it has collected, as well as a message
1598 indicating that the list may be truncated.
1601 (@value{GDBP}) b m@key{TAB}@key{TAB}
1603 <... the rest of the possible completions ...>
1604 *** List may be truncated, max-completions reached. ***
1609 This behavior can be controlled with the following commands:
1612 @kindex set max-completions
1613 @item set max-completions @var{limit}
1614 @itemx set max-completions unlimited
1615 Set the maximum number of completion candidates. @value{GDBN} will
1616 stop looking for more completions once it collects this many candidates.
1617 This is useful when completing on things like function names as collecting
1618 all the possible candidates can be time consuming.
1619 The default value is 200. A value of zero disables tab-completion.
1620 Note that setting either no limit or a very large limit can make
1622 @kindex show max-completions
1623 @item show max-completions
1624 Show the maximum number of candidates that @value{GDBN} will collect and show
1628 @cindex quotes in commands
1629 @cindex completion of quoted strings
1630 Sometimes the string you need, while logically a ``word'', may contain
1631 parentheses or other characters that @value{GDBN} normally excludes from
1632 its notion of a word. To permit word completion to work in this
1633 situation, you may enclose words in @code{'} (single quote marks) in
1634 @value{GDBN} commands.
1636 The most likely situation where you might need this is in typing the
1637 name of a C@t{++} function. This is because C@t{++} allows function
1638 overloading (multiple definitions of the same function, distinguished
1639 by argument type). For example, when you want to set a breakpoint you
1640 may need to distinguish whether you mean the version of @code{name}
1641 that takes an @code{int} parameter, @code{name(int)}, or the version
1642 that takes a @code{float} parameter, @code{name(float)}. To use the
1643 word-completion facilities in this situation, type a single quote
1644 @code{'} at the beginning of the function name. This alerts
1645 @value{GDBN} that it may need to consider more information than usual
1646 when you press @key{TAB} or @kbd{M-?} to request word completion:
1649 (@value{GDBP}) b 'bubble( @kbd{M-?}
1650 bubble(double,double) bubble(int,int)
1651 (@value{GDBP}) b 'bubble(
1654 In some cases, @value{GDBN} can tell that completing a name requires using
1655 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1656 completing as much as it can) if you do not type the quote in the first
1660 (@value{GDBP}) b bub @key{TAB}
1661 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1662 (@value{GDBP}) b 'bubble(
1666 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1667 you have not yet started typing the argument list when you ask for
1668 completion on an overloaded symbol.
1670 For more information about overloaded functions, see @ref{C Plus Plus
1671 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1672 overload-resolution off} to disable overload resolution;
1673 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1675 @cindex completion of structure field names
1676 @cindex structure field name completion
1677 @cindex completion of union field names
1678 @cindex union field name completion
1679 When completing in an expression which looks up a field in a
1680 structure, @value{GDBN} also tries@footnote{The completer can be
1681 confused by certain kinds of invalid expressions. Also, it only
1682 examines the static type of the expression, not the dynamic type.} to
1683 limit completions to the field names available in the type of the
1687 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1688 magic to_fputs to_rewind
1689 to_data to_isatty to_write
1690 to_delete to_put to_write_async_safe
1695 This is because the @code{gdb_stdout} is a variable of the type
1696 @code{struct ui_file} that is defined in @value{GDBN} sources as
1703 ui_file_flush_ftype *to_flush;
1704 ui_file_write_ftype *to_write;
1705 ui_file_write_async_safe_ftype *to_write_async_safe;
1706 ui_file_fputs_ftype *to_fputs;
1707 ui_file_read_ftype *to_read;
1708 ui_file_delete_ftype *to_delete;
1709 ui_file_isatty_ftype *to_isatty;
1710 ui_file_rewind_ftype *to_rewind;
1711 ui_file_put_ftype *to_put;
1718 @section Getting Help
1719 @cindex online documentation
1722 You can always ask @value{GDBN} itself for information on its commands,
1723 using the command @code{help}.
1726 @kindex h @r{(@code{help})}
1729 You can use @code{help} (abbreviated @code{h}) with no arguments to
1730 display a short list of named classes of commands:
1734 List of classes of commands:
1736 aliases -- Aliases of other commands
1737 breakpoints -- Making program stop at certain points
1738 data -- Examining data
1739 files -- Specifying and examining files
1740 internals -- Maintenance commands
1741 obscure -- Obscure features
1742 running -- Running the program
1743 stack -- Examining the stack
1744 status -- Status inquiries
1745 support -- Support facilities
1746 tracepoints -- Tracing of program execution without
1747 stopping the program
1748 user-defined -- User-defined commands
1750 Type "help" followed by a class name for a list of
1751 commands in that class.
1752 Type "help" followed by command name for full
1754 Command name abbreviations are allowed if unambiguous.
1757 @c the above line break eliminates huge line overfull...
1759 @item help @var{class}
1760 Using one of the general help classes as an argument, you can get a
1761 list of the individual commands in that class. For example, here is the
1762 help display for the class @code{status}:
1765 (@value{GDBP}) help status
1770 @c Line break in "show" line falsifies real output, but needed
1771 @c to fit in smallbook page size.
1772 info -- Generic command for showing things
1773 about the program being debugged
1774 show -- Generic command for showing things
1777 Type "help" followed by command name for full
1779 Command name abbreviations are allowed if unambiguous.
1783 @item help @var{command}
1784 With a command name as @code{help} argument, @value{GDBN} displays a
1785 short paragraph on how to use that command.
1788 @item apropos @var{args}
1789 The @code{apropos} command searches through all of the @value{GDBN}
1790 commands, and their documentation, for the regular expression specified in
1791 @var{args}. It prints out all matches found. For example:
1802 alias -- Define a new command that is an alias of an existing command
1803 aliases -- Aliases of other commands
1804 d -- Delete some breakpoints or auto-display expressions
1805 del -- Delete some breakpoints or auto-display expressions
1806 delete -- Delete some breakpoints or auto-display expressions
1811 @item complete @var{args}
1812 The @code{complete @var{args}} command lists all the possible completions
1813 for the beginning of a command. Use @var{args} to specify the beginning of the
1814 command you want completed. For example:
1820 @noindent results in:
1831 @noindent This is intended for use by @sc{gnu} Emacs.
1834 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1835 and @code{show} to inquire about the state of your program, or the state
1836 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1837 manual introduces each of them in the appropriate context. The listings
1838 under @code{info} and under @code{show} in the Command, Variable, and
1839 Function Index point to all the sub-commands. @xref{Command and Variable
1845 @kindex i @r{(@code{info})}
1847 This command (abbreviated @code{i}) is for describing the state of your
1848 program. For example, you can show the arguments passed to a function
1849 with @code{info args}, list the registers currently in use with @code{info
1850 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1851 You can get a complete list of the @code{info} sub-commands with
1852 @w{@code{help info}}.
1856 You can assign the result of an expression to an environment variable with
1857 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1858 @code{set prompt $}.
1862 In contrast to @code{info}, @code{show} is for describing the state of
1863 @value{GDBN} itself.
1864 You can change most of the things you can @code{show}, by using the
1865 related command @code{set}; for example, you can control what number
1866 system is used for displays with @code{set radix}, or simply inquire
1867 which is currently in use with @code{show radix}.
1870 To display all the settable parameters and their current
1871 values, you can use @code{show} with no arguments; you may also use
1872 @code{info set}. Both commands produce the same display.
1873 @c FIXME: "info set" violates the rule that "info" is for state of
1874 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1875 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1879 Here are several miscellaneous @code{show} subcommands, all of which are
1880 exceptional in lacking corresponding @code{set} commands:
1883 @kindex show version
1884 @cindex @value{GDBN} version number
1886 Show what version of @value{GDBN} is running. You should include this
1887 information in @value{GDBN} bug-reports. If multiple versions of
1888 @value{GDBN} are in use at your site, you may need to determine which
1889 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1890 commands are introduced, and old ones may wither away. Also, many
1891 system vendors ship variant versions of @value{GDBN}, and there are
1892 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1893 The version number is the same as the one announced when you start
1896 @kindex show copying
1897 @kindex info copying
1898 @cindex display @value{GDBN} copyright
1901 Display information about permission for copying @value{GDBN}.
1903 @kindex show warranty
1904 @kindex info warranty
1906 @itemx info warranty
1907 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1908 if your version of @value{GDBN} comes with one.
1910 @kindex show configuration
1911 @item show configuration
1912 Display detailed information about the way @value{GDBN} was configured
1913 when it was built. This displays the optional arguments passed to the
1914 @file{configure} script and also configuration parameters detected
1915 automatically by @command{configure}. When reporting a @value{GDBN}
1916 bug (@pxref{GDB Bugs}), it is important to include this information in
1922 @chapter Running Programs Under @value{GDBN}
1924 When you run a program under @value{GDBN}, you must first generate
1925 debugging information when you compile it.
1927 You may start @value{GDBN} with its arguments, if any, in an environment
1928 of your choice. If you are doing native debugging, you may redirect
1929 your program's input and output, debug an already running process, or
1930 kill a child process.
1933 * Compilation:: Compiling for debugging
1934 * Starting:: Starting your program
1935 * Arguments:: Your program's arguments
1936 * Environment:: Your program's environment
1938 * Working Directory:: Your program's working directory
1939 * Input/Output:: Your program's input and output
1940 * Attach:: Debugging an already-running process
1941 * Kill Process:: Killing the child process
1943 * Inferiors and Programs:: Debugging multiple inferiors and programs
1944 * Threads:: Debugging programs with multiple threads
1945 * Forks:: Debugging forks
1946 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1950 @section Compiling for Debugging
1952 In order to debug a program effectively, you need to generate
1953 debugging information when you compile it. This debugging information
1954 is stored in the object file; it describes the data type of each
1955 variable or function and the correspondence between source line numbers
1956 and addresses in the executable code.
1958 To request debugging information, specify the @samp{-g} option when you run
1961 Programs that are to be shipped to your customers are compiled with
1962 optimizations, using the @samp{-O} compiler option. However, some
1963 compilers are unable to handle the @samp{-g} and @samp{-O} options
1964 together. Using those compilers, you cannot generate optimized
1965 executables containing debugging information.
1967 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1968 without @samp{-O}, making it possible to debug optimized code. We
1969 recommend that you @emph{always} use @samp{-g} whenever you compile a
1970 program. You may think your program is correct, but there is no sense
1971 in pushing your luck. For more information, see @ref{Optimized Code}.
1973 Older versions of the @sc{gnu} C compiler permitted a variant option
1974 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1975 format; if your @sc{gnu} C compiler has this option, do not use it.
1977 @value{GDBN} knows about preprocessor macros and can show you their
1978 expansion (@pxref{Macros}). Most compilers do not include information
1979 about preprocessor macros in the debugging information if you specify
1980 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
1981 the @sc{gnu} C compiler, provides macro information if you are using
1982 the DWARF debugging format, and specify the option @option{-g3}.
1984 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
1985 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}, for more
1986 information on @value{NGCC} options affecting debug information.
1988 You will have the best debugging experience if you use the latest
1989 version of the DWARF debugging format that your compiler supports.
1990 DWARF is currently the most expressive and best supported debugging
1991 format in @value{GDBN}.
1995 @section Starting your Program
2001 @kindex r @r{(@code{run})}
2004 Use the @code{run} command to start your program under @value{GDBN}.
2005 You must first specify the program name with an argument to
2006 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
2007 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
2008 command (@pxref{Files, ,Commands to Specify Files}).
2012 If you are running your program in an execution environment that
2013 supports processes, @code{run} creates an inferior process and makes
2014 that process run your program. In some environments without processes,
2015 @code{run} jumps to the start of your program. Other targets,
2016 like @samp{remote}, are always running. If you get an error
2017 message like this one:
2020 The "remote" target does not support "run".
2021 Try "help target" or "continue".
2025 then use @code{continue} to run your program. You may need @code{load}
2026 first (@pxref{load}).
2028 The execution of a program is affected by certain information it
2029 receives from its superior. @value{GDBN} provides ways to specify this
2030 information, which you must do @emph{before} starting your program. (You
2031 can change it after starting your program, but such changes only affect
2032 your program the next time you start it.) This information may be
2033 divided into four categories:
2036 @item The @emph{arguments.}
2037 Specify the arguments to give your program as the arguments of the
2038 @code{run} command. If a shell is available on your target, the shell
2039 is used to pass the arguments, so that you may use normal conventions
2040 (such as wildcard expansion or variable substitution) in describing
2042 In Unix systems, you can control which shell is used with the
2043 @code{SHELL} environment variable. If you do not define @code{SHELL},
2044 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2045 use of any shell with the @code{set startup-with-shell} command (see
2048 @item The @emph{environment.}
2049 Your program normally inherits its environment from @value{GDBN}, but you can
2050 use the @value{GDBN} commands @code{set environment} and @code{unset
2051 environment} to change parts of the environment that affect
2052 your program. @xref{Environment, ,Your Program's Environment}.
2054 @item The @emph{working directory.}
2055 Your program inherits its working directory from @value{GDBN}. You can set
2056 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
2057 @xref{Working Directory, ,Your Program's Working Directory}.
2059 @item The @emph{standard input and output.}
2060 Your program normally uses the same device for standard input and
2061 standard output as @value{GDBN} is using. You can redirect input and output
2062 in the @code{run} command line, or you can use the @code{tty} command to
2063 set a different device for your program.
2064 @xref{Input/Output, ,Your Program's Input and Output}.
2067 @emph{Warning:} While input and output redirection work, you cannot use
2068 pipes to pass the output of the program you are debugging to another
2069 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2073 When you issue the @code{run} command, your program begins to execute
2074 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2075 of how to arrange for your program to stop. Once your program has
2076 stopped, you may call functions in your program, using the @code{print}
2077 or @code{call} commands. @xref{Data, ,Examining Data}.
2079 If the modification time of your symbol file has changed since the last
2080 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2081 table, and reads it again. When it does this, @value{GDBN} tries to retain
2082 your current breakpoints.
2087 @cindex run to main procedure
2088 The name of the main procedure can vary from language to language.
2089 With C or C@t{++}, the main procedure name is always @code{main}, but
2090 other languages such as Ada do not require a specific name for their
2091 main procedure. The debugger provides a convenient way to start the
2092 execution of the program and to stop at the beginning of the main
2093 procedure, depending on the language used.
2095 The @samp{start} command does the equivalent of setting a temporary
2096 breakpoint at the beginning of the main procedure and then invoking
2097 the @samp{run} command.
2099 @cindex elaboration phase
2100 Some programs contain an @dfn{elaboration} phase where some startup code is
2101 executed before the main procedure is called. This depends on the
2102 languages used to write your program. In C@t{++}, for instance,
2103 constructors for static and global objects are executed before
2104 @code{main} is called. It is therefore possible that the debugger stops
2105 before reaching the main procedure. However, the temporary breakpoint
2106 will remain to halt execution.
2108 Specify the arguments to give to your program as arguments to the
2109 @samp{start} command. These arguments will be given verbatim to the
2110 underlying @samp{run} command. Note that the same arguments will be
2111 reused if no argument is provided during subsequent calls to
2112 @samp{start} or @samp{run}.
2114 It is sometimes necessary to debug the program during elaboration. In
2115 these cases, using the @code{start} command would stop the execution of
2116 your program too late, as the program would have already completed the
2117 elaboration phase. Under these circumstances, insert breakpoints in your
2118 elaboration code before running your program.
2120 @anchor{set exec-wrapper}
2121 @kindex set exec-wrapper
2122 @item set exec-wrapper @var{wrapper}
2123 @itemx show exec-wrapper
2124 @itemx unset exec-wrapper
2125 When @samp{exec-wrapper} is set, the specified wrapper is used to
2126 launch programs for debugging. @value{GDBN} starts your program
2127 with a shell command of the form @kbd{exec @var{wrapper}
2128 @var{program}}. Quoting is added to @var{program} and its
2129 arguments, but not to @var{wrapper}, so you should add quotes if
2130 appropriate for your shell. The wrapper runs until it executes
2131 your program, and then @value{GDBN} takes control.
2133 You can use any program that eventually calls @code{execve} with
2134 its arguments as a wrapper. Several standard Unix utilities do
2135 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2136 with @code{exec "$@@"} will also work.
2138 For example, you can use @code{env} to pass an environment variable to
2139 the debugged program, without setting the variable in your shell's
2143 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2147 This command is available when debugging locally on most targets, excluding
2148 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2150 @kindex set startup-with-shell
2151 @item set startup-with-shell
2152 @itemx set startup-with-shell on
2153 @itemx set startup-with-shell off
2154 @itemx show set startup-with-shell
2155 On Unix systems, by default, if a shell is available on your target,
2156 @value{GDBN}) uses it to start your program. Arguments of the
2157 @code{run} command are passed to the shell, which does variable
2158 substitution, expands wildcard characters and performs redirection of
2159 I/O. In some circumstances, it may be useful to disable such use of a
2160 shell, for example, when debugging the shell itself or diagnosing
2161 startup failures such as:
2165 Starting program: ./a.out
2166 During startup program terminated with signal SIGSEGV, Segmentation fault.
2170 which indicates the shell or the wrapper specified with
2171 @samp{exec-wrapper} crashed, not your program. Most often, this is
2172 caused by something odd in your shell's non-interactive mode
2173 initialization file---such as @file{.cshrc} for C-shell,
2174 $@file{.zshenv} for the Z shell, or the file specified in the
2175 @samp{BASH_ENV} environment variable for BASH.
2177 @anchor{set auto-connect-native-target}
2178 @kindex set auto-connect-native-target
2179 @item set auto-connect-native-target
2180 @itemx set auto-connect-native-target on
2181 @itemx set auto-connect-native-target off
2182 @itemx show auto-connect-native-target
2184 By default, if not connected to any target yet (e.g., with
2185 @code{target remote}), the @code{run} command starts your program as a
2186 native process under @value{GDBN}, on your local machine. If you're
2187 sure you don't want to debug programs on your local machine, you can
2188 tell @value{GDBN} to not connect to the native target automatically
2189 with the @code{set auto-connect-native-target off} command.
2191 If @code{on}, which is the default, and if @value{GDBN} is not
2192 connected to a target already, the @code{run} command automaticaly
2193 connects to the native target, if one is available.
2195 If @code{off}, and if @value{GDBN} is not connected to a target
2196 already, the @code{run} command fails with an error:
2200 Don't know how to run. Try "help target".
2203 If @value{GDBN} is already connected to a target, @value{GDBN} always
2204 uses it with the @code{run} command.
2206 In any case, you can explicitly connect to the native target with the
2207 @code{target native} command. For example,
2210 (@value{GDBP}) set auto-connect-native-target off
2212 Don't know how to run. Try "help target".
2213 (@value{GDBP}) target native
2215 Starting program: ./a.out
2216 [Inferior 1 (process 10421) exited normally]
2219 In case you connected explicitly to the @code{native} target,
2220 @value{GDBN} remains connected even if all inferiors exit, ready for
2221 the next @code{run} command. Use the @code{disconnect} command to
2224 Examples of other commands that likewise respect the
2225 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2226 proc}, @code{info os}.
2228 @kindex set disable-randomization
2229 @item set disable-randomization
2230 @itemx set disable-randomization on
2231 This option (enabled by default in @value{GDBN}) will turn off the native
2232 randomization of the virtual address space of the started program. This option
2233 is useful for multiple debugging sessions to make the execution better
2234 reproducible and memory addresses reusable across debugging sessions.
2236 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2237 On @sc{gnu}/Linux you can get the same behavior using
2240 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2243 @item set disable-randomization off
2244 Leave the behavior of the started executable unchanged. Some bugs rear their
2245 ugly heads only when the program is loaded at certain addresses. If your bug
2246 disappears when you run the program under @value{GDBN}, that might be because
2247 @value{GDBN} by default disables the address randomization on platforms, such
2248 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2249 disable-randomization off} to try to reproduce such elusive bugs.
2251 On targets where it is available, virtual address space randomization
2252 protects the programs against certain kinds of security attacks. In these
2253 cases the attacker needs to know the exact location of a concrete executable
2254 code. Randomizing its location makes it impossible to inject jumps misusing
2255 a code at its expected addresses.
2257 Prelinking shared libraries provides a startup performance advantage but it
2258 makes addresses in these libraries predictable for privileged processes by
2259 having just unprivileged access at the target system. Reading the shared
2260 library binary gives enough information for assembling the malicious code
2261 misusing it. Still even a prelinked shared library can get loaded at a new
2262 random address just requiring the regular relocation process during the
2263 startup. Shared libraries not already prelinked are always loaded at
2264 a randomly chosen address.
2266 Position independent executables (PIE) contain position independent code
2267 similar to the shared libraries and therefore such executables get loaded at
2268 a randomly chosen address upon startup. PIE executables always load even
2269 already prelinked shared libraries at a random address. You can build such
2270 executable using @command{gcc -fPIE -pie}.
2272 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2273 (as long as the randomization is enabled).
2275 @item show disable-randomization
2276 Show the current setting of the explicit disable of the native randomization of
2277 the virtual address space of the started program.
2282 @section Your Program's Arguments
2284 @cindex arguments (to your program)
2285 The arguments to your program can be specified by the arguments of the
2287 They are passed to a shell, which expands wildcard characters and
2288 performs redirection of I/O, and thence to your program. Your
2289 @code{SHELL} environment variable (if it exists) specifies what shell
2290 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2291 the default shell (@file{/bin/sh} on Unix).
2293 On non-Unix systems, the program is usually invoked directly by
2294 @value{GDBN}, which emulates I/O redirection via the appropriate system
2295 calls, and the wildcard characters are expanded by the startup code of
2296 the program, not by the shell.
2298 @code{run} with no arguments uses the same arguments used by the previous
2299 @code{run}, or those set by the @code{set args} command.
2304 Specify the arguments to be used the next time your program is run. If
2305 @code{set args} has no arguments, @code{run} executes your program
2306 with no arguments. Once you have run your program with arguments,
2307 using @code{set args} before the next @code{run} is the only way to run
2308 it again without arguments.
2312 Show the arguments to give your program when it is started.
2316 @section Your Program's Environment
2318 @cindex environment (of your program)
2319 The @dfn{environment} consists of a set of environment variables and
2320 their values. Environment variables conventionally record such things as
2321 your user name, your home directory, your terminal type, and your search
2322 path for programs to run. Usually you set up environment variables with
2323 the shell and they are inherited by all the other programs you run. When
2324 debugging, it can be useful to try running your program with a modified
2325 environment without having to start @value{GDBN} over again.
2329 @item path @var{directory}
2330 Add @var{directory} to the front of the @code{PATH} environment variable
2331 (the search path for executables) that will be passed to your program.
2332 The value of @code{PATH} used by @value{GDBN} does not change.
2333 You may specify several directory names, separated by whitespace or by a
2334 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2335 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2336 is moved to the front, so it is searched sooner.
2338 You can use the string @samp{$cwd} to refer to whatever is the current
2339 working directory at the time @value{GDBN} searches the path. If you
2340 use @samp{.} instead, it refers to the directory where you executed the
2341 @code{path} command. @value{GDBN} replaces @samp{.} in the
2342 @var{directory} argument (with the current path) before adding
2343 @var{directory} to the search path.
2344 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2345 @c document that, since repeating it would be a no-op.
2349 Display the list of search paths for executables (the @code{PATH}
2350 environment variable).
2352 @kindex show environment
2353 @item show environment @r{[}@var{varname}@r{]}
2354 Print the value of environment variable @var{varname} to be given to
2355 your program when it starts. If you do not supply @var{varname},
2356 print the names and values of all environment variables to be given to
2357 your program. You can abbreviate @code{environment} as @code{env}.
2359 @kindex set environment
2360 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2361 Set environment variable @var{varname} to @var{value}. The value
2362 changes for your program (and the shell @value{GDBN} uses to launch
2363 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2364 values of environment variables are just strings, and any
2365 interpretation is supplied by your program itself. The @var{value}
2366 parameter is optional; if it is eliminated, the variable is set to a
2368 @c "any string" here does not include leading, trailing
2369 @c blanks. Gnu asks: does anyone care?
2371 For example, this command:
2378 tells the debugged program, when subsequently run, that its user is named
2379 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2380 are not actually required.)
2382 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2383 which also inherits the environment set with @code{set environment}.
2384 If necessary, you can avoid that by using the @samp{env} program as a
2385 wrapper instead of using @code{set environment}. @xref{set
2386 exec-wrapper}, for an example doing just that.
2388 @kindex unset environment
2389 @item unset environment @var{varname}
2390 Remove variable @var{varname} from the environment to be passed to your
2391 program. This is different from @samp{set env @var{varname} =};
2392 @code{unset environment} removes the variable from the environment,
2393 rather than assigning it an empty value.
2396 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2397 the shell indicated by your @code{SHELL} environment variable if it
2398 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2399 names a shell that runs an initialization file when started
2400 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2401 for the Z shell, or the file specified in the @samp{BASH_ENV}
2402 environment variable for BASH---any variables you set in that file
2403 affect your program. You may wish to move setting of environment
2404 variables to files that are only run when you sign on, such as
2405 @file{.login} or @file{.profile}.
2407 @node Working Directory
2408 @section Your Program's Working Directory
2410 @cindex working directory (of your program)
2411 Each time you start your program with @code{run}, it inherits its
2412 working directory from the current working directory of @value{GDBN}.
2413 The @value{GDBN} working directory is initially whatever it inherited
2414 from its parent process (typically the shell), but you can specify a new
2415 working directory in @value{GDBN} with the @code{cd} command.
2417 The @value{GDBN} working directory also serves as a default for the commands
2418 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2423 @cindex change working directory
2424 @item cd @r{[}@var{directory}@r{]}
2425 Set the @value{GDBN} working directory to @var{directory}. If not
2426 given, @var{directory} uses @file{'~'}.
2430 Print the @value{GDBN} working directory.
2433 It is generally impossible to find the current working directory of
2434 the process being debugged (since a program can change its directory
2435 during its run). If you work on a system where @value{GDBN} is
2436 configured with the @file{/proc} support, you can use the @code{info
2437 proc} command (@pxref{SVR4 Process Information}) to find out the
2438 current working directory of the debuggee.
2441 @section Your Program's Input and Output
2446 By default, the program you run under @value{GDBN} does input and output to
2447 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2448 to its own terminal modes to interact with you, but it records the terminal
2449 modes your program was using and switches back to them when you continue
2450 running your program.
2453 @kindex info terminal
2455 Displays information recorded by @value{GDBN} about the terminal modes your
2459 You can redirect your program's input and/or output using shell
2460 redirection with the @code{run} command. For example,
2467 starts your program, diverting its output to the file @file{outfile}.
2470 @cindex controlling terminal
2471 Another way to specify where your program should do input and output is
2472 with the @code{tty} command. This command accepts a file name as
2473 argument, and causes this file to be the default for future @code{run}
2474 commands. It also resets the controlling terminal for the child
2475 process, for future @code{run} commands. For example,
2482 directs that processes started with subsequent @code{run} commands
2483 default to do input and output on the terminal @file{/dev/ttyb} and have
2484 that as their controlling terminal.
2486 An explicit redirection in @code{run} overrides the @code{tty} command's
2487 effect on the input/output device, but not its effect on the controlling
2490 When you use the @code{tty} command or redirect input in the @code{run}
2491 command, only the input @emph{for your program} is affected. The input
2492 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2493 for @code{set inferior-tty}.
2495 @cindex inferior tty
2496 @cindex set inferior controlling terminal
2497 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2498 display the name of the terminal that will be used for future runs of your
2502 @item set inferior-tty /dev/ttyb
2503 @kindex set inferior-tty
2504 Set the tty for the program being debugged to /dev/ttyb.
2506 @item show inferior-tty
2507 @kindex show inferior-tty
2508 Show the current tty for the program being debugged.
2512 @section Debugging an Already-running Process
2517 @item attach @var{process-id}
2518 This command attaches to a running process---one that was started
2519 outside @value{GDBN}. (@code{info files} shows your active
2520 targets.) The command takes as argument a process ID. The usual way to
2521 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2522 or with the @samp{jobs -l} shell command.
2524 @code{attach} does not repeat if you press @key{RET} a second time after
2525 executing the command.
2528 To use @code{attach}, your program must be running in an environment
2529 which supports processes; for example, @code{attach} does not work for
2530 programs on bare-board targets that lack an operating system. You must
2531 also have permission to send the process a signal.
2533 When you use @code{attach}, the debugger finds the program running in
2534 the process first by looking in the current working directory, then (if
2535 the program is not found) by using the source file search path
2536 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2537 the @code{file} command to load the program. @xref{Files, ,Commands to
2540 The first thing @value{GDBN} does after arranging to debug the specified
2541 process is to stop it. You can examine and modify an attached process
2542 with all the @value{GDBN} commands that are ordinarily available when
2543 you start processes with @code{run}. You can insert breakpoints; you
2544 can step and continue; you can modify storage. If you would rather the
2545 process continue running, you may use the @code{continue} command after
2546 attaching @value{GDBN} to the process.
2551 When you have finished debugging the attached process, you can use the
2552 @code{detach} command to release it from @value{GDBN} control. Detaching
2553 the process continues its execution. After the @code{detach} command,
2554 that process and @value{GDBN} become completely independent once more, and you
2555 are ready to @code{attach} another process or start one with @code{run}.
2556 @code{detach} does not repeat if you press @key{RET} again after
2557 executing the command.
2560 If you exit @value{GDBN} while you have an attached process, you detach
2561 that process. If you use the @code{run} command, you kill that process.
2562 By default, @value{GDBN} asks for confirmation if you try to do either of these
2563 things; you can control whether or not you need to confirm by using the
2564 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2568 @section Killing the Child Process
2573 Kill the child process in which your program is running under @value{GDBN}.
2576 This command is useful if you wish to debug a core dump instead of a
2577 running process. @value{GDBN} ignores any core dump file while your program
2580 On some operating systems, a program cannot be executed outside @value{GDBN}
2581 while you have breakpoints set on it inside @value{GDBN}. You can use the
2582 @code{kill} command in this situation to permit running your program
2583 outside the debugger.
2585 The @code{kill} command is also useful if you wish to recompile and
2586 relink your program, since on many systems it is impossible to modify an
2587 executable file while it is running in a process. In this case, when you
2588 next type @code{run}, @value{GDBN} notices that the file has changed, and
2589 reads the symbol table again (while trying to preserve your current
2590 breakpoint settings).
2592 @node Inferiors and Programs
2593 @section Debugging Multiple Inferiors and Programs
2595 @value{GDBN} lets you run and debug multiple programs in a single
2596 session. In addition, @value{GDBN} on some systems may let you run
2597 several programs simultaneously (otherwise you have to exit from one
2598 before starting another). In the most general case, you can have
2599 multiple threads of execution in each of multiple processes, launched
2600 from multiple executables.
2603 @value{GDBN} represents the state of each program execution with an
2604 object called an @dfn{inferior}. An inferior typically corresponds to
2605 a process, but is more general and applies also to targets that do not
2606 have processes. Inferiors may be created before a process runs, and
2607 may be retained after a process exits. Inferiors have unique
2608 identifiers that are different from process ids. Usually each
2609 inferior will also have its own distinct address space, although some
2610 embedded targets may have several inferiors running in different parts
2611 of a single address space. Each inferior may in turn have multiple
2612 threads running in it.
2614 To find out what inferiors exist at any moment, use @w{@code{info
2618 @kindex info inferiors
2619 @item info inferiors
2620 Print a list of all inferiors currently being managed by @value{GDBN}.
2622 @value{GDBN} displays for each inferior (in this order):
2626 the inferior number assigned by @value{GDBN}
2629 the target system's inferior identifier
2632 the name of the executable the inferior is running.
2637 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2638 indicates the current inferior.
2642 @c end table here to get a little more width for example
2645 (@value{GDBP}) info inferiors
2646 Num Description Executable
2647 2 process 2307 hello
2648 * 1 process 3401 goodbye
2651 To switch focus between inferiors, use the @code{inferior} command:
2654 @kindex inferior @var{infno}
2655 @item inferior @var{infno}
2656 Make inferior number @var{infno} the current inferior. The argument
2657 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2658 in the first field of the @samp{info inferiors} display.
2661 @vindex $_inferior@r{, convenience variable}
2662 The debugger convenience variable @samp{$_inferior} contains the
2663 number of the current inferior. You may find this useful in writing
2664 breakpoint conditional expressions, command scripts, and so forth.
2665 @xref{Convenience Vars,, Convenience Variables}, for general
2666 information on convenience variables.
2668 You can get multiple executables into a debugging session via the
2669 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2670 systems @value{GDBN} can add inferiors to the debug session
2671 automatically by following calls to @code{fork} and @code{exec}. To
2672 remove inferiors from the debugging session use the
2673 @w{@code{remove-inferiors}} command.
2676 @kindex add-inferior
2677 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2678 Adds @var{n} inferiors to be run using @var{executable} as the
2679 executable; @var{n} defaults to 1. If no executable is specified,
2680 the inferiors begins empty, with no program. You can still assign or
2681 change the program assigned to the inferior at any time by using the
2682 @code{file} command with the executable name as its argument.
2684 @kindex clone-inferior
2685 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2686 Adds @var{n} inferiors ready to execute the same program as inferior
2687 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
2688 number of the current inferior. This is a convenient command when you
2689 want to run another instance of the inferior you are debugging.
2692 (@value{GDBP}) info inferiors
2693 Num Description Executable
2694 * 1 process 29964 helloworld
2695 (@value{GDBP}) clone-inferior
2698 (@value{GDBP}) info inferiors
2699 Num Description Executable
2701 * 1 process 29964 helloworld
2704 You can now simply switch focus to inferior 2 and run it.
2706 @kindex remove-inferiors
2707 @item remove-inferiors @var{infno}@dots{}
2708 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2709 possible to remove an inferior that is running with this command. For
2710 those, use the @code{kill} or @code{detach} command first.
2714 To quit debugging one of the running inferiors that is not the current
2715 inferior, you can either detach from it by using the @w{@code{detach
2716 inferior}} command (allowing it to run independently), or kill it
2717 using the @w{@code{kill inferiors}} command:
2720 @kindex detach inferiors @var{infno}@dots{}
2721 @item detach inferior @var{infno}@dots{}
2722 Detach from the inferior or inferiors identified by @value{GDBN}
2723 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2724 still stays on the list of inferiors shown by @code{info inferiors},
2725 but its Description will show @samp{<null>}.
2727 @kindex kill inferiors @var{infno}@dots{}
2728 @item kill inferiors @var{infno}@dots{}
2729 Kill the inferior or inferiors identified by @value{GDBN} inferior
2730 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2731 stays on the list of inferiors shown by @code{info inferiors}, but its
2732 Description will show @samp{<null>}.
2735 After the successful completion of a command such as @code{detach},
2736 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2737 a normal process exit, the inferior is still valid and listed with
2738 @code{info inferiors}, ready to be restarted.
2741 To be notified when inferiors are started or exit under @value{GDBN}'s
2742 control use @w{@code{set print inferior-events}}:
2745 @kindex set print inferior-events
2746 @cindex print messages on inferior start and exit
2747 @item set print inferior-events
2748 @itemx set print inferior-events on
2749 @itemx set print inferior-events off
2750 The @code{set print inferior-events} command allows you to enable or
2751 disable printing of messages when @value{GDBN} notices that new
2752 inferiors have started or that inferiors have exited or have been
2753 detached. By default, these messages will not be printed.
2755 @kindex show print inferior-events
2756 @item show print inferior-events
2757 Show whether messages will be printed when @value{GDBN} detects that
2758 inferiors have started, exited or have been detached.
2761 Many commands will work the same with multiple programs as with a
2762 single program: e.g., @code{print myglobal} will simply display the
2763 value of @code{myglobal} in the current inferior.
2766 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2767 get more info about the relationship of inferiors, programs, address
2768 spaces in a debug session. You can do that with the @w{@code{maint
2769 info program-spaces}} command.
2772 @kindex maint info program-spaces
2773 @item maint info program-spaces
2774 Print a list of all program spaces currently being managed by
2777 @value{GDBN} displays for each program space (in this order):
2781 the program space number assigned by @value{GDBN}
2784 the name of the executable loaded into the program space, with e.g.,
2785 the @code{file} command.
2790 An asterisk @samp{*} preceding the @value{GDBN} program space number
2791 indicates the current program space.
2793 In addition, below each program space line, @value{GDBN} prints extra
2794 information that isn't suitable to display in tabular form. For
2795 example, the list of inferiors bound to the program space.
2798 (@value{GDBP}) maint info program-spaces
2802 Bound inferiors: ID 1 (process 21561)
2805 Here we can see that no inferior is running the program @code{hello},
2806 while @code{process 21561} is running the program @code{goodbye}. On
2807 some targets, it is possible that multiple inferiors are bound to the
2808 same program space. The most common example is that of debugging both
2809 the parent and child processes of a @code{vfork} call. For example,
2812 (@value{GDBP}) maint info program-spaces
2815 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2818 Here, both inferior 2 and inferior 1 are running in the same program
2819 space as a result of inferior 1 having executed a @code{vfork} call.
2823 @section Debugging Programs with Multiple Threads
2825 @cindex threads of execution
2826 @cindex multiple threads
2827 @cindex switching threads
2828 In some operating systems, such as GNU/Linux and Solaris, a single program
2829 may have more than one @dfn{thread} of execution. The precise semantics
2830 of threads differ from one operating system to another, but in general
2831 the threads of a single program are akin to multiple processes---except
2832 that they share one address space (that is, they can all examine and
2833 modify the same variables). On the other hand, each thread has its own
2834 registers and execution stack, and perhaps private memory.
2836 @value{GDBN} provides these facilities for debugging multi-thread
2840 @item automatic notification of new threads
2841 @item @samp{thread @var{thread-id}}, a command to switch among threads
2842 @item @samp{info threads}, a command to inquire about existing threads
2843 @item @samp{thread apply [@var{thread-id-list}] [@var{all}] @var{args}},
2844 a command to apply a command to a list of threads
2845 @item thread-specific breakpoints
2846 @item @samp{set print thread-events}, which controls printing of
2847 messages on thread start and exit.
2848 @item @samp{set libthread-db-search-path @var{path}}, which lets
2849 the user specify which @code{libthread_db} to use if the default choice
2850 isn't compatible with the program.
2853 @cindex focus of debugging
2854 @cindex current thread
2855 The @value{GDBN} thread debugging facility allows you to observe all
2856 threads while your program runs---but whenever @value{GDBN} takes
2857 control, one thread in particular is always the focus of debugging.
2858 This thread is called the @dfn{current thread}. Debugging commands show
2859 program information from the perspective of the current thread.
2861 @cindex @code{New} @var{systag} message
2862 @cindex thread identifier (system)
2863 @c FIXME-implementors!! It would be more helpful if the [New...] message
2864 @c included GDB's numeric thread handle, so you could just go to that
2865 @c thread without first checking `info threads'.
2866 Whenever @value{GDBN} detects a new thread in your program, it displays
2867 the target system's identification for the thread with a message in the
2868 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
2869 whose form varies depending on the particular system. For example, on
2870 @sc{gnu}/Linux, you might see
2873 [New Thread 0x41e02940 (LWP 25582)]
2877 when @value{GDBN} notices a new thread. In contrast, on other systems,
2878 the @var{systag} is simply something like @samp{process 368}, with no
2881 @c FIXME!! (1) Does the [New...] message appear even for the very first
2882 @c thread of a program, or does it only appear for the
2883 @c second---i.e.@: when it becomes obvious we have a multithread
2885 @c (2) *Is* there necessarily a first thread always? Or do some
2886 @c multithread systems permit starting a program with multiple
2887 @c threads ab initio?
2889 @anchor{thread numbers}
2890 @cindex thread number, per inferior
2891 @cindex thread identifier (GDB)
2892 For debugging purposes, @value{GDBN} associates its own thread number
2893 ---always a single integer---with each thread of an inferior. This
2894 number is unique between all threads of an inferior, but not unique
2895 between threads of different inferiors.
2897 @cindex qualified thread ID
2898 You can refer to a given thread in an inferior using the qualified
2899 @var{inferior-num}.@var{thread-num} syntax, also known as
2900 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
2901 number and @var{thread-num} being the thread number of the given
2902 inferior. For example, thread @code{2.3} refers to thread number 3 of
2903 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
2904 then @value{GDBN} infers you're referring to a thread of the current
2907 Until you create a second inferior, @value{GDBN} does not show the
2908 @var{inferior-num} part of thread IDs, even though you can always use
2909 the full @var{inferior-num}.@var{thread-num} form to refer to threads
2910 of inferior 1, the initial inferior.
2912 @anchor{thread ID lists}
2913 @cindex thread ID lists
2914 Some commands accept a space-separated @dfn{thread ID list} as
2915 argument. A list element can be:
2919 A thread ID as shown in the first field of the @samp{info threads}
2920 display, with or without an inferior qualifier. E.g., @samp{2.1} or
2924 A range of thread numbers, again with or without an inferior
2925 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
2926 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
2929 All threads of an inferior, specified with a star wildcard, with or
2930 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
2931 @samp{1.*}) or @code{*}. The former refers to all threads of the
2932 given inferior, and the latter form without an inferior qualifier
2933 refers to all threads of the current inferior.
2937 For example, if the current inferior is 1, and inferior 7 has one
2938 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
2939 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
2940 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
2941 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
2945 @anchor{global thread numbers}
2946 @cindex global thread number
2947 @cindex global thread identifier (GDB)
2948 In addition to a @emph{per-inferior} number, each thread is also
2949 assigned a unique @emph{global} number, also known as @dfn{global
2950 thread ID}, a single integer. Unlike the thread number component of
2951 the thread ID, no two threads have the same global ID, even when
2952 you're debugging multiple inferiors.
2954 From @value{GDBN}'s perspective, a process always has at least one
2955 thread. In other words, @value{GDBN} assigns a thread number to the
2956 program's ``main thread'' even if the program is not multi-threaded.
2958 @vindex $_thread@r{, convenience variable}
2959 @vindex $_gthread@r{, convenience variable}
2960 The debugger convenience variables @samp{$_thread} and
2961 @samp{$_gthread} contain, respectively, the per-inferior thread number
2962 and the global thread number of the current thread. You may find this
2963 useful in writing breakpoint conditional expressions, command scripts,
2964 and so forth. @xref{Convenience Vars,, Convenience Variables}, for
2965 general information on convenience variables.
2967 If @value{GDBN} detects the program is multi-threaded, it augments the
2968 usual message about stopping at a breakpoint with the ID and name of
2969 the thread that hit the breakpoint.
2972 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
2975 Likewise when the program receives a signal:
2978 Thread 1 "main" received signal SIGINT, Interrupt.
2982 @kindex info threads
2983 @item info threads @r{[}@var{thread-id-list}@r{]}
2985 Display information about one or more threads. With no arguments
2986 displays information about all threads. You can specify the list of
2987 threads that you want to display using the thread ID list syntax
2988 (@pxref{thread ID lists}).
2990 @value{GDBN} displays for each thread (in this order):
2994 the per-inferior thread number assigned by @value{GDBN}
2997 the global thread number assigned by @value{GDBN}, if the @samp{-gid}
2998 option was specified
3001 the target system's thread identifier (@var{systag})
3004 the thread's name, if one is known. A thread can either be named by
3005 the user (see @code{thread name}, below), or, in some cases, by the
3009 the current stack frame summary for that thread
3013 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3014 indicates the current thread.
3018 @c end table here to get a little more width for example
3021 (@value{GDBP}) info threads
3023 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3024 2 process 35 thread 23 0x34e5 in sigpause ()
3025 3 process 35 thread 27 0x34e5 in sigpause ()
3029 If you're debugging multiple inferiors, @value{GDBN} displays thread
3030 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3031 Otherwise, only @var{thread-num} is shown.
3033 If you specify the @samp{-gid} option, @value{GDBN} displays a column
3034 indicating each thread's global thread ID:
3037 (@value{GDBP}) info threads
3038 Id GId Target Id Frame
3039 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3040 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3041 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3042 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3045 On Solaris, you can display more information about user threads with a
3046 Solaris-specific command:
3049 @item maint info sol-threads
3050 @kindex maint info sol-threads
3051 @cindex thread info (Solaris)
3052 Display info on Solaris user threads.
3056 @kindex thread @var{thread-id}
3057 @item thread @var{thread-id}
3058 Make thread ID @var{thread-id} the current thread. The command
3059 argument @var{thread-id} is the @value{GDBN} thread ID, as shown in
3060 the first field of the @samp{info threads} display, with or without an
3061 inferior qualifier (e.g., @samp{2.1} or @samp{1}).
3063 @value{GDBN} responds by displaying the system identifier of the
3064 thread you selected, and its current stack frame summary:
3067 (@value{GDBP}) thread 2
3068 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3069 #0 some_function (ignore=0x0) at example.c:8
3070 8 printf ("hello\n");
3074 As with the @samp{[New @dots{}]} message, the form of the text after
3075 @samp{Switching to} depends on your system's conventions for identifying
3078 @kindex thread apply
3079 @cindex apply command to several threads
3080 @item thread apply [@var{thread-id-list} | all [-ascending]] @var{command}
3081 The @code{thread apply} command allows you to apply the named
3082 @var{command} to one or more threads. Specify the threads that you
3083 want affected using the thread ID list syntax (@pxref{thread ID
3084 lists}), or specify @code{all} to apply to all threads. To apply a
3085 command to all threads in descending order, type @kbd{thread apply all
3086 @var{command}}. To apply a command to all threads in ascending order,
3087 type @kbd{thread apply all -ascending @var{command}}.
3091 @cindex name a thread
3092 @item thread name [@var{name}]
3093 This command assigns a name to the current thread. If no argument is
3094 given, any existing user-specified name is removed. The thread name
3095 appears in the @samp{info threads} display.
3097 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3098 determine the name of the thread as given by the OS. On these
3099 systems, a name specified with @samp{thread name} will override the
3100 system-give name, and removing the user-specified name will cause
3101 @value{GDBN} to once again display the system-specified name.
3104 @cindex search for a thread
3105 @item thread find [@var{regexp}]
3106 Search for and display thread ids whose name or @var{systag}
3107 matches the supplied regular expression.
3109 As well as being the complement to the @samp{thread name} command,
3110 this command also allows you to identify a thread by its target
3111 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3115 (@value{GDBN}) thread find 26688
3116 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3117 (@value{GDBN}) info thread 4
3119 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3122 @kindex set print thread-events
3123 @cindex print messages on thread start and exit
3124 @item set print thread-events
3125 @itemx set print thread-events on
3126 @itemx set print thread-events off
3127 The @code{set print thread-events} command allows you to enable or
3128 disable printing of messages when @value{GDBN} notices that new threads have
3129 started or that threads have exited. By default, these messages will
3130 be printed if detection of these events is supported by the target.
3131 Note that these messages cannot be disabled on all targets.
3133 @kindex show print thread-events
3134 @item show print thread-events
3135 Show whether messages will be printed when @value{GDBN} detects that threads
3136 have started and exited.
3139 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3140 more information about how @value{GDBN} behaves when you stop and start
3141 programs with multiple threads.
3143 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3144 watchpoints in programs with multiple threads.
3146 @anchor{set libthread-db-search-path}
3148 @kindex set libthread-db-search-path
3149 @cindex search path for @code{libthread_db}
3150 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3151 If this variable is set, @var{path} is a colon-separated list of
3152 directories @value{GDBN} will use to search for @code{libthread_db}.
3153 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3154 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3155 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3158 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3159 @code{libthread_db} library to obtain information about threads in the
3160 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3161 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3162 specific thread debugging library loading is enabled
3163 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3165 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3166 refers to the default system directories that are
3167 normally searched for loading shared libraries. The @samp{$sdir} entry
3168 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3169 (@pxref{libthread_db.so.1 file}).
3171 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3172 refers to the directory from which @code{libpthread}
3173 was loaded in the inferior process.
3175 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3176 @value{GDBN} attempts to initialize it with the current inferior process.
3177 If this initialization fails (which could happen because of a version
3178 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3179 will unload @code{libthread_db}, and continue with the next directory.
3180 If none of @code{libthread_db} libraries initialize successfully,
3181 @value{GDBN} will issue a warning and thread debugging will be disabled.
3183 Setting @code{libthread-db-search-path} is currently implemented
3184 only on some platforms.
3186 @kindex show libthread-db-search-path
3187 @item show libthread-db-search-path
3188 Display current libthread_db search path.
3190 @kindex set debug libthread-db
3191 @kindex show debug libthread-db
3192 @cindex debugging @code{libthread_db}
3193 @item set debug libthread-db
3194 @itemx show debug libthread-db
3195 Turns on or off display of @code{libthread_db}-related events.
3196 Use @code{1} to enable, @code{0} to disable.
3200 @section Debugging Forks
3202 @cindex fork, debugging programs which call
3203 @cindex multiple processes
3204 @cindex processes, multiple
3205 On most systems, @value{GDBN} has no special support for debugging
3206 programs which create additional processes using the @code{fork}
3207 function. When a program forks, @value{GDBN} will continue to debug the
3208 parent process and the child process will run unimpeded. If you have
3209 set a breakpoint in any code which the child then executes, the child
3210 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3211 will cause it to terminate.
3213 However, if you want to debug the child process there is a workaround
3214 which isn't too painful. Put a call to @code{sleep} in the code which
3215 the child process executes after the fork. It may be useful to sleep
3216 only if a certain environment variable is set, or a certain file exists,
3217 so that the delay need not occur when you don't want to run @value{GDBN}
3218 on the child. While the child is sleeping, use the @code{ps} program to
3219 get its process ID. Then tell @value{GDBN} (a new invocation of
3220 @value{GDBN} if you are also debugging the parent process) to attach to
3221 the child process (@pxref{Attach}). From that point on you can debug
3222 the child process just like any other process which you attached to.
3224 On some systems, @value{GDBN} provides support for debugging programs
3225 that create additional processes using the @code{fork} or @code{vfork}
3226 functions. On @sc{gnu}/Linux platforms, this feature is supported
3227 with kernel version 2.5.46 and later.
3229 The fork debugging commands are supported in native mode and when
3230 connected to @code{gdbserver} in either @code{target remote} mode or
3231 @code{target extended-remote} mode.
3233 By default, when a program forks, @value{GDBN} will continue to debug
3234 the parent process and the child process will run unimpeded.
3236 If you want to follow the child process instead of the parent process,
3237 use the command @w{@code{set follow-fork-mode}}.
3240 @kindex set follow-fork-mode
3241 @item set follow-fork-mode @var{mode}
3242 Set the debugger response to a program call of @code{fork} or
3243 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3244 process. The @var{mode} argument can be:
3248 The original process is debugged after a fork. The child process runs
3249 unimpeded. This is the default.
3252 The new process is debugged after a fork. The parent process runs
3257 @kindex show follow-fork-mode
3258 @item show follow-fork-mode
3259 Display the current debugger response to a @code{fork} or @code{vfork} call.
3262 @cindex debugging multiple processes
3263 On Linux, if you want to debug both the parent and child processes, use the
3264 command @w{@code{set detach-on-fork}}.
3267 @kindex set detach-on-fork
3268 @item set detach-on-fork @var{mode}
3269 Tells gdb whether to detach one of the processes after a fork, or
3270 retain debugger control over them both.
3274 The child process (or parent process, depending on the value of
3275 @code{follow-fork-mode}) will be detached and allowed to run
3276 independently. This is the default.
3279 Both processes will be held under the control of @value{GDBN}.
3280 One process (child or parent, depending on the value of
3281 @code{follow-fork-mode}) is debugged as usual, while the other
3286 @kindex show detach-on-fork
3287 @item show detach-on-fork
3288 Show whether detach-on-fork mode is on/off.
3291 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3292 will retain control of all forked processes (including nested forks).
3293 You can list the forked processes under the control of @value{GDBN} by
3294 using the @w{@code{info inferiors}} command, and switch from one fork
3295 to another by using the @code{inferior} command (@pxref{Inferiors and
3296 Programs, ,Debugging Multiple Inferiors and Programs}).
3298 To quit debugging one of the forked processes, you can either detach
3299 from it by using the @w{@code{detach inferiors}} command (allowing it
3300 to run independently), or kill it using the @w{@code{kill inferiors}}
3301 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3304 If you ask to debug a child process and a @code{vfork} is followed by an
3305 @code{exec}, @value{GDBN} executes the new target up to the first
3306 breakpoint in the new target. If you have a breakpoint set on
3307 @code{main} in your original program, the breakpoint will also be set on
3308 the child process's @code{main}.
3310 On some systems, when a child process is spawned by @code{vfork}, you
3311 cannot debug the child or parent until an @code{exec} call completes.
3313 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3314 call executes, the new target restarts. To restart the parent
3315 process, use the @code{file} command with the parent executable name
3316 as its argument. By default, after an @code{exec} call executes,
3317 @value{GDBN} discards the symbols of the previous executable image.
3318 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3322 @kindex set follow-exec-mode
3323 @item set follow-exec-mode @var{mode}
3325 Set debugger response to a program call of @code{exec}. An
3326 @code{exec} call replaces the program image of a process.
3328 @code{follow-exec-mode} can be:
3332 @value{GDBN} creates a new inferior and rebinds the process to this
3333 new inferior. The program the process was running before the
3334 @code{exec} call can be restarted afterwards by restarting the
3340 (@value{GDBP}) info inferiors
3342 Id Description Executable
3345 process 12020 is executing new program: prog2
3346 Program exited normally.
3347 (@value{GDBP}) info inferiors
3348 Id Description Executable
3354 @value{GDBN} keeps the process bound to the same inferior. The new
3355 executable image replaces the previous executable loaded in the
3356 inferior. Restarting the inferior after the @code{exec} call, with
3357 e.g., the @code{run} command, restarts the executable the process was
3358 running after the @code{exec} call. This is the default mode.
3363 (@value{GDBP}) info inferiors
3364 Id Description Executable
3367 process 12020 is executing new program: prog2
3368 Program exited normally.
3369 (@value{GDBP}) info inferiors
3370 Id Description Executable
3377 @code{follow-exec-mode} is supported in native mode and
3378 @code{target extended-remote} mode.
3380 You can use the @code{catch} command to make @value{GDBN} stop whenever
3381 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3382 Catchpoints, ,Setting Catchpoints}.
3384 @node Checkpoint/Restart
3385 @section Setting a @emph{Bookmark} to Return to Later
3390 @cindex snapshot of a process
3391 @cindex rewind program state
3393 On certain operating systems@footnote{Currently, only
3394 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3395 program's state, called a @dfn{checkpoint}, and come back to it
3398 Returning to a checkpoint effectively undoes everything that has
3399 happened in the program since the @code{checkpoint} was saved. This
3400 includes changes in memory, registers, and even (within some limits)
3401 system state. Effectively, it is like going back in time to the
3402 moment when the checkpoint was saved.
3404 Thus, if you're stepping thru a program and you think you're
3405 getting close to the point where things go wrong, you can save
3406 a checkpoint. Then, if you accidentally go too far and miss
3407 the critical statement, instead of having to restart your program
3408 from the beginning, you can just go back to the checkpoint and
3409 start again from there.
3411 This can be especially useful if it takes a lot of time or
3412 steps to reach the point where you think the bug occurs.
3414 To use the @code{checkpoint}/@code{restart} method of debugging:
3419 Save a snapshot of the debugged program's current execution state.
3420 The @code{checkpoint} command takes no arguments, but each checkpoint
3421 is assigned a small integer id, similar to a breakpoint id.
3423 @kindex info checkpoints
3424 @item info checkpoints
3425 List the checkpoints that have been saved in the current debugging
3426 session. For each checkpoint, the following information will be
3433 @item Source line, or label
3436 @kindex restart @var{checkpoint-id}
3437 @item restart @var{checkpoint-id}
3438 Restore the program state that was saved as checkpoint number
3439 @var{checkpoint-id}. All program variables, registers, stack frames
3440 etc.@: will be returned to the values that they had when the checkpoint
3441 was saved. In essence, gdb will ``wind back the clock'' to the point
3442 in time when the checkpoint was saved.
3444 Note that breakpoints, @value{GDBN} variables, command history etc.
3445 are not affected by restoring a checkpoint. In general, a checkpoint
3446 only restores things that reside in the program being debugged, not in
3449 @kindex delete checkpoint @var{checkpoint-id}
3450 @item delete checkpoint @var{checkpoint-id}
3451 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3455 Returning to a previously saved checkpoint will restore the user state
3456 of the program being debugged, plus a significant subset of the system
3457 (OS) state, including file pointers. It won't ``un-write'' data from
3458 a file, but it will rewind the file pointer to the previous location,
3459 so that the previously written data can be overwritten. For files
3460 opened in read mode, the pointer will also be restored so that the
3461 previously read data can be read again.
3463 Of course, characters that have been sent to a printer (or other
3464 external device) cannot be ``snatched back'', and characters received
3465 from eg.@: a serial device can be removed from internal program buffers,
3466 but they cannot be ``pushed back'' into the serial pipeline, ready to
3467 be received again. Similarly, the actual contents of files that have
3468 been changed cannot be restored (at this time).
3470 However, within those constraints, you actually can ``rewind'' your
3471 program to a previously saved point in time, and begin debugging it
3472 again --- and you can change the course of events so as to debug a
3473 different execution path this time.
3475 @cindex checkpoints and process id
3476 Finally, there is one bit of internal program state that will be
3477 different when you return to a checkpoint --- the program's process
3478 id. Each checkpoint will have a unique process id (or @var{pid}),
3479 and each will be different from the program's original @var{pid}.
3480 If your program has saved a local copy of its process id, this could
3481 potentially pose a problem.
3483 @subsection A Non-obvious Benefit of Using Checkpoints
3485 On some systems such as @sc{gnu}/Linux, address space randomization
3486 is performed on new processes for security reasons. This makes it
3487 difficult or impossible to set a breakpoint, or watchpoint, on an
3488 absolute address if you have to restart the program, since the
3489 absolute location of a symbol will change from one execution to the
3492 A checkpoint, however, is an @emph{identical} copy of a process.
3493 Therefore if you create a checkpoint at (eg.@:) the start of main,
3494 and simply return to that checkpoint instead of restarting the
3495 process, you can avoid the effects of address randomization and
3496 your symbols will all stay in the same place.
3499 @chapter Stopping and Continuing
3501 The principal purposes of using a debugger are so that you can stop your
3502 program before it terminates; or so that, if your program runs into
3503 trouble, you can investigate and find out why.
3505 Inside @value{GDBN}, your program may stop for any of several reasons,
3506 such as a signal, a breakpoint, or reaching a new line after a
3507 @value{GDBN} command such as @code{step}. You may then examine and
3508 change variables, set new breakpoints or remove old ones, and then
3509 continue execution. Usually, the messages shown by @value{GDBN} provide
3510 ample explanation of the status of your program---but you can also
3511 explicitly request this information at any time.
3514 @kindex info program
3516 Display information about the status of your program: whether it is
3517 running or not, what process it is, and why it stopped.
3521 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3522 * Continuing and Stepping:: Resuming execution
3523 * Skipping Over Functions and Files::
3524 Skipping over functions and files
3526 * Thread Stops:: Stopping and starting multi-thread programs
3530 @section Breakpoints, Watchpoints, and Catchpoints
3533 A @dfn{breakpoint} makes your program stop whenever a certain point in
3534 the program is reached. For each breakpoint, you can add conditions to
3535 control in finer detail whether your program stops. You can set
3536 breakpoints with the @code{break} command and its variants (@pxref{Set
3537 Breaks, ,Setting Breakpoints}), to specify the place where your program
3538 should stop by line number, function name or exact address in the
3541 On some systems, you can set breakpoints in shared libraries before
3542 the executable is run.
3545 @cindex data breakpoints
3546 @cindex memory tracing
3547 @cindex breakpoint on memory address
3548 @cindex breakpoint on variable modification
3549 A @dfn{watchpoint} is a special breakpoint that stops your program
3550 when the value of an expression changes. The expression may be a value
3551 of a variable, or it could involve values of one or more variables
3552 combined by operators, such as @samp{a + b}. This is sometimes called
3553 @dfn{data breakpoints}. You must use a different command to set
3554 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3555 from that, you can manage a watchpoint like any other breakpoint: you
3556 enable, disable, and delete both breakpoints and watchpoints using the
3559 You can arrange to have values from your program displayed automatically
3560 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3564 @cindex breakpoint on events
3565 A @dfn{catchpoint} is another special breakpoint that stops your program
3566 when a certain kind of event occurs, such as the throwing of a C@t{++}
3567 exception or the loading of a library. As with watchpoints, you use a
3568 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3569 Catchpoints}), but aside from that, you can manage a catchpoint like any
3570 other breakpoint. (To stop when your program receives a signal, use the
3571 @code{handle} command; see @ref{Signals, ,Signals}.)
3573 @cindex breakpoint numbers
3574 @cindex numbers for breakpoints
3575 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3576 catchpoint when you create it; these numbers are successive integers
3577 starting with one. In many of the commands for controlling various
3578 features of breakpoints you use the breakpoint number to say which
3579 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3580 @dfn{disabled}; if disabled, it has no effect on your program until you
3583 @cindex breakpoint ranges
3584 @cindex ranges of breakpoints
3585 Some @value{GDBN} commands accept a range of breakpoints on which to
3586 operate. A breakpoint range is either a single breakpoint number, like
3587 @samp{5}, or two such numbers, in increasing order, separated by a
3588 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3589 all breakpoints in that range are operated on.
3592 * Set Breaks:: Setting breakpoints
3593 * Set Watchpoints:: Setting watchpoints
3594 * Set Catchpoints:: Setting catchpoints
3595 * Delete Breaks:: Deleting breakpoints
3596 * Disabling:: Disabling breakpoints
3597 * Conditions:: Break conditions
3598 * Break Commands:: Breakpoint command lists
3599 * Dynamic Printf:: Dynamic printf
3600 * Save Breakpoints:: How to save breakpoints in a file
3601 * Static Probe Points:: Listing static probe points
3602 * Error in Breakpoints:: ``Cannot insert breakpoints''
3603 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3607 @subsection Setting Breakpoints
3609 @c FIXME LMB what does GDB do if no code on line of breakpt?
3610 @c consider in particular declaration with/without initialization.
3612 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3615 @kindex b @r{(@code{break})}
3616 @vindex $bpnum@r{, convenience variable}
3617 @cindex latest breakpoint
3618 Breakpoints are set with the @code{break} command (abbreviated
3619 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3620 number of the breakpoint you've set most recently; see @ref{Convenience
3621 Vars,, Convenience Variables}, for a discussion of what you can do with
3622 convenience variables.
3625 @item break @var{location}
3626 Set a breakpoint at the given @var{location}, which can specify a
3627 function name, a line number, or an address of an instruction.
3628 (@xref{Specify Location}, for a list of all the possible ways to
3629 specify a @var{location}.) The breakpoint will stop your program just
3630 before it executes any of the code in the specified @var{location}.
3632 When using source languages that permit overloading of symbols, such as
3633 C@t{++}, a function name may refer to more than one possible place to break.
3634 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3637 It is also possible to insert a breakpoint that will stop the program
3638 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3639 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3642 When called without any arguments, @code{break} sets a breakpoint at
3643 the next instruction to be executed in the selected stack frame
3644 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3645 innermost, this makes your program stop as soon as control
3646 returns to that frame. This is similar to the effect of a
3647 @code{finish} command in the frame inside the selected frame---except
3648 that @code{finish} does not leave an active breakpoint. If you use
3649 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3650 the next time it reaches the current location; this may be useful
3653 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3654 least one instruction has been executed. If it did not do this, you
3655 would be unable to proceed past a breakpoint without first disabling the
3656 breakpoint. This rule applies whether or not the breakpoint already
3657 existed when your program stopped.
3659 @item break @dots{} if @var{cond}
3660 Set a breakpoint with condition @var{cond}; evaluate the expression
3661 @var{cond} each time the breakpoint is reached, and stop only if the
3662 value is nonzero---that is, if @var{cond} evaluates as true.
3663 @samp{@dots{}} stands for one of the possible arguments described
3664 above (or no argument) specifying where to break. @xref{Conditions,
3665 ,Break Conditions}, for more information on breakpoint conditions.
3668 @item tbreak @var{args}
3669 Set a breakpoint enabled only for one stop. The @var{args} are the
3670 same as for the @code{break} command, and the breakpoint is set in the same
3671 way, but the breakpoint is automatically deleted after the first time your
3672 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3675 @cindex hardware breakpoints
3676 @item hbreak @var{args}
3677 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
3678 @code{break} command and the breakpoint is set in the same way, but the
3679 breakpoint requires hardware support and some target hardware may not
3680 have this support. The main purpose of this is EPROM/ROM code
3681 debugging, so you can set a breakpoint at an instruction without
3682 changing the instruction. This can be used with the new trap-generation
3683 provided by SPARClite DSU and most x86-based targets. These targets
3684 will generate traps when a program accesses some data or instruction
3685 address that is assigned to the debug registers. However the hardware
3686 breakpoint registers can take a limited number of breakpoints. For
3687 example, on the DSU, only two data breakpoints can be set at a time, and
3688 @value{GDBN} will reject this command if more than two are used. Delete
3689 or disable unused hardware breakpoints before setting new ones
3690 (@pxref{Disabling, ,Disabling Breakpoints}).
3691 @xref{Conditions, ,Break Conditions}.
3692 For remote targets, you can restrict the number of hardware
3693 breakpoints @value{GDBN} will use, see @ref{set remote
3694 hardware-breakpoint-limit}.
3697 @item thbreak @var{args}
3698 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
3699 are the same as for the @code{hbreak} command and the breakpoint is set in
3700 the same way. However, like the @code{tbreak} command,
3701 the breakpoint is automatically deleted after the
3702 first time your program stops there. Also, like the @code{hbreak}
3703 command, the breakpoint requires hardware support and some target hardware
3704 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3705 See also @ref{Conditions, ,Break Conditions}.
3708 @cindex regular expression
3709 @cindex breakpoints at functions matching a regexp
3710 @cindex set breakpoints in many functions
3711 @item rbreak @var{regex}
3712 Set breakpoints on all functions matching the regular expression
3713 @var{regex}. This command sets an unconditional breakpoint on all
3714 matches, printing a list of all breakpoints it set. Once these
3715 breakpoints are set, they are treated just like the breakpoints set with
3716 the @code{break} command. You can delete them, disable them, or make
3717 them conditional the same way as any other breakpoint.
3719 The syntax of the regular expression is the standard one used with tools
3720 like @file{grep}. Note that this is different from the syntax used by
3721 shells, so for instance @code{foo*} matches all functions that include
3722 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3723 @code{.*} leading and trailing the regular expression you supply, so to
3724 match only functions that begin with @code{foo}, use @code{^foo}.
3726 @cindex non-member C@t{++} functions, set breakpoint in
3727 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3728 breakpoints on overloaded functions that are not members of any special
3731 @cindex set breakpoints on all functions
3732 The @code{rbreak} command can be used to set breakpoints in
3733 @strong{all} the functions in a program, like this:
3736 (@value{GDBP}) rbreak .
3739 @item rbreak @var{file}:@var{regex}
3740 If @code{rbreak} is called with a filename qualification, it limits
3741 the search for functions matching the given regular expression to the
3742 specified @var{file}. This can be used, for example, to set breakpoints on
3743 every function in a given file:
3746 (@value{GDBP}) rbreak file.c:.
3749 The colon separating the filename qualifier from the regex may
3750 optionally be surrounded by spaces.
3752 @kindex info breakpoints
3753 @cindex @code{$_} and @code{info breakpoints}
3754 @item info breakpoints @r{[}@var{n}@dots{}@r{]}
3755 @itemx info break @r{[}@var{n}@dots{}@r{]}
3756 Print a table of all breakpoints, watchpoints, and catchpoints set and
3757 not deleted. Optional argument @var{n} means print information only
3758 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
3759 For each breakpoint, following columns are printed:
3762 @item Breakpoint Numbers
3764 Breakpoint, watchpoint, or catchpoint.
3766 Whether the breakpoint is marked to be disabled or deleted when hit.
3767 @item Enabled or Disabled
3768 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3769 that are not enabled.
3771 Where the breakpoint is in your program, as a memory address. For a
3772 pending breakpoint whose address is not yet known, this field will
3773 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3774 library that has the symbol or line referred by breakpoint is loaded.
3775 See below for details. A breakpoint with several locations will
3776 have @samp{<MULTIPLE>} in this field---see below for details.
3778 Where the breakpoint is in the source for your program, as a file and
3779 line number. For a pending breakpoint, the original string passed to
3780 the breakpoint command will be listed as it cannot be resolved until
3781 the appropriate shared library is loaded in the future.
3785 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
3786 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
3787 @value{GDBN} on the host's side. If it is ``target'', then the condition
3788 is evaluated by the target. The @code{info break} command shows
3789 the condition on the line following the affected breakpoint, together with
3790 its condition evaluation mode in between parentheses.
3792 Breakpoint commands, if any, are listed after that. A pending breakpoint is
3793 allowed to have a condition specified for it. The condition is not parsed for
3794 validity until a shared library is loaded that allows the pending
3795 breakpoint to resolve to a valid location.
3798 @code{info break} with a breakpoint
3799 number @var{n} as argument lists only that breakpoint. The
3800 convenience variable @code{$_} and the default examining-address for
3801 the @code{x} command are set to the address of the last breakpoint
3802 listed (@pxref{Memory, ,Examining Memory}).
3805 @code{info break} displays a count of the number of times the breakpoint
3806 has been hit. This is especially useful in conjunction with the
3807 @code{ignore} command. You can ignore a large number of breakpoint
3808 hits, look at the breakpoint info to see how many times the breakpoint
3809 was hit, and then run again, ignoring one less than that number. This
3810 will get you quickly to the last hit of that breakpoint.
3813 For a breakpoints with an enable count (xref) greater than 1,
3814 @code{info break} also displays that count.
3818 @value{GDBN} allows you to set any number of breakpoints at the same place in
3819 your program. There is nothing silly or meaningless about this. When
3820 the breakpoints are conditional, this is even useful
3821 (@pxref{Conditions, ,Break Conditions}).
3823 @cindex multiple locations, breakpoints
3824 @cindex breakpoints, multiple locations
3825 It is possible that a breakpoint corresponds to several locations
3826 in your program. Examples of this situation are:
3830 Multiple functions in the program may have the same name.
3833 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3834 instances of the function body, used in different cases.
3837 For a C@t{++} template function, a given line in the function can
3838 correspond to any number of instantiations.
3841 For an inlined function, a given source line can correspond to
3842 several places where that function is inlined.
3845 In all those cases, @value{GDBN} will insert a breakpoint at all
3846 the relevant locations.
3848 A breakpoint with multiple locations is displayed in the breakpoint
3849 table using several rows---one header row, followed by one row for
3850 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3851 address column. The rows for individual locations contain the actual
3852 addresses for locations, and show the functions to which those
3853 locations belong. The number column for a location is of the form
3854 @var{breakpoint-number}.@var{location-number}.
3859 Num Type Disp Enb Address What
3860 1 breakpoint keep y <MULTIPLE>
3862 breakpoint already hit 1 time
3863 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3864 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3867 Each location can be individually enabled or disabled by passing
3868 @var{breakpoint-number}.@var{location-number} as argument to the
3869 @code{enable} and @code{disable} commands. Note that you cannot
3870 delete the individual locations from the list, you can only delete the
3871 entire list of locations that belong to their parent breakpoint (with
3872 the @kbd{delete @var{num}} command, where @var{num} is the number of
3873 the parent breakpoint, 1 in the above example). Disabling or enabling
3874 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3875 that belong to that breakpoint.
3877 @cindex pending breakpoints
3878 It's quite common to have a breakpoint inside a shared library.
3879 Shared libraries can be loaded and unloaded explicitly,
3880 and possibly repeatedly, as the program is executed. To support
3881 this use case, @value{GDBN} updates breakpoint locations whenever
3882 any shared library is loaded or unloaded. Typically, you would
3883 set a breakpoint in a shared library at the beginning of your
3884 debugging session, when the library is not loaded, and when the
3885 symbols from the library are not available. When you try to set
3886 breakpoint, @value{GDBN} will ask you if you want to set
3887 a so called @dfn{pending breakpoint}---breakpoint whose address
3888 is not yet resolved.
3890 After the program is run, whenever a new shared library is loaded,
3891 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3892 shared library contains the symbol or line referred to by some
3893 pending breakpoint, that breakpoint is resolved and becomes an
3894 ordinary breakpoint. When a library is unloaded, all breakpoints
3895 that refer to its symbols or source lines become pending again.
3897 This logic works for breakpoints with multiple locations, too. For
3898 example, if you have a breakpoint in a C@t{++} template function, and
3899 a newly loaded shared library has an instantiation of that template,
3900 a new location is added to the list of locations for the breakpoint.
3902 Except for having unresolved address, pending breakpoints do not
3903 differ from regular breakpoints. You can set conditions or commands,
3904 enable and disable them and perform other breakpoint operations.
3906 @value{GDBN} provides some additional commands for controlling what
3907 happens when the @samp{break} command cannot resolve breakpoint
3908 address specification to an address:
3910 @kindex set breakpoint pending
3911 @kindex show breakpoint pending
3913 @item set breakpoint pending auto
3914 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3915 location, it queries you whether a pending breakpoint should be created.
3917 @item set breakpoint pending on
3918 This indicates that an unrecognized breakpoint location should automatically
3919 result in a pending breakpoint being created.
3921 @item set breakpoint pending off
3922 This indicates that pending breakpoints are not to be created. Any
3923 unrecognized breakpoint location results in an error. This setting does
3924 not affect any pending breakpoints previously created.
3926 @item show breakpoint pending
3927 Show the current behavior setting for creating pending breakpoints.
3930 The settings above only affect the @code{break} command and its
3931 variants. Once breakpoint is set, it will be automatically updated
3932 as shared libraries are loaded and unloaded.
3934 @cindex automatic hardware breakpoints
3935 For some targets, @value{GDBN} can automatically decide if hardware or
3936 software breakpoints should be used, depending on whether the
3937 breakpoint address is read-only or read-write. This applies to
3938 breakpoints set with the @code{break} command as well as to internal
3939 breakpoints set by commands like @code{next} and @code{finish}. For
3940 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3943 You can control this automatic behaviour with the following commands::
3945 @kindex set breakpoint auto-hw
3946 @kindex show breakpoint auto-hw
3948 @item set breakpoint auto-hw on
3949 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3950 will try to use the target memory map to decide if software or hardware
3951 breakpoint must be used.
3953 @item set breakpoint auto-hw off
3954 This indicates @value{GDBN} should not automatically select breakpoint
3955 type. If the target provides a memory map, @value{GDBN} will warn when
3956 trying to set software breakpoint at a read-only address.
3959 @value{GDBN} normally implements breakpoints by replacing the program code
3960 at the breakpoint address with a special instruction, which, when
3961 executed, given control to the debugger. By default, the program
3962 code is so modified only when the program is resumed. As soon as
3963 the program stops, @value{GDBN} restores the original instructions. This
3964 behaviour guards against leaving breakpoints inserted in the
3965 target should gdb abrubptly disconnect. However, with slow remote
3966 targets, inserting and removing breakpoint can reduce the performance.
3967 This behavior can be controlled with the following commands::
3969 @kindex set breakpoint always-inserted
3970 @kindex show breakpoint always-inserted
3972 @item set breakpoint always-inserted off
3973 All breakpoints, including newly added by the user, are inserted in
3974 the target only when the target is resumed. All breakpoints are
3975 removed from the target when it stops. This is the default mode.
3977 @item set breakpoint always-inserted on
3978 Causes all breakpoints to be inserted in the target at all times. If
3979 the user adds a new breakpoint, or changes an existing breakpoint, the
3980 breakpoints in the target are updated immediately. A breakpoint is
3981 removed from the target only when breakpoint itself is deleted.
3984 @value{GDBN} handles conditional breakpoints by evaluating these conditions
3985 when a breakpoint breaks. If the condition is true, then the process being
3986 debugged stops, otherwise the process is resumed.
3988 If the target supports evaluating conditions on its end, @value{GDBN} may
3989 download the breakpoint, together with its conditions, to it.
3991 This feature can be controlled via the following commands:
3993 @kindex set breakpoint condition-evaluation
3994 @kindex show breakpoint condition-evaluation
3996 @item set breakpoint condition-evaluation host
3997 This option commands @value{GDBN} to evaluate the breakpoint
3998 conditions on the host's side. Unconditional breakpoints are sent to
3999 the target which in turn receives the triggers and reports them back to GDB
4000 for condition evaluation. This is the standard evaluation mode.
4002 @item set breakpoint condition-evaluation target
4003 This option commands @value{GDBN} to download breakpoint conditions
4004 to the target at the moment of their insertion. The target
4005 is responsible for evaluating the conditional expression and reporting
4006 breakpoint stop events back to @value{GDBN} whenever the condition
4007 is true. Due to limitations of target-side evaluation, some conditions
4008 cannot be evaluated there, e.g., conditions that depend on local data
4009 that is only known to the host. Examples include
4010 conditional expressions involving convenience variables, complex types
4011 that cannot be handled by the agent expression parser and expressions
4012 that are too long to be sent over to the target, specially when the
4013 target is a remote system. In these cases, the conditions will be
4014 evaluated by @value{GDBN}.
4016 @item set breakpoint condition-evaluation auto
4017 This is the default mode. If the target supports evaluating breakpoint
4018 conditions on its end, @value{GDBN} will download breakpoint conditions to
4019 the target (limitations mentioned previously apply). If the target does
4020 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4021 to evaluating all these conditions on the host's side.
4025 @cindex negative breakpoint numbers
4026 @cindex internal @value{GDBN} breakpoints
4027 @value{GDBN} itself sometimes sets breakpoints in your program for
4028 special purposes, such as proper handling of @code{longjmp} (in C
4029 programs). These internal breakpoints are assigned negative numbers,
4030 starting with @code{-1}; @samp{info breakpoints} does not display them.
4031 You can see these breakpoints with the @value{GDBN} maintenance command
4032 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
4035 @node Set Watchpoints
4036 @subsection Setting Watchpoints
4038 @cindex setting watchpoints
4039 You can use a watchpoint to stop execution whenever the value of an
4040 expression changes, without having to predict a particular place where
4041 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
4042 The expression may be as simple as the value of a single variable, or
4043 as complex as many variables combined by operators. Examples include:
4047 A reference to the value of a single variable.
4050 An address cast to an appropriate data type. For example,
4051 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
4052 address (assuming an @code{int} occupies 4 bytes).
4055 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
4056 expression can use any operators valid in the program's native
4057 language (@pxref{Languages}).
4060 You can set a watchpoint on an expression even if the expression can
4061 not be evaluated yet. For instance, you can set a watchpoint on
4062 @samp{*global_ptr} before @samp{global_ptr} is initialized.
4063 @value{GDBN} will stop when your program sets @samp{global_ptr} and
4064 the expression produces a valid value. If the expression becomes
4065 valid in some other way than changing a variable (e.g.@: if the memory
4066 pointed to by @samp{*global_ptr} becomes readable as the result of a
4067 @code{malloc} call), @value{GDBN} may not stop until the next time
4068 the expression changes.
4070 @cindex software watchpoints
4071 @cindex hardware watchpoints
4072 Depending on your system, watchpoints may be implemented in software or
4073 hardware. @value{GDBN} does software watchpointing by single-stepping your
4074 program and testing the variable's value each time, which is hundreds of
4075 times slower than normal execution. (But this may still be worth it, to
4076 catch errors where you have no clue what part of your program is the
4079 On some systems, such as most PowerPC or x86-based targets,
4080 @value{GDBN} includes support for hardware watchpoints, which do not
4081 slow down the running of your program.
4085 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4086 Set a watchpoint for an expression. @value{GDBN} will break when the
4087 expression @var{expr} is written into by the program and its value
4088 changes. The simplest (and the most popular) use of this command is
4089 to watch the value of a single variable:
4092 (@value{GDBP}) watch foo
4095 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
4096 argument, @value{GDBN} breaks only when the thread identified by
4097 @var{thread-id} changes the value of @var{expr}. If any other threads
4098 change the value of @var{expr}, @value{GDBN} will not break. Note
4099 that watchpoints restricted to a single thread in this way only work
4100 with Hardware Watchpoints.
4102 Ordinarily a watchpoint respects the scope of variables in @var{expr}
4103 (see below). The @code{-location} argument tells @value{GDBN} to
4104 instead watch the memory referred to by @var{expr}. In this case,
4105 @value{GDBN} will evaluate @var{expr}, take the address of the result,
4106 and watch the memory at that address. The type of the result is used
4107 to determine the size of the watched memory. If the expression's
4108 result does not have an address, then @value{GDBN} will print an
4111 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
4112 of masked watchpoints, if the current architecture supports this
4113 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
4114 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
4115 to an address to watch. The mask specifies that some bits of an address
4116 (the bits which are reset in the mask) should be ignored when matching
4117 the address accessed by the inferior against the watchpoint address.
4118 Thus, a masked watchpoint watches many addresses simultaneously---those
4119 addresses whose unmasked bits are identical to the unmasked bits in the
4120 watchpoint address. The @code{mask} argument implies @code{-location}.
4124 (@value{GDBP}) watch foo mask 0xffff00ff
4125 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4129 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4130 Set a watchpoint that will break when the value of @var{expr} is read
4134 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4135 Set a watchpoint that will break when @var{expr} is either read from
4136 or written into by the program.
4138 @kindex info watchpoints @r{[}@var{n}@dots{}@r{]}
4139 @item info watchpoints @r{[}@var{n}@dots{}@r{]}
4140 This command prints a list of watchpoints, using the same format as
4141 @code{info break} (@pxref{Set Breaks}).
4144 If you watch for a change in a numerically entered address you need to
4145 dereference it, as the address itself is just a constant number which will
4146 never change. @value{GDBN} refuses to create a watchpoint that watches
4147 a never-changing value:
4150 (@value{GDBP}) watch 0x600850
4151 Cannot watch constant value 0x600850.
4152 (@value{GDBP}) watch *(int *) 0x600850
4153 Watchpoint 1: *(int *) 6293584
4156 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4157 watchpoints execute very quickly, and the debugger reports a change in
4158 value at the exact instruction where the change occurs. If @value{GDBN}
4159 cannot set a hardware watchpoint, it sets a software watchpoint, which
4160 executes more slowly and reports the change in value at the next
4161 @emph{statement}, not the instruction, after the change occurs.
4163 @cindex use only software watchpoints
4164 You can force @value{GDBN} to use only software watchpoints with the
4165 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4166 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4167 the underlying system supports them. (Note that hardware-assisted
4168 watchpoints that were set @emph{before} setting
4169 @code{can-use-hw-watchpoints} to zero will still use the hardware
4170 mechanism of watching expression values.)
4173 @item set can-use-hw-watchpoints
4174 @kindex set can-use-hw-watchpoints
4175 Set whether or not to use hardware watchpoints.
4177 @item show can-use-hw-watchpoints
4178 @kindex show can-use-hw-watchpoints
4179 Show the current mode of using hardware watchpoints.
4182 For remote targets, you can restrict the number of hardware
4183 watchpoints @value{GDBN} will use, see @ref{set remote
4184 hardware-breakpoint-limit}.
4186 When you issue the @code{watch} command, @value{GDBN} reports
4189 Hardware watchpoint @var{num}: @var{expr}
4193 if it was able to set a hardware watchpoint.
4195 Currently, the @code{awatch} and @code{rwatch} commands can only set
4196 hardware watchpoints, because accesses to data that don't change the
4197 value of the watched expression cannot be detected without examining
4198 every instruction as it is being executed, and @value{GDBN} does not do
4199 that currently. If @value{GDBN} finds that it is unable to set a
4200 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4201 will print a message like this:
4204 Expression cannot be implemented with read/access watchpoint.
4207 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4208 data type of the watched expression is wider than what a hardware
4209 watchpoint on the target machine can handle. For example, some systems
4210 can only watch regions that are up to 4 bytes wide; on such systems you
4211 cannot set hardware watchpoints for an expression that yields a
4212 double-precision floating-point number (which is typically 8 bytes
4213 wide). As a work-around, it might be possible to break the large region
4214 into a series of smaller ones and watch them with separate watchpoints.
4216 If you set too many hardware watchpoints, @value{GDBN} might be unable
4217 to insert all of them when you resume the execution of your program.
4218 Since the precise number of active watchpoints is unknown until such
4219 time as the program is about to be resumed, @value{GDBN} might not be
4220 able to warn you about this when you set the watchpoints, and the
4221 warning will be printed only when the program is resumed:
4224 Hardware watchpoint @var{num}: Could not insert watchpoint
4228 If this happens, delete or disable some of the watchpoints.
4230 Watching complex expressions that reference many variables can also
4231 exhaust the resources available for hardware-assisted watchpoints.
4232 That's because @value{GDBN} needs to watch every variable in the
4233 expression with separately allocated resources.
4235 If you call a function interactively using @code{print} or @code{call},
4236 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4237 kind of breakpoint or the call completes.
4239 @value{GDBN} automatically deletes watchpoints that watch local
4240 (automatic) variables, or expressions that involve such variables, when
4241 they go out of scope, that is, when the execution leaves the block in
4242 which these variables were defined. In particular, when the program
4243 being debugged terminates, @emph{all} local variables go out of scope,
4244 and so only watchpoints that watch global variables remain set. If you
4245 rerun the program, you will need to set all such watchpoints again. One
4246 way of doing that would be to set a code breakpoint at the entry to the
4247 @code{main} function and when it breaks, set all the watchpoints.
4249 @cindex watchpoints and threads
4250 @cindex threads and watchpoints
4251 In multi-threaded programs, watchpoints will detect changes to the
4252 watched expression from every thread.
4255 @emph{Warning:} In multi-threaded programs, software watchpoints
4256 have only limited usefulness. If @value{GDBN} creates a software
4257 watchpoint, it can only watch the value of an expression @emph{in a
4258 single thread}. If you are confident that the expression can only
4259 change due to the current thread's activity (and if you are also
4260 confident that no other thread can become current), then you can use
4261 software watchpoints as usual. However, @value{GDBN} may not notice
4262 when a non-current thread's activity changes the expression. (Hardware
4263 watchpoints, in contrast, watch an expression in all threads.)
4266 @xref{set remote hardware-watchpoint-limit}.
4268 @node Set Catchpoints
4269 @subsection Setting Catchpoints
4270 @cindex catchpoints, setting
4271 @cindex exception handlers
4272 @cindex event handling
4274 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4275 kinds of program events, such as C@t{++} exceptions or the loading of a
4276 shared library. Use the @code{catch} command to set a catchpoint.
4280 @item catch @var{event}
4281 Stop when @var{event} occurs. The @var{event} can be any of the following:
4284 @item throw @r{[}@var{regexp}@r{]}
4285 @itemx rethrow @r{[}@var{regexp}@r{]}
4286 @itemx catch @r{[}@var{regexp}@r{]}
4288 @kindex catch rethrow
4290 @cindex stop on C@t{++} exceptions
4291 The throwing, re-throwing, or catching of a C@t{++} exception.
4293 If @var{regexp} is given, then only exceptions whose type matches the
4294 regular expression will be caught.
4296 @vindex $_exception@r{, convenience variable}
4297 The convenience variable @code{$_exception} is available at an
4298 exception-related catchpoint, on some systems. This holds the
4299 exception being thrown.
4301 There are currently some limitations to C@t{++} exception handling in
4306 The support for these commands is system-dependent. Currently, only
4307 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4311 The regular expression feature and the @code{$_exception} convenience
4312 variable rely on the presence of some SDT probes in @code{libstdc++}.
4313 If these probes are not present, then these features cannot be used.
4314 These probes were first available in the GCC 4.8 release, but whether
4315 or not they are available in your GCC also depends on how it was
4319 The @code{$_exception} convenience variable is only valid at the
4320 instruction at which an exception-related catchpoint is set.
4323 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4324 location in the system library which implements runtime exception
4325 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4326 (@pxref{Selection}) to get to your code.
4329 If you call a function interactively, @value{GDBN} normally returns
4330 control to you when the function has finished executing. If the call
4331 raises an exception, however, the call may bypass the mechanism that
4332 returns control to you and cause your program either to abort or to
4333 simply continue running until it hits a breakpoint, catches a signal
4334 that @value{GDBN} is listening for, or exits. This is the case even if
4335 you set a catchpoint for the exception; catchpoints on exceptions are
4336 disabled within interactive calls. @xref{Calling}, for information on
4337 controlling this with @code{set unwind-on-terminating-exception}.
4340 You cannot raise an exception interactively.
4343 You cannot install an exception handler interactively.
4347 @kindex catch exception
4348 @cindex Ada exception catching
4349 @cindex catch Ada exceptions
4350 An Ada exception being raised. If an exception name is specified
4351 at the end of the command (eg @code{catch exception Program_Error}),
4352 the debugger will stop only when this specific exception is raised.
4353 Otherwise, the debugger stops execution when any Ada exception is raised.
4355 When inserting an exception catchpoint on a user-defined exception whose
4356 name is identical to one of the exceptions defined by the language, the
4357 fully qualified name must be used as the exception name. Otherwise,
4358 @value{GDBN} will assume that it should stop on the pre-defined exception
4359 rather than the user-defined one. For instance, assuming an exception
4360 called @code{Constraint_Error} is defined in package @code{Pck}, then
4361 the command to use to catch such exceptions is @kbd{catch exception
4362 Pck.Constraint_Error}.
4364 @item exception unhandled
4365 @kindex catch exception unhandled
4366 An exception that was raised but is not handled by the program.
4369 @kindex catch assert
4370 A failed Ada assertion.
4374 @cindex break on fork/exec
4375 A call to @code{exec}.
4378 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @dots{}
4379 @kindex catch syscall
4380 @cindex break on a system call.
4381 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4382 syscall is a mechanism for application programs to request a service
4383 from the operating system (OS) or one of the OS system services.
4384 @value{GDBN} can catch some or all of the syscalls issued by the
4385 debuggee, and show the related information for each syscall. If no
4386 argument is specified, calls to and returns from all system calls
4389 @var{name} can be any system call name that is valid for the
4390 underlying OS. Just what syscalls are valid depends on the OS. On
4391 GNU and Unix systems, you can find the full list of valid syscall
4392 names on @file{/usr/include/asm/unistd.h}.
4394 @c For MS-Windows, the syscall names and the corresponding numbers
4395 @c can be found, e.g., on this URL:
4396 @c http://www.metasploit.com/users/opcode/syscalls.html
4397 @c but we don't support Windows syscalls yet.
4399 Normally, @value{GDBN} knows in advance which syscalls are valid for
4400 each OS, so you can use the @value{GDBN} command-line completion
4401 facilities (@pxref{Completion,, command completion}) to list the
4404 You may also specify the system call numerically. A syscall's
4405 number is the value passed to the OS's syscall dispatcher to
4406 identify the requested service. When you specify the syscall by its
4407 name, @value{GDBN} uses its database of syscalls to convert the name
4408 into the corresponding numeric code, but using the number directly
4409 may be useful if @value{GDBN}'s database does not have the complete
4410 list of syscalls on your system (e.g., because @value{GDBN} lags
4411 behind the OS upgrades).
4413 The example below illustrates how this command works if you don't provide
4417 (@value{GDBP}) catch syscall
4418 Catchpoint 1 (syscall)
4420 Starting program: /tmp/catch-syscall
4422 Catchpoint 1 (call to syscall 'close'), \
4423 0xffffe424 in __kernel_vsyscall ()
4427 Catchpoint 1 (returned from syscall 'close'), \
4428 0xffffe424 in __kernel_vsyscall ()
4432 Here is an example of catching a system call by name:
4435 (@value{GDBP}) catch syscall chroot
4436 Catchpoint 1 (syscall 'chroot' [61])
4438 Starting program: /tmp/catch-syscall
4440 Catchpoint 1 (call to syscall 'chroot'), \
4441 0xffffe424 in __kernel_vsyscall ()
4445 Catchpoint 1 (returned from syscall 'chroot'), \
4446 0xffffe424 in __kernel_vsyscall ()
4450 An example of specifying a system call numerically. In the case
4451 below, the syscall number has a corresponding entry in the XML
4452 file, so @value{GDBN} finds its name and prints it:
4455 (@value{GDBP}) catch syscall 252
4456 Catchpoint 1 (syscall(s) 'exit_group')
4458 Starting program: /tmp/catch-syscall
4460 Catchpoint 1 (call to syscall 'exit_group'), \
4461 0xffffe424 in __kernel_vsyscall ()
4465 Program exited normally.
4469 However, there can be situations when there is no corresponding name
4470 in XML file for that syscall number. In this case, @value{GDBN} prints
4471 a warning message saying that it was not able to find the syscall name,
4472 but the catchpoint will be set anyway. See the example below:
4475 (@value{GDBP}) catch syscall 764
4476 warning: The number '764' does not represent a known syscall.
4477 Catchpoint 2 (syscall 764)
4481 If you configure @value{GDBN} using the @samp{--without-expat} option,
4482 it will not be able to display syscall names. Also, if your
4483 architecture does not have an XML file describing its system calls,
4484 you will not be able to see the syscall names. It is important to
4485 notice that these two features are used for accessing the syscall
4486 name database. In either case, you will see a warning like this:
4489 (@value{GDBP}) catch syscall
4490 warning: Could not open "syscalls/i386-linux.xml"
4491 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4492 GDB will not be able to display syscall names.
4493 Catchpoint 1 (syscall)
4497 Of course, the file name will change depending on your architecture and system.
4499 Still using the example above, you can also try to catch a syscall by its
4500 number. In this case, you would see something like:
4503 (@value{GDBP}) catch syscall 252
4504 Catchpoint 1 (syscall(s) 252)
4507 Again, in this case @value{GDBN} would not be able to display syscall's names.
4511 A call to @code{fork}.
4515 A call to @code{vfork}.
4517 @item load @r{[}regexp@r{]}
4518 @itemx unload @r{[}regexp@r{]}
4520 @kindex catch unload
4521 The loading or unloading of a shared library. If @var{regexp} is
4522 given, then the catchpoint will stop only if the regular expression
4523 matches one of the affected libraries.
4525 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
4526 @kindex catch signal
4527 The delivery of a signal.
4529 With no arguments, this catchpoint will catch any signal that is not
4530 used internally by @value{GDBN}, specifically, all signals except
4531 @samp{SIGTRAP} and @samp{SIGINT}.
4533 With the argument @samp{all}, all signals, including those used by
4534 @value{GDBN}, will be caught. This argument cannot be used with other
4537 Otherwise, the arguments are a list of signal names as given to
4538 @code{handle} (@pxref{Signals}). Only signals specified in this list
4541 One reason that @code{catch signal} can be more useful than
4542 @code{handle} is that you can attach commands and conditions to the
4545 When a signal is caught by a catchpoint, the signal's @code{stop} and
4546 @code{print} settings, as specified by @code{handle}, are ignored.
4547 However, whether the signal is still delivered to the inferior depends
4548 on the @code{pass} setting; this can be changed in the catchpoint's
4553 @item tcatch @var{event}
4555 Set a catchpoint that is enabled only for one stop. The catchpoint is
4556 automatically deleted after the first time the event is caught.
4560 Use the @code{info break} command to list the current catchpoints.
4564 @subsection Deleting Breakpoints
4566 @cindex clearing breakpoints, watchpoints, catchpoints
4567 @cindex deleting breakpoints, watchpoints, catchpoints
4568 It is often necessary to eliminate a breakpoint, watchpoint, or
4569 catchpoint once it has done its job and you no longer want your program
4570 to stop there. This is called @dfn{deleting} the breakpoint. A
4571 breakpoint that has been deleted no longer exists; it is forgotten.
4573 With the @code{clear} command you can delete breakpoints according to
4574 where they are in your program. With the @code{delete} command you can
4575 delete individual breakpoints, watchpoints, or catchpoints by specifying
4576 their breakpoint numbers.
4578 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4579 automatically ignores breakpoints on the first instruction to be executed
4580 when you continue execution without changing the execution address.
4585 Delete any breakpoints at the next instruction to be executed in the
4586 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4587 the innermost frame is selected, this is a good way to delete a
4588 breakpoint where your program just stopped.
4590 @item clear @var{location}
4591 Delete any breakpoints set at the specified @var{location}.
4592 @xref{Specify Location}, for the various forms of @var{location}; the
4593 most useful ones are listed below:
4596 @item clear @var{function}
4597 @itemx clear @var{filename}:@var{function}
4598 Delete any breakpoints set at entry to the named @var{function}.
4600 @item clear @var{linenum}
4601 @itemx clear @var{filename}:@var{linenum}
4602 Delete any breakpoints set at or within the code of the specified
4603 @var{linenum} of the specified @var{filename}.
4606 @cindex delete breakpoints
4608 @kindex d @r{(@code{delete})}
4609 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4610 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4611 ranges specified as arguments. If no argument is specified, delete all
4612 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4613 confirm off}). You can abbreviate this command as @code{d}.
4617 @subsection Disabling Breakpoints
4619 @cindex enable/disable a breakpoint
4620 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4621 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4622 it had been deleted, but remembers the information on the breakpoint so
4623 that you can @dfn{enable} it again later.
4625 You disable and enable breakpoints, watchpoints, and catchpoints with
4626 the @code{enable} and @code{disable} commands, optionally specifying
4627 one or more breakpoint numbers as arguments. Use @code{info break} to
4628 print a list of all breakpoints, watchpoints, and catchpoints if you
4629 do not know which numbers to use.
4631 Disabling and enabling a breakpoint that has multiple locations
4632 affects all of its locations.
4634 A breakpoint, watchpoint, or catchpoint can have any of several
4635 different states of enablement:
4639 Enabled. The breakpoint stops your program. A breakpoint set
4640 with the @code{break} command starts out in this state.
4642 Disabled. The breakpoint has no effect on your program.
4644 Enabled once. The breakpoint stops your program, but then becomes
4647 Enabled for a count. The breakpoint stops your program for the next
4648 N times, then becomes disabled.
4650 Enabled for deletion. The breakpoint stops your program, but
4651 immediately after it does so it is deleted permanently. A breakpoint
4652 set with the @code{tbreak} command starts out in this state.
4655 You can use the following commands to enable or disable breakpoints,
4656 watchpoints, and catchpoints:
4660 @kindex dis @r{(@code{disable})}
4661 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4662 Disable the specified breakpoints---or all breakpoints, if none are
4663 listed. A disabled breakpoint has no effect but is not forgotten. All
4664 options such as ignore-counts, conditions and commands are remembered in
4665 case the breakpoint is enabled again later. You may abbreviate
4666 @code{disable} as @code{dis}.
4669 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4670 Enable the specified breakpoints (or all defined breakpoints). They
4671 become effective once again in stopping your program.
4673 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4674 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4675 of these breakpoints immediately after stopping your program.
4677 @item enable @r{[}breakpoints@r{]} count @var{count} @var{range}@dots{}
4678 Enable the specified breakpoints temporarily. @value{GDBN} records
4679 @var{count} with each of the specified breakpoints, and decrements a
4680 breakpoint's count when it is hit. When any count reaches 0,
4681 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
4682 count (@pxref{Conditions, ,Break Conditions}), that will be
4683 decremented to 0 before @var{count} is affected.
4685 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4686 Enable the specified breakpoints to work once, then die. @value{GDBN}
4687 deletes any of these breakpoints as soon as your program stops there.
4688 Breakpoints set by the @code{tbreak} command start out in this state.
4691 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4692 @c confusing: tbreak is also initially enabled.
4693 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4694 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4695 subsequently, they become disabled or enabled only when you use one of
4696 the commands above. (The command @code{until} can set and delete a
4697 breakpoint of its own, but it does not change the state of your other
4698 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4702 @subsection Break Conditions
4703 @cindex conditional breakpoints
4704 @cindex breakpoint conditions
4706 @c FIXME what is scope of break condition expr? Context where wanted?
4707 @c in particular for a watchpoint?
4708 The simplest sort of breakpoint breaks every time your program reaches a
4709 specified place. You can also specify a @dfn{condition} for a
4710 breakpoint. A condition is just a Boolean expression in your
4711 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4712 a condition evaluates the expression each time your program reaches it,
4713 and your program stops only if the condition is @emph{true}.
4715 This is the converse of using assertions for program validation; in that
4716 situation, you want to stop when the assertion is violated---that is,
4717 when the condition is false. In C, if you want to test an assertion expressed
4718 by the condition @var{assert}, you should set the condition
4719 @samp{! @var{assert}} on the appropriate breakpoint.
4721 Conditions are also accepted for watchpoints; you may not need them,
4722 since a watchpoint is inspecting the value of an expression anyhow---but
4723 it might be simpler, say, to just set a watchpoint on a variable name,
4724 and specify a condition that tests whether the new value is an interesting
4727 Break conditions can have side effects, and may even call functions in
4728 your program. This can be useful, for example, to activate functions
4729 that log program progress, or to use your own print functions to
4730 format special data structures. The effects are completely predictable
4731 unless there is another enabled breakpoint at the same address. (In
4732 that case, @value{GDBN} might see the other breakpoint first and stop your
4733 program without checking the condition of this one.) Note that
4734 breakpoint commands are usually more convenient and flexible than break
4736 purpose of performing side effects when a breakpoint is reached
4737 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4739 Breakpoint conditions can also be evaluated on the target's side if
4740 the target supports it. Instead of evaluating the conditions locally,
4741 @value{GDBN} encodes the expression into an agent expression
4742 (@pxref{Agent Expressions}) suitable for execution on the target,
4743 independently of @value{GDBN}. Global variables become raw memory
4744 locations, locals become stack accesses, and so forth.
4746 In this case, @value{GDBN} will only be notified of a breakpoint trigger
4747 when its condition evaluates to true. This mechanism may provide faster
4748 response times depending on the performance characteristics of the target
4749 since it does not need to keep @value{GDBN} informed about
4750 every breakpoint trigger, even those with false conditions.
4752 Break conditions can be specified when a breakpoint is set, by using
4753 @samp{if} in the arguments to the @code{break} command. @xref{Set
4754 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4755 with the @code{condition} command.
4757 You can also use the @code{if} keyword with the @code{watch} command.
4758 The @code{catch} command does not recognize the @code{if} keyword;
4759 @code{condition} is the only way to impose a further condition on a
4764 @item condition @var{bnum} @var{expression}
4765 Specify @var{expression} as the break condition for breakpoint,
4766 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4767 breakpoint @var{bnum} stops your program only if the value of
4768 @var{expression} is true (nonzero, in C). When you use
4769 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4770 syntactic correctness, and to determine whether symbols in it have
4771 referents in the context of your breakpoint. If @var{expression} uses
4772 symbols not referenced in the context of the breakpoint, @value{GDBN}
4773 prints an error message:
4776 No symbol "foo" in current context.
4781 not actually evaluate @var{expression} at the time the @code{condition}
4782 command (or a command that sets a breakpoint with a condition, like
4783 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4785 @item condition @var{bnum}
4786 Remove the condition from breakpoint number @var{bnum}. It becomes
4787 an ordinary unconditional breakpoint.
4790 @cindex ignore count (of breakpoint)
4791 A special case of a breakpoint condition is to stop only when the
4792 breakpoint has been reached a certain number of times. This is so
4793 useful that there is a special way to do it, using the @dfn{ignore
4794 count} of the breakpoint. Every breakpoint has an ignore count, which
4795 is an integer. Most of the time, the ignore count is zero, and
4796 therefore has no effect. But if your program reaches a breakpoint whose
4797 ignore count is positive, then instead of stopping, it just decrements
4798 the ignore count by one and continues. As a result, if the ignore count
4799 value is @var{n}, the breakpoint does not stop the next @var{n} times
4800 your program reaches it.
4804 @item ignore @var{bnum} @var{count}
4805 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4806 The next @var{count} times the breakpoint is reached, your program's
4807 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4810 To make the breakpoint stop the next time it is reached, specify
4813 When you use @code{continue} to resume execution of your program from a
4814 breakpoint, you can specify an ignore count directly as an argument to
4815 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4816 Stepping,,Continuing and Stepping}.
4818 If a breakpoint has a positive ignore count and a condition, the
4819 condition is not checked. Once the ignore count reaches zero,
4820 @value{GDBN} resumes checking the condition.
4822 You could achieve the effect of the ignore count with a condition such
4823 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4824 is decremented each time. @xref{Convenience Vars, ,Convenience
4828 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4831 @node Break Commands
4832 @subsection Breakpoint Command Lists
4834 @cindex breakpoint commands
4835 You can give any breakpoint (or watchpoint or catchpoint) a series of
4836 commands to execute when your program stops due to that breakpoint. For
4837 example, you might want to print the values of certain expressions, or
4838 enable other breakpoints.
4842 @kindex end@r{ (breakpoint commands)}
4843 @item commands @r{[}@var{range}@dots{}@r{]}
4844 @itemx @dots{} @var{command-list} @dots{}
4846 Specify a list of commands for the given breakpoints. The commands
4847 themselves appear on the following lines. Type a line containing just
4848 @code{end} to terminate the commands.
4850 To remove all commands from a breakpoint, type @code{commands} and
4851 follow it immediately with @code{end}; that is, give no commands.
4853 With no argument, @code{commands} refers to the last breakpoint,
4854 watchpoint, or catchpoint set (not to the breakpoint most recently
4855 encountered). If the most recent breakpoints were set with a single
4856 command, then the @code{commands} will apply to all the breakpoints
4857 set by that command. This applies to breakpoints set by
4858 @code{rbreak}, and also applies when a single @code{break} command
4859 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4863 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4864 disabled within a @var{command-list}.
4866 You can use breakpoint commands to start your program up again. Simply
4867 use the @code{continue} command, or @code{step}, or any other command
4868 that resumes execution.
4870 Any other commands in the command list, after a command that resumes
4871 execution, are ignored. This is because any time you resume execution
4872 (even with a simple @code{next} or @code{step}), you may encounter
4873 another breakpoint---which could have its own command list, leading to
4874 ambiguities about which list to execute.
4877 If the first command you specify in a command list is @code{silent}, the
4878 usual message about stopping at a breakpoint is not printed. This may
4879 be desirable for breakpoints that are to print a specific message and
4880 then continue. If none of the remaining commands print anything, you
4881 see no sign that the breakpoint was reached. @code{silent} is
4882 meaningful only at the beginning of a breakpoint command list.
4884 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4885 print precisely controlled output, and are often useful in silent
4886 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4888 For example, here is how you could use breakpoint commands to print the
4889 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4895 printf "x is %d\n",x
4900 One application for breakpoint commands is to compensate for one bug so
4901 you can test for another. Put a breakpoint just after the erroneous line
4902 of code, give it a condition to detect the case in which something
4903 erroneous has been done, and give it commands to assign correct values
4904 to any variables that need them. End with the @code{continue} command
4905 so that your program does not stop, and start with the @code{silent}
4906 command so that no output is produced. Here is an example:
4917 @node Dynamic Printf
4918 @subsection Dynamic Printf
4920 @cindex dynamic printf
4922 The dynamic printf command @code{dprintf} combines a breakpoint with
4923 formatted printing of your program's data to give you the effect of
4924 inserting @code{printf} calls into your program on-the-fly, without
4925 having to recompile it.
4927 In its most basic form, the output goes to the GDB console. However,
4928 you can set the variable @code{dprintf-style} for alternate handling.
4929 For instance, you can ask to format the output by calling your
4930 program's @code{printf} function. This has the advantage that the
4931 characters go to the program's output device, so they can recorded in
4932 redirects to files and so forth.
4934 If you are doing remote debugging with a stub or agent, you can also
4935 ask to have the printf handled by the remote agent. In addition to
4936 ensuring that the output goes to the remote program's device along
4937 with any other output the program might produce, you can also ask that
4938 the dprintf remain active even after disconnecting from the remote
4939 target. Using the stub/agent is also more efficient, as it can do
4940 everything without needing to communicate with @value{GDBN}.
4944 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
4945 Whenever execution reaches @var{location}, print the values of one or
4946 more @var{expressions} under the control of the string @var{template}.
4947 To print several values, separate them with commas.
4949 @item set dprintf-style @var{style}
4950 Set the dprintf output to be handled in one of several different
4951 styles enumerated below. A change of style affects all existing
4952 dynamic printfs immediately. (If you need individual control over the
4953 print commands, simply define normal breakpoints with
4954 explicitly-supplied command lists.)
4957 @kindex dprintf-style gdb
4958 Handle the output using the @value{GDBN} @code{printf} command.
4961 @kindex dprintf-style call
4962 Handle the output by calling a function in your program (normally
4966 @kindex dprintf-style agent
4967 Have the remote debugging agent (such as @code{gdbserver}) handle
4968 the output itself. This style is only available for agents that
4969 support running commands on the target.
4971 @item set dprintf-function @var{function}
4972 Set the function to call if the dprintf style is @code{call}. By
4973 default its value is @code{printf}. You may set it to any expression.
4974 that @value{GDBN} can evaluate to a function, as per the @code{call}
4977 @item set dprintf-channel @var{channel}
4978 Set a ``channel'' for dprintf. If set to a non-empty value,
4979 @value{GDBN} will evaluate it as an expression and pass the result as
4980 a first argument to the @code{dprintf-function}, in the manner of
4981 @code{fprintf} and similar functions. Otherwise, the dprintf format
4982 string will be the first argument, in the manner of @code{printf}.
4984 As an example, if you wanted @code{dprintf} output to go to a logfile
4985 that is a standard I/O stream assigned to the variable @code{mylog},
4986 you could do the following:
4989 (gdb) set dprintf-style call
4990 (gdb) set dprintf-function fprintf
4991 (gdb) set dprintf-channel mylog
4992 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
4993 Dprintf 1 at 0x123456: file main.c, line 25.
4995 1 dprintf keep y 0x00123456 in main at main.c:25
4996 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
5001 Note that the @code{info break} displays the dynamic printf commands
5002 as normal breakpoint commands; you can thus easily see the effect of
5003 the variable settings.
5005 @item set disconnected-dprintf on
5006 @itemx set disconnected-dprintf off
5007 @kindex set disconnected-dprintf
5008 Choose whether @code{dprintf} commands should continue to run if
5009 @value{GDBN} has disconnected from the target. This only applies
5010 if the @code{dprintf-style} is @code{agent}.
5012 @item show disconnected-dprintf off
5013 @kindex show disconnected-dprintf
5014 Show the current choice for disconnected @code{dprintf}.
5018 @value{GDBN} does not check the validity of function and channel,
5019 relying on you to supply values that are meaningful for the contexts
5020 in which they are being used. For instance, the function and channel
5021 may be the values of local variables, but if that is the case, then
5022 all enabled dynamic prints must be at locations within the scope of
5023 those locals. If evaluation fails, @value{GDBN} will report an error.
5025 @node Save Breakpoints
5026 @subsection How to save breakpoints to a file
5028 To save breakpoint definitions to a file use the @w{@code{save
5029 breakpoints}} command.
5032 @kindex save breakpoints
5033 @cindex save breakpoints to a file for future sessions
5034 @item save breakpoints [@var{filename}]
5035 This command saves all current breakpoint definitions together with
5036 their commands and ignore counts, into a file @file{@var{filename}}
5037 suitable for use in a later debugging session. This includes all
5038 types of breakpoints (breakpoints, watchpoints, catchpoints,
5039 tracepoints). To read the saved breakpoint definitions, use the
5040 @code{source} command (@pxref{Command Files}). Note that watchpoints
5041 with expressions involving local variables may fail to be recreated
5042 because it may not be possible to access the context where the
5043 watchpoint is valid anymore. Because the saved breakpoint definitions
5044 are simply a sequence of @value{GDBN} commands that recreate the
5045 breakpoints, you can edit the file in your favorite editing program,
5046 and remove the breakpoint definitions you're not interested in, or
5047 that can no longer be recreated.
5050 @node Static Probe Points
5051 @subsection Static Probe Points
5053 @cindex static probe point, SystemTap
5054 @cindex static probe point, DTrace
5055 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
5056 for Statically Defined Tracing, and the probes are designed to have a tiny
5057 runtime code and data footprint, and no dynamic relocations.
5059 Currently, the following types of probes are supported on
5060 ELF-compatible systems:
5064 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
5065 @acronym{SDT} probes@footnote{See
5066 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
5067 for more information on how to add @code{SystemTap} @acronym{SDT}
5068 probes in your applications.}. @code{SystemTap} probes are usable
5069 from assembly, C and C@t{++} languages@footnote{See
5070 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
5071 for a good reference on how the @acronym{SDT} probes are implemented.}.
5073 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
5074 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
5078 @cindex semaphores on static probe points
5079 Some @code{SystemTap} probes have an associated semaphore variable;
5080 for instance, this happens automatically if you defined your probe
5081 using a DTrace-style @file{.d} file. If your probe has a semaphore,
5082 @value{GDBN} will automatically enable it when you specify a
5083 breakpoint using the @samp{-probe-stap} notation. But, if you put a
5084 breakpoint at a probe's location by some other method (e.g.,
5085 @code{break file:line}), then @value{GDBN} will not automatically set
5086 the semaphore. @code{DTrace} probes do not support semaphores.
5088 You can examine the available static static probes using @code{info
5089 probes}, with optional arguments:
5093 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5094 If given, @var{type} is either @code{stap} for listing
5095 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
5096 probes. If omitted all probes are listed regardless of their types.
5098 If given, @var{provider} is a regular expression used to match against provider
5099 names when selecting which probes to list. If omitted, probes by all
5100 probes from all providers are listed.
5102 If given, @var{name} is a regular expression to match against probe names
5103 when selecting which probes to list. If omitted, probe names are not
5104 considered when deciding whether to display them.
5106 If given, @var{objfile} is a regular expression used to select which
5107 object files (executable or shared libraries) to examine. If not
5108 given, all object files are considered.
5110 @item info probes all
5111 List the available static probes, from all types.
5114 @cindex enabling and disabling probes
5115 Some probe points can be enabled and/or disabled. The effect of
5116 enabling or disabling a probe depends on the type of probe being
5117 handled. Some @code{DTrace} probes can be enabled or
5118 disabled, but @code{SystemTap} probes cannot be disabled.
5120 You can enable (or disable) one or more probes using the following
5121 commands, with optional arguments:
5124 @kindex enable probes
5125 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5126 If given, @var{provider} is a regular expression used to match against
5127 provider names when selecting which probes to enable. If omitted,
5128 all probes from all providers are enabled.
5130 If given, @var{name} is a regular expression to match against probe
5131 names when selecting which probes to enable. If omitted, probe names
5132 are not considered when deciding whether to enable them.
5134 If given, @var{objfile} is a regular expression used to select which
5135 object files (executable or shared libraries) to examine. If not
5136 given, all object files are considered.
5138 @kindex disable probes
5139 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5140 See the @code{enable probes} command above for a description of the
5141 optional arguments accepted by this command.
5144 @vindex $_probe_arg@r{, convenience variable}
5145 A probe may specify up to twelve arguments. These are available at the
5146 point at which the probe is defined---that is, when the current PC is
5147 at the probe's location. The arguments are available using the
5148 convenience variables (@pxref{Convenience Vars})
5149 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
5150 probes each probe argument is an integer of the appropriate size;
5151 types are not preserved. In @code{DTrace} probes types are preserved
5152 provided that they are recognized as such by @value{GDBN}; otherwise
5153 the value of the probe argument will be a long integer. The
5154 convenience variable @code{$_probe_argc} holds the number of arguments
5155 at the current probe point.
5157 These variables are always available, but attempts to access them at
5158 any location other than a probe point will cause @value{GDBN} to give
5162 @c @ifclear BARETARGET
5163 @node Error in Breakpoints
5164 @subsection ``Cannot insert breakpoints''
5166 If you request too many active hardware-assisted breakpoints and
5167 watchpoints, you will see this error message:
5169 @c FIXME: the precise wording of this message may change; the relevant
5170 @c source change is not committed yet (Sep 3, 1999).
5172 Stopped; cannot insert breakpoints.
5173 You may have requested too many hardware breakpoints and watchpoints.
5177 This message is printed when you attempt to resume the program, since
5178 only then @value{GDBN} knows exactly how many hardware breakpoints and
5179 watchpoints it needs to insert.
5181 When this message is printed, you need to disable or remove some of the
5182 hardware-assisted breakpoints and watchpoints, and then continue.
5184 @node Breakpoint-related Warnings
5185 @subsection ``Breakpoint address adjusted...''
5186 @cindex breakpoint address adjusted
5188 Some processor architectures place constraints on the addresses at
5189 which breakpoints may be placed. For architectures thus constrained,
5190 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5191 with the constraints dictated by the architecture.
5193 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5194 a VLIW architecture in which a number of RISC-like instructions may be
5195 bundled together for parallel execution. The FR-V architecture
5196 constrains the location of a breakpoint instruction within such a
5197 bundle to the instruction with the lowest address. @value{GDBN}
5198 honors this constraint by adjusting a breakpoint's address to the
5199 first in the bundle.
5201 It is not uncommon for optimized code to have bundles which contain
5202 instructions from different source statements, thus it may happen that
5203 a breakpoint's address will be adjusted from one source statement to
5204 another. Since this adjustment may significantly alter @value{GDBN}'s
5205 breakpoint related behavior from what the user expects, a warning is
5206 printed when the breakpoint is first set and also when the breakpoint
5209 A warning like the one below is printed when setting a breakpoint
5210 that's been subject to address adjustment:
5213 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5216 Such warnings are printed both for user settable and @value{GDBN}'s
5217 internal breakpoints. If you see one of these warnings, you should
5218 verify that a breakpoint set at the adjusted address will have the
5219 desired affect. If not, the breakpoint in question may be removed and
5220 other breakpoints may be set which will have the desired behavior.
5221 E.g., it may be sufficient to place the breakpoint at a later
5222 instruction. A conditional breakpoint may also be useful in some
5223 cases to prevent the breakpoint from triggering too often.
5225 @value{GDBN} will also issue a warning when stopping at one of these
5226 adjusted breakpoints:
5229 warning: Breakpoint 1 address previously adjusted from 0x00010414
5233 When this warning is encountered, it may be too late to take remedial
5234 action except in cases where the breakpoint is hit earlier or more
5235 frequently than expected.
5237 @node Continuing and Stepping
5238 @section Continuing and Stepping
5242 @cindex resuming execution
5243 @dfn{Continuing} means resuming program execution until your program
5244 completes normally. In contrast, @dfn{stepping} means executing just
5245 one more ``step'' of your program, where ``step'' may mean either one
5246 line of source code, or one machine instruction (depending on what
5247 particular command you use). Either when continuing or when stepping,
5248 your program may stop even sooner, due to a breakpoint or a signal. (If
5249 it stops due to a signal, you may want to use @code{handle}, or use
5250 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5251 or you may step into the signal's handler (@pxref{stepping and signal
5256 @kindex c @r{(@code{continue})}
5257 @kindex fg @r{(resume foreground execution)}
5258 @item continue @r{[}@var{ignore-count}@r{]}
5259 @itemx c @r{[}@var{ignore-count}@r{]}
5260 @itemx fg @r{[}@var{ignore-count}@r{]}
5261 Resume program execution, at the address where your program last stopped;
5262 any breakpoints set at that address are bypassed. The optional argument
5263 @var{ignore-count} allows you to specify a further number of times to
5264 ignore a breakpoint at this location; its effect is like that of
5265 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5267 The argument @var{ignore-count} is meaningful only when your program
5268 stopped due to a breakpoint. At other times, the argument to
5269 @code{continue} is ignored.
5271 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5272 debugged program is deemed to be the foreground program) are provided
5273 purely for convenience, and have exactly the same behavior as
5277 To resume execution at a different place, you can use @code{return}
5278 (@pxref{Returning, ,Returning from a Function}) to go back to the
5279 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5280 Different Address}) to go to an arbitrary location in your program.
5282 A typical technique for using stepping is to set a breakpoint
5283 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5284 beginning of the function or the section of your program where a problem
5285 is believed to lie, run your program until it stops at that breakpoint,
5286 and then step through the suspect area, examining the variables that are
5287 interesting, until you see the problem happen.
5291 @kindex s @r{(@code{step})}
5293 Continue running your program until control reaches a different source
5294 line, then stop it and return control to @value{GDBN}. This command is
5295 abbreviated @code{s}.
5298 @c "without debugging information" is imprecise; actually "without line
5299 @c numbers in the debugging information". (gcc -g1 has debugging info but
5300 @c not line numbers). But it seems complex to try to make that
5301 @c distinction here.
5302 @emph{Warning:} If you use the @code{step} command while control is
5303 within a function that was compiled without debugging information,
5304 execution proceeds until control reaches a function that does have
5305 debugging information. Likewise, it will not step into a function which
5306 is compiled without debugging information. To step through functions
5307 without debugging information, use the @code{stepi} command, described
5311 The @code{step} command only stops at the first instruction of a source
5312 line. This prevents the multiple stops that could otherwise occur in
5313 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5314 to stop if a function that has debugging information is called within
5315 the line. In other words, @code{step} @emph{steps inside} any functions
5316 called within the line.
5318 Also, the @code{step} command only enters a function if there is line
5319 number information for the function. Otherwise it acts like the
5320 @code{next} command. This avoids problems when using @code{cc -gl}
5321 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5322 was any debugging information about the routine.
5324 @item step @var{count}
5325 Continue running as in @code{step}, but do so @var{count} times. If a
5326 breakpoint is reached, or a signal not related to stepping occurs before
5327 @var{count} steps, stepping stops right away.
5330 @kindex n @r{(@code{next})}
5331 @item next @r{[}@var{count}@r{]}
5332 Continue to the next source line in the current (innermost) stack frame.
5333 This is similar to @code{step}, but function calls that appear within
5334 the line of code are executed without stopping. Execution stops when
5335 control reaches a different line of code at the original stack level
5336 that was executing when you gave the @code{next} command. This command
5337 is abbreviated @code{n}.
5339 An argument @var{count} is a repeat count, as for @code{step}.
5342 @c FIX ME!! Do we delete this, or is there a way it fits in with
5343 @c the following paragraph? --- Vctoria
5345 @c @code{next} within a function that lacks debugging information acts like
5346 @c @code{step}, but any function calls appearing within the code of the
5347 @c function are executed without stopping.
5349 The @code{next} command only stops at the first instruction of a
5350 source line. This prevents multiple stops that could otherwise occur in
5351 @code{switch} statements, @code{for} loops, etc.
5353 @kindex set step-mode
5355 @cindex functions without line info, and stepping
5356 @cindex stepping into functions with no line info
5357 @itemx set step-mode on
5358 The @code{set step-mode on} command causes the @code{step} command to
5359 stop at the first instruction of a function which contains no debug line
5360 information rather than stepping over it.
5362 This is useful in cases where you may be interested in inspecting the
5363 machine instructions of a function which has no symbolic info and do not
5364 want @value{GDBN} to automatically skip over this function.
5366 @item set step-mode off
5367 Causes the @code{step} command to step over any functions which contains no
5368 debug information. This is the default.
5370 @item show step-mode
5371 Show whether @value{GDBN} will stop in or step over functions without
5372 source line debug information.
5375 @kindex fin @r{(@code{finish})}
5377 Continue running until just after function in the selected stack frame
5378 returns. Print the returned value (if any). This command can be
5379 abbreviated as @code{fin}.
5381 Contrast this with the @code{return} command (@pxref{Returning,
5382 ,Returning from a Function}).
5385 @kindex u @r{(@code{until})}
5386 @cindex run until specified location
5389 Continue running until a source line past the current line, in the
5390 current stack frame, is reached. This command is used to avoid single
5391 stepping through a loop more than once. It is like the @code{next}
5392 command, except that when @code{until} encounters a jump, it
5393 automatically continues execution until the program counter is greater
5394 than the address of the jump.
5396 This means that when you reach the end of a loop after single stepping
5397 though it, @code{until} makes your program continue execution until it
5398 exits the loop. In contrast, a @code{next} command at the end of a loop
5399 simply steps back to the beginning of the loop, which forces you to step
5400 through the next iteration.
5402 @code{until} always stops your program if it attempts to exit the current
5405 @code{until} may produce somewhat counterintuitive results if the order
5406 of machine code does not match the order of the source lines. For
5407 example, in the following excerpt from a debugging session, the @code{f}
5408 (@code{frame}) command shows that execution is stopped at line
5409 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5413 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5415 (@value{GDBP}) until
5416 195 for ( ; argc > 0; NEXTARG) @{
5419 This happened because, for execution efficiency, the compiler had
5420 generated code for the loop closure test at the end, rather than the
5421 start, of the loop---even though the test in a C @code{for}-loop is
5422 written before the body of the loop. The @code{until} command appeared
5423 to step back to the beginning of the loop when it advanced to this
5424 expression; however, it has not really gone to an earlier
5425 statement---not in terms of the actual machine code.
5427 @code{until} with no argument works by means of single
5428 instruction stepping, and hence is slower than @code{until} with an
5431 @item until @var{location}
5432 @itemx u @var{location}
5433 Continue running your program until either the specified @var{location} is
5434 reached, or the current stack frame returns. The location is any of
5435 the forms described in @ref{Specify Location}.
5436 This form of the command uses temporary breakpoints, and
5437 hence is quicker than @code{until} without an argument. The specified
5438 location is actually reached only if it is in the current frame. This
5439 implies that @code{until} can be used to skip over recursive function
5440 invocations. For instance in the code below, if the current location is
5441 line @code{96}, issuing @code{until 99} will execute the program up to
5442 line @code{99} in the same invocation of factorial, i.e., after the inner
5443 invocations have returned.
5446 94 int factorial (int value)
5448 96 if (value > 1) @{
5449 97 value *= factorial (value - 1);
5456 @kindex advance @var{location}
5457 @item advance @var{location}
5458 Continue running the program up to the given @var{location}. An argument is
5459 required, which should be of one of the forms described in
5460 @ref{Specify Location}.
5461 Execution will also stop upon exit from the current stack
5462 frame. This command is similar to @code{until}, but @code{advance} will
5463 not skip over recursive function calls, and the target location doesn't
5464 have to be in the same frame as the current one.
5468 @kindex si @r{(@code{stepi})}
5470 @itemx stepi @var{arg}
5472 Execute one machine instruction, then stop and return to the debugger.
5474 It is often useful to do @samp{display/i $pc} when stepping by machine
5475 instructions. This makes @value{GDBN} automatically display the next
5476 instruction to be executed, each time your program stops. @xref{Auto
5477 Display,, Automatic Display}.
5479 An argument is a repeat count, as in @code{step}.
5483 @kindex ni @r{(@code{nexti})}
5485 @itemx nexti @var{arg}
5487 Execute one machine instruction, but if it is a function call,
5488 proceed until the function returns.
5490 An argument is a repeat count, as in @code{next}.
5494 @anchor{range stepping}
5495 @cindex range stepping
5496 @cindex target-assisted range stepping
5497 By default, and if available, @value{GDBN} makes use of
5498 target-assisted @dfn{range stepping}. In other words, whenever you
5499 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
5500 tells the target to step the corresponding range of instruction
5501 addresses instead of issuing multiple single-steps. This speeds up
5502 line stepping, particularly for remote targets. Ideally, there should
5503 be no reason you would want to turn range stepping off. However, it's
5504 possible that a bug in the debug info, a bug in the remote stub (for
5505 remote targets), or even a bug in @value{GDBN} could make line
5506 stepping behave incorrectly when target-assisted range stepping is
5507 enabled. You can use the following command to turn off range stepping
5511 @kindex set range-stepping
5512 @kindex show range-stepping
5513 @item set range-stepping
5514 @itemx show range-stepping
5515 Control whether range stepping is enabled.
5517 If @code{on}, and the target supports it, @value{GDBN} tells the
5518 target to step a range of addresses itself, instead of issuing
5519 multiple single-steps. If @code{off}, @value{GDBN} always issues
5520 single-steps, even if range stepping is supported by the target. The
5521 default is @code{on}.
5525 @node Skipping Over Functions and Files
5526 @section Skipping Over Functions and Files
5527 @cindex skipping over functions and files
5529 The program you are debugging may contain some functions which are
5530 uninteresting to debug. The @code{skip} comand lets you tell @value{GDBN} to
5531 skip a function or all functions in a file when stepping.
5533 For example, consider the following C function:
5544 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5545 are not interested in stepping through @code{boring}. If you run @code{step}
5546 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5547 step over both @code{foo} and @code{boring}!
5549 One solution is to @code{step} into @code{boring} and use the @code{finish}
5550 command to immediately exit it. But this can become tedious if @code{boring}
5551 is called from many places.
5553 A more flexible solution is to execute @kbd{skip boring}. This instructs
5554 @value{GDBN} never to step into @code{boring}. Now when you execute
5555 @code{step} at line 103, you'll step over @code{boring} and directly into
5558 You can also instruct @value{GDBN} to skip all functions in a file, with, for
5559 example, @code{skip file boring.c}.
5562 @kindex skip function
5563 @item skip @r{[}@var{linespec}@r{]}
5564 @itemx skip function @r{[}@var{linespec}@r{]}
5565 After running this command, the function named by @var{linespec} or the
5566 function containing the line named by @var{linespec} will be skipped over when
5567 stepping. @xref{Specify Location}.
5569 If you do not specify @var{linespec}, the function you're currently debugging
5572 (If you have a function called @code{file} that you want to skip, use
5573 @kbd{skip function file}.)
5576 @item skip file @r{[}@var{filename}@r{]}
5577 After running this command, any function whose source lives in @var{filename}
5578 will be skipped over when stepping.
5580 If you do not specify @var{filename}, functions whose source lives in the file
5581 you're currently debugging will be skipped.
5584 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
5585 These are the commands for managing your list of skips:
5589 @item info skip @r{[}@var{range}@r{]}
5590 Print details about the specified skip(s). If @var{range} is not specified,
5591 print a table with details about all functions and files marked for skipping.
5592 @code{info skip} prints the following information about each skip:
5596 A number identifying this skip.
5598 The type of this skip, either @samp{function} or @samp{file}.
5599 @item Enabled or Disabled
5600 Enabled skips are marked with @samp{y}. Disabled skips are marked with @samp{n}.
5602 For function skips, this column indicates the address in memory of the function
5603 being skipped. If you've set a function skip on a function which has not yet
5604 been loaded, this field will contain @samp{<PENDING>}. Once a shared library
5605 which has the function is loaded, @code{info skip} will show the function's
5608 For file skips, this field contains the filename being skipped. For functions
5609 skips, this field contains the function name and its line number in the file
5610 where it is defined.
5614 @item skip delete @r{[}@var{range}@r{]}
5615 Delete the specified skip(s). If @var{range} is not specified, delete all
5619 @item skip enable @r{[}@var{range}@r{]}
5620 Enable the specified skip(s). If @var{range} is not specified, enable all
5623 @kindex skip disable
5624 @item skip disable @r{[}@var{range}@r{]}
5625 Disable the specified skip(s). If @var{range} is not specified, disable all
5634 A signal is an asynchronous event that can happen in a program. The
5635 operating system defines the possible kinds of signals, and gives each
5636 kind a name and a number. For example, in Unix @code{SIGINT} is the
5637 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
5638 @code{SIGSEGV} is the signal a program gets from referencing a place in
5639 memory far away from all the areas in use; @code{SIGALRM} occurs when
5640 the alarm clock timer goes off (which happens only if your program has
5641 requested an alarm).
5643 @cindex fatal signals
5644 Some signals, including @code{SIGALRM}, are a normal part of the
5645 functioning of your program. Others, such as @code{SIGSEGV}, indicate
5646 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
5647 program has not specified in advance some other way to handle the signal.
5648 @code{SIGINT} does not indicate an error in your program, but it is normally
5649 fatal so it can carry out the purpose of the interrupt: to kill the program.
5651 @value{GDBN} has the ability to detect any occurrence of a signal in your
5652 program. You can tell @value{GDBN} in advance what to do for each kind of
5655 @cindex handling signals
5656 Normally, @value{GDBN} is set up to let the non-erroneous signals like
5657 @code{SIGALRM} be silently passed to your program
5658 (so as not to interfere with their role in the program's functioning)
5659 but to stop your program immediately whenever an error signal happens.
5660 You can change these settings with the @code{handle} command.
5663 @kindex info signals
5667 Print a table of all the kinds of signals and how @value{GDBN} has been told to
5668 handle each one. You can use this to see the signal numbers of all
5669 the defined types of signals.
5671 @item info signals @var{sig}
5672 Similar, but print information only about the specified signal number.
5674 @code{info handle} is an alias for @code{info signals}.
5676 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5677 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
5678 for details about this command.
5681 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
5682 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
5683 can be the number of a signal or its name (with or without the
5684 @samp{SIG} at the beginning); a list of signal numbers of the form
5685 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
5686 known signals. Optional arguments @var{keywords}, described below,
5687 say what change to make.
5691 The keywords allowed by the @code{handle} command can be abbreviated.
5692 Their full names are:
5696 @value{GDBN} should not stop your program when this signal happens. It may
5697 still print a message telling you that the signal has come in.
5700 @value{GDBN} should stop your program when this signal happens. This implies
5701 the @code{print} keyword as well.
5704 @value{GDBN} should print a message when this signal happens.
5707 @value{GDBN} should not mention the occurrence of the signal at all. This
5708 implies the @code{nostop} keyword as well.
5712 @value{GDBN} should allow your program to see this signal; your program
5713 can handle the signal, or else it may terminate if the signal is fatal
5714 and not handled. @code{pass} and @code{noignore} are synonyms.
5718 @value{GDBN} should not allow your program to see this signal.
5719 @code{nopass} and @code{ignore} are synonyms.
5723 When a signal stops your program, the signal is not visible to the
5725 continue. Your program sees the signal then, if @code{pass} is in
5726 effect for the signal in question @emph{at that time}. In other words,
5727 after @value{GDBN} reports a signal, you can use the @code{handle}
5728 command with @code{pass} or @code{nopass} to control whether your
5729 program sees that signal when you continue.
5731 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
5732 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
5733 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
5736 You can also use the @code{signal} command to prevent your program from
5737 seeing a signal, or cause it to see a signal it normally would not see,
5738 or to give it any signal at any time. For example, if your program stopped
5739 due to some sort of memory reference error, you might store correct
5740 values into the erroneous variables and continue, hoping to see more
5741 execution; but your program would probably terminate immediately as
5742 a result of the fatal signal once it saw the signal. To prevent this,
5743 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
5746 @cindex stepping and signal handlers
5747 @anchor{stepping and signal handlers}
5749 @value{GDBN} optimizes for stepping the mainline code. If a signal
5750 that has @code{handle nostop} and @code{handle pass} set arrives while
5751 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
5752 in progress, @value{GDBN} lets the signal handler run and then resumes
5753 stepping the mainline code once the signal handler returns. In other
5754 words, @value{GDBN} steps over the signal handler. This prevents
5755 signals that you've specified as not interesting (with @code{handle
5756 nostop}) from changing the focus of debugging unexpectedly. Note that
5757 the signal handler itself may still hit a breakpoint, stop for another
5758 signal that has @code{handle stop} in effect, or for any other event
5759 that normally results in stopping the stepping command sooner. Also
5760 note that @value{GDBN} still informs you that the program received a
5761 signal if @code{handle print} is set.
5763 @anchor{stepping into signal handlers}
5765 If you set @code{handle pass} for a signal, and your program sets up a
5766 handler for it, then issuing a stepping command, such as @code{step}
5767 or @code{stepi}, when your program is stopped due to the signal will
5768 step @emph{into} the signal handler (if the target supports that).
5770 Likewise, if you use the @code{queue-signal} command to queue a signal
5771 to be delivered to the current thread when execution of the thread
5772 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
5773 stepping command will step into the signal handler.
5775 Here's an example, using @code{stepi} to step to the first instruction
5776 of @code{SIGUSR1}'s handler:
5779 (@value{GDBP}) handle SIGUSR1
5780 Signal Stop Print Pass to program Description
5781 SIGUSR1 Yes Yes Yes User defined signal 1
5785 Program received signal SIGUSR1, User defined signal 1.
5786 main () sigusr1.c:28
5789 sigusr1_handler () at sigusr1.c:9
5793 The same, but using @code{queue-signal} instead of waiting for the
5794 program to receive the signal first:
5799 (@value{GDBP}) queue-signal SIGUSR1
5801 sigusr1_handler () at sigusr1.c:9
5806 @cindex extra signal information
5807 @anchor{extra signal information}
5809 On some targets, @value{GDBN} can inspect extra signal information
5810 associated with the intercepted signal, before it is actually
5811 delivered to the program being debugged. This information is exported
5812 by the convenience variable @code{$_siginfo}, and consists of data
5813 that is passed by the kernel to the signal handler at the time of the
5814 receipt of a signal. The data type of the information itself is
5815 target dependent. You can see the data type using the @code{ptype
5816 $_siginfo} command. On Unix systems, it typically corresponds to the
5817 standard @code{siginfo_t} type, as defined in the @file{signal.h}
5820 Here's an example, on a @sc{gnu}/Linux system, printing the stray
5821 referenced address that raised a segmentation fault.
5825 (@value{GDBP}) continue
5826 Program received signal SIGSEGV, Segmentation fault.
5827 0x0000000000400766 in main ()
5829 (@value{GDBP}) ptype $_siginfo
5836 struct @{...@} _kill;
5837 struct @{...@} _timer;
5839 struct @{...@} _sigchld;
5840 struct @{...@} _sigfault;
5841 struct @{...@} _sigpoll;
5844 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
5848 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
5849 $1 = (void *) 0x7ffff7ff7000
5853 Depending on target support, @code{$_siginfo} may also be writable.
5856 @section Stopping and Starting Multi-thread Programs
5858 @cindex stopped threads
5859 @cindex threads, stopped
5861 @cindex continuing threads
5862 @cindex threads, continuing
5864 @value{GDBN} supports debugging programs with multiple threads
5865 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
5866 are two modes of controlling execution of your program within the
5867 debugger. In the default mode, referred to as @dfn{all-stop mode},
5868 when any thread in your program stops (for example, at a breakpoint
5869 or while being stepped), all other threads in the program are also stopped by
5870 @value{GDBN}. On some targets, @value{GDBN} also supports
5871 @dfn{non-stop mode}, in which other threads can continue to run freely while
5872 you examine the stopped thread in the debugger.
5875 * All-Stop Mode:: All threads stop when GDB takes control
5876 * Non-Stop Mode:: Other threads continue to execute
5877 * Background Execution:: Running your program asynchronously
5878 * Thread-Specific Breakpoints:: Controlling breakpoints
5879 * Interrupted System Calls:: GDB may interfere with system calls
5880 * Observer Mode:: GDB does not alter program behavior
5884 @subsection All-Stop Mode
5886 @cindex all-stop mode
5888 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
5889 @emph{all} threads of execution stop, not just the current thread. This
5890 allows you to examine the overall state of the program, including
5891 switching between threads, without worrying that things may change
5894 Conversely, whenever you restart the program, @emph{all} threads start
5895 executing. @emph{This is true even when single-stepping} with commands
5896 like @code{step} or @code{next}.
5898 In particular, @value{GDBN} cannot single-step all threads in lockstep.
5899 Since thread scheduling is up to your debugging target's operating
5900 system (not controlled by @value{GDBN}), other threads may
5901 execute more than one statement while the current thread completes a
5902 single step. Moreover, in general other threads stop in the middle of a
5903 statement, rather than at a clean statement boundary, when the program
5906 You might even find your program stopped in another thread after
5907 continuing or even single-stepping. This happens whenever some other
5908 thread runs into a breakpoint, a signal, or an exception before the
5909 first thread completes whatever you requested.
5911 @cindex automatic thread selection
5912 @cindex switching threads automatically
5913 @cindex threads, automatic switching
5914 Whenever @value{GDBN} stops your program, due to a breakpoint or a
5915 signal, it automatically selects the thread where that breakpoint or
5916 signal happened. @value{GDBN} alerts you to the context switch with a
5917 message such as @samp{[Switching to Thread @var{n}]} to identify the
5920 On some OSes, you can modify @value{GDBN}'s default behavior by
5921 locking the OS scheduler to allow only a single thread to run.
5924 @item set scheduler-locking @var{mode}
5925 @cindex scheduler locking mode
5926 @cindex lock scheduler
5927 Set the scheduler locking mode. It applies to normal execution,
5928 record mode, and replay mode. If it is @code{off}, then there is no
5929 locking and any thread may run at any time. If @code{on}, then only
5930 the current thread may run when the inferior is resumed. The
5931 @code{step} mode optimizes for single-stepping; it prevents other
5932 threads from preempting the current thread while you are stepping, so
5933 that the focus of debugging does not change unexpectedly. Other
5934 threads never get a chance to run when you step, and they are
5935 completely free to run when you use commands like @samp{continue},
5936 @samp{until}, or @samp{finish}. However, unless another thread hits a
5937 breakpoint during its timeslice, @value{GDBN} does not change the
5938 current thread away from the thread that you are debugging. The
5939 @code{replay} mode behaves like @code{off} in record mode and like
5940 @code{on} in replay mode.
5942 @item show scheduler-locking
5943 Display the current scheduler locking mode.
5946 @cindex resume threads of multiple processes simultaneously
5947 By default, when you issue one of the execution commands such as
5948 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
5949 threads of the current inferior to run. For example, if @value{GDBN}
5950 is attached to two inferiors, each with two threads, the
5951 @code{continue} command resumes only the two threads of the current
5952 inferior. This is useful, for example, when you debug a program that
5953 forks and you want to hold the parent stopped (so that, for instance,
5954 it doesn't run to exit), while you debug the child. In other
5955 situations, you may not be interested in inspecting the current state
5956 of any of the processes @value{GDBN} is attached to, and you may want
5957 to resume them all until some breakpoint is hit. In the latter case,
5958 you can instruct @value{GDBN} to allow all threads of all the
5959 inferiors to run with the @w{@code{set schedule-multiple}} command.
5962 @kindex set schedule-multiple
5963 @item set schedule-multiple
5964 Set the mode for allowing threads of multiple processes to be resumed
5965 when an execution command is issued. When @code{on}, all threads of
5966 all processes are allowed to run. When @code{off}, only the threads
5967 of the current process are resumed. The default is @code{off}. The
5968 @code{scheduler-locking} mode takes precedence when set to @code{on},
5969 or while you are stepping and set to @code{step}.
5971 @item show schedule-multiple
5972 Display the current mode for resuming the execution of threads of
5977 @subsection Non-Stop Mode
5979 @cindex non-stop mode
5981 @c This section is really only a place-holder, and needs to be expanded
5982 @c with more details.
5984 For some multi-threaded targets, @value{GDBN} supports an optional
5985 mode of operation in which you can examine stopped program threads in
5986 the debugger while other threads continue to execute freely. This
5987 minimizes intrusion when debugging live systems, such as programs
5988 where some threads have real-time constraints or must continue to
5989 respond to external events. This is referred to as @dfn{non-stop} mode.
5991 In non-stop mode, when a thread stops to report a debugging event,
5992 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5993 threads as well, in contrast to the all-stop mode behavior. Additionally,
5994 execution commands such as @code{continue} and @code{step} apply by default
5995 only to the current thread in non-stop mode, rather than all threads as
5996 in all-stop mode. This allows you to control threads explicitly in
5997 ways that are not possible in all-stop mode --- for example, stepping
5998 one thread while allowing others to run freely, stepping
5999 one thread while holding all others stopped, or stepping several threads
6000 independently and simultaneously.
6002 To enter non-stop mode, use this sequence of commands before you run
6003 or attach to your program:
6006 # If using the CLI, pagination breaks non-stop.
6009 # Finally, turn it on!
6013 You can use these commands to manipulate the non-stop mode setting:
6016 @kindex set non-stop
6017 @item set non-stop on
6018 Enable selection of non-stop mode.
6019 @item set non-stop off
6020 Disable selection of non-stop mode.
6021 @kindex show non-stop
6023 Show the current non-stop enablement setting.
6026 Note these commands only reflect whether non-stop mode is enabled,
6027 not whether the currently-executing program is being run in non-stop mode.
6028 In particular, the @code{set non-stop} preference is only consulted when
6029 @value{GDBN} starts or connects to the target program, and it is generally
6030 not possible to switch modes once debugging has started. Furthermore,
6031 since not all targets support non-stop mode, even when you have enabled
6032 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
6035 In non-stop mode, all execution commands apply only to the current thread
6036 by default. That is, @code{continue} only continues one thread.
6037 To continue all threads, issue @code{continue -a} or @code{c -a}.
6039 You can use @value{GDBN}'s background execution commands
6040 (@pxref{Background Execution}) to run some threads in the background
6041 while you continue to examine or step others from @value{GDBN}.
6042 The MI execution commands (@pxref{GDB/MI Program Execution}) are
6043 always executed asynchronously in non-stop mode.
6045 Suspending execution is done with the @code{interrupt} command when
6046 running in the background, or @kbd{Ctrl-c} during foreground execution.
6047 In all-stop mode, this stops the whole process;
6048 but in non-stop mode the interrupt applies only to the current thread.
6049 To stop the whole program, use @code{interrupt -a}.
6051 Other execution commands do not currently support the @code{-a} option.
6053 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
6054 that thread current, as it does in all-stop mode. This is because the
6055 thread stop notifications are asynchronous with respect to @value{GDBN}'s
6056 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
6057 changed to a different thread just as you entered a command to operate on the
6058 previously current thread.
6060 @node Background Execution
6061 @subsection Background Execution
6063 @cindex foreground execution
6064 @cindex background execution
6065 @cindex asynchronous execution
6066 @cindex execution, foreground, background and asynchronous
6068 @value{GDBN}'s execution commands have two variants: the normal
6069 foreground (synchronous) behavior, and a background
6070 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
6071 the program to report that some thread has stopped before prompting for
6072 another command. In background execution, @value{GDBN} immediately gives
6073 a command prompt so that you can issue other commands while your program runs.
6075 If the target doesn't support async mode, @value{GDBN} issues an error
6076 message if you attempt to use the background execution commands.
6078 To specify background execution, add a @code{&} to the command. For example,
6079 the background form of the @code{continue} command is @code{continue&}, or
6080 just @code{c&}. The execution commands that accept background execution
6086 @xref{Starting, , Starting your Program}.
6090 @xref{Attach, , Debugging an Already-running Process}.
6094 @xref{Continuing and Stepping, step}.
6098 @xref{Continuing and Stepping, stepi}.
6102 @xref{Continuing and Stepping, next}.
6106 @xref{Continuing and Stepping, nexti}.
6110 @xref{Continuing and Stepping, continue}.
6114 @xref{Continuing and Stepping, finish}.
6118 @xref{Continuing and Stepping, until}.
6122 Background execution is especially useful in conjunction with non-stop
6123 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
6124 However, you can also use these commands in the normal all-stop mode with
6125 the restriction that you cannot issue another execution command until the
6126 previous one finishes. Examples of commands that are valid in all-stop
6127 mode while the program is running include @code{help} and @code{info break}.
6129 You can interrupt your program while it is running in the background by
6130 using the @code{interrupt} command.
6137 Suspend execution of the running program. In all-stop mode,
6138 @code{interrupt} stops the whole process, but in non-stop mode, it stops
6139 only the current thread. To stop the whole program in non-stop mode,
6140 use @code{interrupt -a}.
6143 @node Thread-Specific Breakpoints
6144 @subsection Thread-Specific Breakpoints
6146 When your program has multiple threads (@pxref{Threads,, Debugging
6147 Programs with Multiple Threads}), you can choose whether to set
6148 breakpoints on all threads, or on a particular thread.
6151 @cindex breakpoints and threads
6152 @cindex thread breakpoints
6153 @kindex break @dots{} thread @var{thread-id}
6154 @item break @var{location} thread @var{thread-id}
6155 @itemx break @var{location} thread @var{thread-id} if @dots{}
6156 @var{location} specifies source lines; there are several ways of
6157 writing them (@pxref{Specify Location}), but the effect is always to
6158 specify some source line.
6160 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
6161 to specify that you only want @value{GDBN} to stop the program when a
6162 particular thread reaches this breakpoint. The @var{thread-id} specifier
6163 is one of the thread identifiers assigned by @value{GDBN}, shown
6164 in the first column of the @samp{info threads} display.
6166 If you do not specify @samp{thread @var{thread-id}} when you set a
6167 breakpoint, the breakpoint applies to @emph{all} threads of your
6170 You can use the @code{thread} qualifier on conditional breakpoints as
6171 well; in this case, place @samp{thread @var{thread-id}} before or
6172 after the breakpoint condition, like this:
6175 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6180 Thread-specific breakpoints are automatically deleted when
6181 @value{GDBN} detects the corresponding thread is no longer in the
6182 thread list. For example:
6186 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6189 There are several ways for a thread to disappear, such as a regular
6190 thread exit, but also when you detach from the process with the
6191 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6192 Process}), or if @value{GDBN} loses the remote connection
6193 (@pxref{Remote Debugging}), etc. Note that with some targets,
6194 @value{GDBN} is only able to detect a thread has exited when the user
6195 explictly asks for the thread list with the @code{info threads}
6198 @node Interrupted System Calls
6199 @subsection Interrupted System Calls
6201 @cindex thread breakpoints and system calls
6202 @cindex system calls and thread breakpoints
6203 @cindex premature return from system calls
6204 There is an unfortunate side effect when using @value{GDBN} to debug
6205 multi-threaded programs. If one thread stops for a
6206 breakpoint, or for some other reason, and another thread is blocked in a
6207 system call, then the system call may return prematurely. This is a
6208 consequence of the interaction between multiple threads and the signals
6209 that @value{GDBN} uses to implement breakpoints and other events that
6212 To handle this problem, your program should check the return value of
6213 each system call and react appropriately. This is good programming
6216 For example, do not write code like this:
6222 The call to @code{sleep} will return early if a different thread stops
6223 at a breakpoint or for some other reason.
6225 Instead, write this:
6230 unslept = sleep (unslept);
6233 A system call is allowed to return early, so the system is still
6234 conforming to its specification. But @value{GDBN} does cause your
6235 multi-threaded program to behave differently than it would without
6238 Also, @value{GDBN} uses internal breakpoints in the thread library to
6239 monitor certain events such as thread creation and thread destruction.
6240 When such an event happens, a system call in another thread may return
6241 prematurely, even though your program does not appear to stop.
6244 @subsection Observer Mode
6246 If you want to build on non-stop mode and observe program behavior
6247 without any chance of disruption by @value{GDBN}, you can set
6248 variables to disable all of the debugger's attempts to modify state,
6249 whether by writing memory, inserting breakpoints, etc. These operate
6250 at a low level, intercepting operations from all commands.
6252 When all of these are set to @code{off}, then @value{GDBN} is said to
6253 be @dfn{observer mode}. As a convenience, the variable
6254 @code{observer} can be set to disable these, plus enable non-stop
6257 Note that @value{GDBN} will not prevent you from making nonsensical
6258 combinations of these settings. For instance, if you have enabled
6259 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6260 then breakpoints that work by writing trap instructions into the code
6261 stream will still not be able to be placed.
6266 @item set observer on
6267 @itemx set observer off
6268 When set to @code{on}, this disables all the permission variables
6269 below (except for @code{insert-fast-tracepoints}), plus enables
6270 non-stop debugging. Setting this to @code{off} switches back to
6271 normal debugging, though remaining in non-stop mode.
6274 Show whether observer mode is on or off.
6276 @kindex may-write-registers
6277 @item set may-write-registers on
6278 @itemx set may-write-registers off
6279 This controls whether @value{GDBN} will attempt to alter the values of
6280 registers, such as with assignment expressions in @code{print}, or the
6281 @code{jump} command. It defaults to @code{on}.
6283 @item show may-write-registers
6284 Show the current permission to write registers.
6286 @kindex may-write-memory
6287 @item set may-write-memory on
6288 @itemx set may-write-memory off
6289 This controls whether @value{GDBN} will attempt to alter the contents
6290 of memory, such as with assignment expressions in @code{print}. It
6291 defaults to @code{on}.
6293 @item show may-write-memory
6294 Show the current permission to write memory.
6296 @kindex may-insert-breakpoints
6297 @item set may-insert-breakpoints on
6298 @itemx set may-insert-breakpoints off
6299 This controls whether @value{GDBN} will attempt to insert breakpoints.
6300 This affects all breakpoints, including internal breakpoints defined
6301 by @value{GDBN}. It defaults to @code{on}.
6303 @item show may-insert-breakpoints
6304 Show the current permission to insert breakpoints.
6306 @kindex may-insert-tracepoints
6307 @item set may-insert-tracepoints on
6308 @itemx set may-insert-tracepoints off
6309 This controls whether @value{GDBN} will attempt to insert (regular)
6310 tracepoints at the beginning of a tracing experiment. It affects only
6311 non-fast tracepoints, fast tracepoints being under the control of
6312 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6314 @item show may-insert-tracepoints
6315 Show the current permission to insert tracepoints.
6317 @kindex may-insert-fast-tracepoints
6318 @item set may-insert-fast-tracepoints on
6319 @itemx set may-insert-fast-tracepoints off
6320 This controls whether @value{GDBN} will attempt to insert fast
6321 tracepoints at the beginning of a tracing experiment. It affects only
6322 fast tracepoints, regular (non-fast) tracepoints being under the
6323 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6325 @item show may-insert-fast-tracepoints
6326 Show the current permission to insert fast tracepoints.
6328 @kindex may-interrupt
6329 @item set may-interrupt on
6330 @itemx set may-interrupt off
6331 This controls whether @value{GDBN} will attempt to interrupt or stop
6332 program execution. When this variable is @code{off}, the
6333 @code{interrupt} command will have no effect, nor will
6334 @kbd{Ctrl-c}. It defaults to @code{on}.
6336 @item show may-interrupt
6337 Show the current permission to interrupt or stop the program.
6341 @node Reverse Execution
6342 @chapter Running programs backward
6343 @cindex reverse execution
6344 @cindex running programs backward
6346 When you are debugging a program, it is not unusual to realize that
6347 you have gone too far, and some event of interest has already happened.
6348 If the target environment supports it, @value{GDBN} can allow you to
6349 ``rewind'' the program by running it backward.
6351 A target environment that supports reverse execution should be able
6352 to ``undo'' the changes in machine state that have taken place as the
6353 program was executing normally. Variables, registers etc.@: should
6354 revert to their previous values. Obviously this requires a great
6355 deal of sophistication on the part of the target environment; not
6356 all target environments can support reverse execution.
6358 When a program is executed in reverse, the instructions that
6359 have most recently been executed are ``un-executed'', in reverse
6360 order. The program counter runs backward, following the previous
6361 thread of execution in reverse. As each instruction is ``un-executed'',
6362 the values of memory and/or registers that were changed by that
6363 instruction are reverted to their previous states. After executing
6364 a piece of source code in reverse, all side effects of that code
6365 should be ``undone'', and all variables should be returned to their
6366 prior values@footnote{
6367 Note that some side effects are easier to undo than others. For instance,
6368 memory and registers are relatively easy, but device I/O is hard. Some
6369 targets may be able undo things like device I/O, and some may not.
6371 The contract between @value{GDBN} and the reverse executing target
6372 requires only that the target do something reasonable when
6373 @value{GDBN} tells it to execute backwards, and then report the
6374 results back to @value{GDBN}. Whatever the target reports back to
6375 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6376 assumes that the memory and registers that the target reports are in a
6377 consistant state, but @value{GDBN} accepts whatever it is given.
6380 If you are debugging in a target environment that supports
6381 reverse execution, @value{GDBN} provides the following commands.
6384 @kindex reverse-continue
6385 @kindex rc @r{(@code{reverse-continue})}
6386 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6387 @itemx rc @r{[}@var{ignore-count}@r{]}
6388 Beginning at the point where your program last stopped, start executing
6389 in reverse. Reverse execution will stop for breakpoints and synchronous
6390 exceptions (signals), just like normal execution. Behavior of
6391 asynchronous signals depends on the target environment.
6393 @kindex reverse-step
6394 @kindex rs @r{(@code{step})}
6395 @item reverse-step @r{[}@var{count}@r{]}
6396 Run the program backward until control reaches the start of a
6397 different source line; then stop it, and return control to @value{GDBN}.
6399 Like the @code{step} command, @code{reverse-step} will only stop
6400 at the beginning of a source line. It ``un-executes'' the previously
6401 executed source line. If the previous source line included calls to
6402 debuggable functions, @code{reverse-step} will step (backward) into
6403 the called function, stopping at the beginning of the @emph{last}
6404 statement in the called function (typically a return statement).
6406 Also, as with the @code{step} command, if non-debuggable functions are
6407 called, @code{reverse-step} will run thru them backward without stopping.
6409 @kindex reverse-stepi
6410 @kindex rsi @r{(@code{reverse-stepi})}
6411 @item reverse-stepi @r{[}@var{count}@r{]}
6412 Reverse-execute one machine instruction. Note that the instruction
6413 to be reverse-executed is @emph{not} the one pointed to by the program
6414 counter, but the instruction executed prior to that one. For instance,
6415 if the last instruction was a jump, @code{reverse-stepi} will take you
6416 back from the destination of the jump to the jump instruction itself.
6418 @kindex reverse-next
6419 @kindex rn @r{(@code{reverse-next})}
6420 @item reverse-next @r{[}@var{count}@r{]}
6421 Run backward to the beginning of the previous line executed in
6422 the current (innermost) stack frame. If the line contains function
6423 calls, they will be ``un-executed'' without stopping. Starting from
6424 the first line of a function, @code{reverse-next} will take you back
6425 to the caller of that function, @emph{before} the function was called,
6426 just as the normal @code{next} command would take you from the last
6427 line of a function back to its return to its caller
6428 @footnote{Unless the code is too heavily optimized.}.
6430 @kindex reverse-nexti
6431 @kindex rni @r{(@code{reverse-nexti})}
6432 @item reverse-nexti @r{[}@var{count}@r{]}
6433 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
6434 in reverse, except that called functions are ``un-executed'' atomically.
6435 That is, if the previously executed instruction was a return from
6436 another function, @code{reverse-nexti} will continue to execute
6437 in reverse until the call to that function (from the current stack
6440 @kindex reverse-finish
6441 @item reverse-finish
6442 Just as the @code{finish} command takes you to the point where the
6443 current function returns, @code{reverse-finish} takes you to the point
6444 where it was called. Instead of ending up at the end of the current
6445 function invocation, you end up at the beginning.
6447 @kindex set exec-direction
6448 @item set exec-direction
6449 Set the direction of target execution.
6450 @item set exec-direction reverse
6451 @cindex execute forward or backward in time
6452 @value{GDBN} will perform all execution commands in reverse, until the
6453 exec-direction mode is changed to ``forward''. Affected commands include
6454 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
6455 command cannot be used in reverse mode.
6456 @item set exec-direction forward
6457 @value{GDBN} will perform all execution commands in the normal fashion.
6458 This is the default.
6462 @node Process Record and Replay
6463 @chapter Recording Inferior's Execution and Replaying It
6464 @cindex process record and replay
6465 @cindex recording inferior's execution and replaying it
6467 On some platforms, @value{GDBN} provides a special @dfn{process record
6468 and replay} target that can record a log of the process execution, and
6469 replay it later with both forward and reverse execution commands.
6472 When this target is in use, if the execution log includes the record
6473 for the next instruction, @value{GDBN} will debug in @dfn{replay
6474 mode}. In the replay mode, the inferior does not really execute code
6475 instructions. Instead, all the events that normally happen during
6476 code execution are taken from the execution log. While code is not
6477 really executed in replay mode, the values of registers (including the
6478 program counter register) and the memory of the inferior are still
6479 changed as they normally would. Their contents are taken from the
6483 If the record for the next instruction is not in the execution log,
6484 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6485 inferior executes normally, and @value{GDBN} records the execution log
6488 The process record and replay target supports reverse execution
6489 (@pxref{Reverse Execution}), even if the platform on which the
6490 inferior runs does not. However, the reverse execution is limited in
6491 this case by the range of the instructions recorded in the execution
6492 log. In other words, reverse execution on platforms that don't
6493 support it directly can only be done in the replay mode.
6495 When debugging in the reverse direction, @value{GDBN} will work in
6496 replay mode as long as the execution log includes the record for the
6497 previous instruction; otherwise, it will work in record mode, if the
6498 platform supports reverse execution, or stop if not.
6500 For architecture environments that support process record and replay,
6501 @value{GDBN} provides the following commands:
6504 @kindex target record
6505 @kindex target record-full
6506 @kindex target record-btrace
6509 @kindex record btrace
6510 @kindex record btrace bts
6511 @kindex record btrace pt
6517 @kindex rec btrace bts
6518 @kindex rec btrace pt
6521 @item record @var{method}
6522 This command starts the process record and replay target. The
6523 recording method can be specified as parameter. Without a parameter
6524 the command uses the @code{full} recording method. The following
6525 recording methods are available:
6529 Full record/replay recording using @value{GDBN}'s software record and
6530 replay implementation. This method allows replaying and reverse
6533 @item btrace @var{format}
6534 Hardware-supported instruction recording. This method does not record
6535 data. Further, the data is collected in a ring buffer so old data will
6536 be overwritten when the buffer is full. It allows limited reverse
6537 execution. Variables and registers are not available during reverse
6540 The recording format can be specified as parameter. Without a parameter
6541 the command chooses the recording format. The following recording
6542 formats are available:
6546 @cindex branch trace store
6547 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
6548 this format, the processor stores a from/to record for each executed
6549 branch in the btrace ring buffer.
6552 @cindex Intel Processor Trace
6553 Use the @dfn{Intel Processor Trace} recording format. In this
6554 format, the processor stores the execution trace in a compressed form
6555 that is afterwards decoded by @value{GDBN}.
6557 The trace can be recorded with very low overhead. The compressed
6558 trace format also allows small trace buffers to already contain a big
6559 number of instructions compared to @acronym{BTS}.
6561 Decoding the recorded execution trace, on the other hand, is more
6562 expensive than decoding @acronym{BTS} trace. This is mostly due to the
6563 increased number of instructions to process. You should increase the
6564 buffer-size with care.
6567 Not all recording formats may be available on all processors.
6570 The process record and replay target can only debug a process that is
6571 already running. Therefore, you need first to start the process with
6572 the @kbd{run} or @kbd{start} commands, and then start the recording
6573 with the @kbd{record @var{method}} command.
6575 @cindex displaced stepping, and process record and replay
6576 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
6577 will be automatically disabled when process record and replay target
6578 is started. That's because the process record and replay target
6579 doesn't support displaced stepping.
6581 @cindex non-stop mode, and process record and replay
6582 @cindex asynchronous execution, and process record and replay
6583 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
6584 the asynchronous execution mode (@pxref{Background Execution}), not
6585 all recording methods are available. The @code{full} recording method
6586 does not support these two modes.
6591 Stop the process record and replay target. When process record and
6592 replay target stops, the entire execution log will be deleted and the
6593 inferior will either be terminated, or will remain in its final state.
6595 When you stop the process record and replay target in record mode (at
6596 the end of the execution log), the inferior will be stopped at the
6597 next instruction that would have been recorded. In other words, if
6598 you record for a while and then stop recording, the inferior process
6599 will be left in the same state as if the recording never happened.
6601 On the other hand, if the process record and replay target is stopped
6602 while in replay mode (that is, not at the end of the execution log,
6603 but at some earlier point), the inferior process will become ``live''
6604 at that earlier state, and it will then be possible to continue the
6605 usual ``live'' debugging of the process from that state.
6607 When the inferior process exits, or @value{GDBN} detaches from it,
6608 process record and replay target will automatically stop itself.
6612 Go to a specific location in the execution log. There are several
6613 ways to specify the location to go to:
6616 @item record goto begin
6617 @itemx record goto start
6618 Go to the beginning of the execution log.
6620 @item record goto end
6621 Go to the end of the execution log.
6623 @item record goto @var{n}
6624 Go to instruction number @var{n} in the execution log.
6628 @item record save @var{filename}
6629 Save the execution log to a file @file{@var{filename}}.
6630 Default filename is @file{gdb_record.@var{process_id}}, where
6631 @var{process_id} is the process ID of the inferior.
6633 This command may not be available for all recording methods.
6635 @kindex record restore
6636 @item record restore @var{filename}
6637 Restore the execution log from a file @file{@var{filename}}.
6638 File must have been created with @code{record save}.
6640 @kindex set record full
6641 @item set record full insn-number-max @var{limit}
6642 @itemx set record full insn-number-max unlimited
6643 Set the limit of instructions to be recorded for the @code{full}
6644 recording method. Default value is 200000.
6646 If @var{limit} is a positive number, then @value{GDBN} will start
6647 deleting instructions from the log once the number of the record
6648 instructions becomes greater than @var{limit}. For every new recorded
6649 instruction, @value{GDBN} will delete the earliest recorded
6650 instruction to keep the number of recorded instructions at the limit.
6651 (Since deleting recorded instructions loses information, @value{GDBN}
6652 lets you control what happens when the limit is reached, by means of
6653 the @code{stop-at-limit} option, described below.)
6655 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
6656 delete recorded instructions from the execution log. The number of
6657 recorded instructions is limited only by the available memory.
6659 @kindex show record full
6660 @item show record full insn-number-max
6661 Show the limit of instructions to be recorded with the @code{full}
6664 @item set record full stop-at-limit
6665 Control the behavior of the @code{full} recording method when the
6666 number of recorded instructions reaches the limit. If ON (the
6667 default), @value{GDBN} will stop when the limit is reached for the
6668 first time and ask you whether you want to stop the inferior or
6669 continue running it and recording the execution log. If you decide
6670 to continue recording, each new recorded instruction will cause the
6671 oldest one to be deleted.
6673 If this option is OFF, @value{GDBN} will automatically delete the
6674 oldest record to make room for each new one, without asking.
6676 @item show record full stop-at-limit
6677 Show the current setting of @code{stop-at-limit}.
6679 @item set record full memory-query
6680 Control the behavior when @value{GDBN} is unable to record memory
6681 changes caused by an instruction for the @code{full} recording method.
6682 If ON, @value{GDBN} will query whether to stop the inferior in that
6685 If this option is OFF (the default), @value{GDBN} will automatically
6686 ignore the effect of such instructions on memory. Later, when
6687 @value{GDBN} replays this execution log, it will mark the log of this
6688 instruction as not accessible, and it will not affect the replay
6691 @item show record full memory-query
6692 Show the current setting of @code{memory-query}.
6694 @kindex set record btrace
6695 The @code{btrace} record target does not trace data. As a
6696 convenience, when replaying, @value{GDBN} reads read-only memory off
6697 the live program directly, assuming that the addresses of the
6698 read-only areas don't change. This for example makes it possible to
6699 disassemble code while replaying, but not to print variables.
6700 In some cases, being able to inspect variables might be useful.
6701 You can use the following command for that:
6703 @item set record btrace replay-memory-access
6704 Control the behavior of the @code{btrace} recording method when
6705 accessing memory during replay. If @code{read-only} (the default),
6706 @value{GDBN} will only allow accesses to read-only memory.
6707 If @code{read-write}, @value{GDBN} will allow accesses to read-only
6708 and to read-write memory. Beware that the accessed memory corresponds
6709 to the live target and not necessarily to the current replay
6712 @kindex show record btrace
6713 @item show record btrace replay-memory-access
6714 Show the current setting of @code{replay-memory-access}.
6716 @kindex set record btrace bts
6717 @item set record btrace bts buffer-size @var{size}
6718 @itemx set record btrace bts buffer-size unlimited
6719 Set the requested ring buffer size for branch tracing in @acronym{BTS}
6720 format. Default is 64KB.
6722 If @var{size} is a positive number, then @value{GDBN} will try to
6723 allocate a buffer of at least @var{size} bytes for each new thread
6724 that uses the btrace recording method and the @acronym{BTS} format.
6725 The actually obtained buffer size may differ from the requested
6726 @var{size}. Use the @code{info record} command to see the actual
6727 buffer size for each thread that uses the btrace recording method and
6728 the @acronym{BTS} format.
6730 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
6731 allocate a buffer of 4MB.
6733 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
6734 also need longer to process the branch trace data before it can be used.
6736 @item show record btrace bts buffer-size @var{size}
6737 Show the current setting of the requested ring buffer size for branch
6738 tracing in @acronym{BTS} format.
6740 @kindex set record btrace pt
6741 @item set record btrace pt buffer-size @var{size}
6742 @itemx set record btrace pt buffer-size unlimited
6743 Set the requested ring buffer size for branch tracing in Intel
6744 Processor Trace format. Default is 16KB.
6746 If @var{size} is a positive number, then @value{GDBN} will try to
6747 allocate a buffer of at least @var{size} bytes for each new thread
6748 that uses the btrace recording method and the Intel Processor Trace
6749 format. The actually obtained buffer size may differ from the
6750 requested @var{size}. Use the @code{info record} command to see the
6751 actual buffer size for each thread.
6753 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
6754 allocate a buffer of 4MB.
6756 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
6757 also need longer to process the branch trace data before it can be used.
6759 @item show record btrace pt buffer-size @var{size}
6760 Show the current setting of the requested ring buffer size for branch
6761 tracing in Intel Processor Trace format.
6765 Show various statistics about the recording depending on the recording
6770 For the @code{full} recording method, it shows the state of process
6771 record and its in-memory execution log buffer, including:
6775 Whether in record mode or replay mode.
6777 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
6779 Highest recorded instruction number.
6781 Current instruction about to be replayed (if in replay mode).
6783 Number of instructions contained in the execution log.
6785 Maximum number of instructions that may be contained in the execution log.
6789 For the @code{btrace} recording method, it shows:
6795 Number of instructions that have been recorded.
6797 Number of blocks of sequential control-flow formed by the recorded
6800 Whether in record mode or replay mode.
6803 For the @code{bts} recording format, it also shows:
6806 Size of the perf ring buffer.
6809 For the @code{pt} recording format, it also shows:
6812 Size of the perf ring buffer.
6816 @kindex record delete
6819 When record target runs in replay mode (``in the past''), delete the
6820 subsequent execution log and begin to record a new execution log starting
6821 from the current address. This means you will abandon the previously
6822 recorded ``future'' and begin recording a new ``future''.
6824 @kindex record instruction-history
6825 @kindex rec instruction-history
6826 @item record instruction-history
6827 Disassembles instructions from the recorded execution log. By
6828 default, ten instructions are disassembled. This can be changed using
6829 the @code{set record instruction-history-size} command. Instructions
6830 are printed in execution order.
6832 It can also print mixed source+disassembly if you specify the the
6833 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
6834 as well as in symbolic form by specifying the @code{/r} modifier.
6836 The current position marker is printed for the instruction at the
6837 current program counter value. This instruction can appear multiple
6838 times in the trace and the current position marker will be printed
6839 every time. To omit the current position marker, specify the
6842 To better align the printed instructions when the trace contains
6843 instructions from more than one function, the function name may be
6844 omitted by specifying the @code{/f} modifier.
6846 Speculatively executed instructions are prefixed with @samp{?}. This
6847 feature is not available for all recording formats.
6849 There are several ways to specify what part of the execution log to
6853 @item record instruction-history @var{insn}
6854 Disassembles ten instructions starting from instruction number
6857 @item record instruction-history @var{insn}, +/-@var{n}
6858 Disassembles @var{n} instructions around instruction number
6859 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
6860 @var{n} instructions after instruction number @var{insn}. If
6861 @var{n} is preceded with @code{-}, disassembles @var{n}
6862 instructions before instruction number @var{insn}.
6864 @item record instruction-history
6865 Disassembles ten more instructions after the last disassembly.
6867 @item record instruction-history -
6868 Disassembles ten more instructions before the last disassembly.
6870 @item record instruction-history @var{begin}, @var{end}
6871 Disassembles instructions beginning with instruction number
6872 @var{begin} until instruction number @var{end}. The instruction
6873 number @var{end} is included.
6876 This command may not be available for all recording methods.
6879 @item set record instruction-history-size @var{size}
6880 @itemx set record instruction-history-size unlimited
6881 Define how many instructions to disassemble in the @code{record
6882 instruction-history} command. The default value is 10.
6883 A @var{size} of @code{unlimited} means unlimited instructions.
6886 @item show record instruction-history-size
6887 Show how many instructions to disassemble in the @code{record
6888 instruction-history} command.
6890 @kindex record function-call-history
6891 @kindex rec function-call-history
6892 @item record function-call-history
6893 Prints the execution history at function granularity. It prints one
6894 line for each sequence of instructions that belong to the same
6895 function giving the name of that function, the source lines
6896 for this instruction sequence (if the @code{/l} modifier is
6897 specified), and the instructions numbers that form the sequence (if
6898 the @code{/i} modifier is specified). The function names are indented
6899 to reflect the call stack depth if the @code{/c} modifier is
6900 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
6904 (@value{GDBP}) @b{list 1, 10}
6915 (@value{GDBP}) @b{record function-call-history /ilc}
6916 1 bar inst 1,4 at foo.c:6,8
6917 2 foo inst 5,10 at foo.c:2,3
6918 3 bar inst 11,13 at foo.c:9,10
6921 By default, ten lines are printed. This can be changed using the
6922 @code{set record function-call-history-size} command. Functions are
6923 printed in execution order. There are several ways to specify what
6927 @item record function-call-history @var{func}
6928 Prints ten functions starting from function number @var{func}.
6930 @item record function-call-history @var{func}, +/-@var{n}
6931 Prints @var{n} functions around function number @var{func}. If
6932 @var{n} is preceded with @code{+}, prints @var{n} functions after
6933 function number @var{func}. If @var{n} is preceded with @code{-},
6934 prints @var{n} functions before function number @var{func}.
6936 @item record function-call-history
6937 Prints ten more functions after the last ten-line print.
6939 @item record function-call-history -
6940 Prints ten more functions before the last ten-line print.
6942 @item record function-call-history @var{begin}, @var{end}
6943 Prints functions beginning with function number @var{begin} until
6944 function number @var{end}. The function number @var{end} is included.
6947 This command may not be available for all recording methods.
6949 @item set record function-call-history-size @var{size}
6950 @itemx set record function-call-history-size unlimited
6951 Define how many lines to print in the
6952 @code{record function-call-history} command. The default value is 10.
6953 A size of @code{unlimited} means unlimited lines.
6955 @item show record function-call-history-size
6956 Show how many lines to print in the
6957 @code{record function-call-history} command.
6962 @chapter Examining the Stack
6964 When your program has stopped, the first thing you need to know is where it
6965 stopped and how it got there.
6968 Each time your program performs a function call, information about the call
6970 That information includes the location of the call in your program,
6971 the arguments of the call,
6972 and the local variables of the function being called.
6973 The information is saved in a block of data called a @dfn{stack frame}.
6974 The stack frames are allocated in a region of memory called the @dfn{call
6977 When your program stops, the @value{GDBN} commands for examining the
6978 stack allow you to see all of this information.
6980 @cindex selected frame
6981 One of the stack frames is @dfn{selected} by @value{GDBN} and many
6982 @value{GDBN} commands refer implicitly to the selected frame. In
6983 particular, whenever you ask @value{GDBN} for the value of a variable in
6984 your program, the value is found in the selected frame. There are
6985 special @value{GDBN} commands to select whichever frame you are
6986 interested in. @xref{Selection, ,Selecting a Frame}.
6988 When your program stops, @value{GDBN} automatically selects the
6989 currently executing frame and describes it briefly, similar to the
6990 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
6993 * Frames:: Stack frames
6994 * Backtrace:: Backtraces
6995 * Selection:: Selecting a frame
6996 * Frame Info:: Information on a frame
6997 * Frame Filter Management:: Managing frame filters
7002 @section Stack Frames
7004 @cindex frame, definition
7006 The call stack is divided up into contiguous pieces called @dfn{stack
7007 frames}, or @dfn{frames} for short; each frame is the data associated
7008 with one call to one function. The frame contains the arguments given
7009 to the function, the function's local variables, and the address at
7010 which the function is executing.
7012 @cindex initial frame
7013 @cindex outermost frame
7014 @cindex innermost frame
7015 When your program is started, the stack has only one frame, that of the
7016 function @code{main}. This is called the @dfn{initial} frame or the
7017 @dfn{outermost} frame. Each time a function is called, a new frame is
7018 made. Each time a function returns, the frame for that function invocation
7019 is eliminated. If a function is recursive, there can be many frames for
7020 the same function. The frame for the function in which execution is
7021 actually occurring is called the @dfn{innermost} frame. This is the most
7022 recently created of all the stack frames that still exist.
7024 @cindex frame pointer
7025 Inside your program, stack frames are identified by their addresses. A
7026 stack frame consists of many bytes, each of which has its own address; each
7027 kind of computer has a convention for choosing one byte whose
7028 address serves as the address of the frame. Usually this address is kept
7029 in a register called the @dfn{frame pointer register}
7030 (@pxref{Registers, $fp}) while execution is going on in that frame.
7032 @cindex frame number
7033 @value{GDBN} assigns numbers to all existing stack frames, starting with
7034 zero for the innermost frame, one for the frame that called it,
7035 and so on upward. These numbers do not really exist in your program;
7036 they are assigned by @value{GDBN} to give you a way of designating stack
7037 frames in @value{GDBN} commands.
7039 @c The -fomit-frame-pointer below perennially causes hbox overflow
7040 @c underflow problems.
7041 @cindex frameless execution
7042 Some compilers provide a way to compile functions so that they operate
7043 without stack frames. (For example, the @value{NGCC} option
7045 @samp{-fomit-frame-pointer}
7047 generates functions without a frame.)
7048 This is occasionally done with heavily used library functions to save
7049 the frame setup time. @value{GDBN} has limited facilities for dealing
7050 with these function invocations. If the innermost function invocation
7051 has no stack frame, @value{GDBN} nevertheless regards it as though
7052 it had a separate frame, which is numbered zero as usual, allowing
7053 correct tracing of the function call chain. However, @value{GDBN} has
7054 no provision for frameless functions elsewhere in the stack.
7060 @cindex call stack traces
7061 A backtrace is a summary of how your program got where it is. It shows one
7062 line per frame, for many frames, starting with the currently executing
7063 frame (frame zero), followed by its caller (frame one), and on up the
7066 @anchor{backtrace-command}
7069 @kindex bt @r{(@code{backtrace})}
7072 Print a backtrace of the entire stack: one line per frame for all
7073 frames in the stack.
7075 You can stop the backtrace at any time by typing the system interrupt
7076 character, normally @kbd{Ctrl-c}.
7078 @item backtrace @var{n}
7080 Similar, but print only the innermost @var{n} frames.
7082 @item backtrace -@var{n}
7084 Similar, but print only the outermost @var{n} frames.
7086 @item backtrace full
7088 @itemx bt full @var{n}
7089 @itemx bt full -@var{n}
7090 Print the values of the local variables also. As described above,
7091 @var{n} specifies the number of frames to print.
7093 @item backtrace no-filters
7094 @itemx bt no-filters
7095 @itemx bt no-filters @var{n}
7096 @itemx bt no-filters -@var{n}
7097 @itemx bt no-filters full
7098 @itemx bt no-filters full @var{n}
7099 @itemx bt no-filters full -@var{n}
7100 Do not run Python frame filters on this backtrace. @xref{Frame
7101 Filter API}, for more information. Additionally use @ref{disable
7102 frame-filter all} to turn off all frame filters. This is only
7103 relevant when @value{GDBN} has been configured with @code{Python}
7109 The names @code{where} and @code{info stack} (abbreviated @code{info s})
7110 are additional aliases for @code{backtrace}.
7112 @cindex multiple threads, backtrace
7113 In a multi-threaded program, @value{GDBN} by default shows the
7114 backtrace only for the current thread. To display the backtrace for
7115 several or all of the threads, use the command @code{thread apply}
7116 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
7117 apply all backtrace}, @value{GDBN} will display the backtrace for all
7118 the threads; this is handy when you debug a core dump of a
7119 multi-threaded program.
7121 Each line in the backtrace shows the frame number and the function name.
7122 The program counter value is also shown---unless you use @code{set
7123 print address off}. The backtrace also shows the source file name and
7124 line number, as well as the arguments to the function. The program
7125 counter value is omitted if it is at the beginning of the code for that
7128 Here is an example of a backtrace. It was made with the command
7129 @samp{bt 3}, so it shows the innermost three frames.
7133 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7135 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
7136 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
7138 (More stack frames follow...)
7143 The display for frame zero does not begin with a program counter
7144 value, indicating that your program has stopped at the beginning of the
7145 code for line @code{993} of @code{builtin.c}.
7148 The value of parameter @code{data} in frame 1 has been replaced by
7149 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
7150 only if it is a scalar (integer, pointer, enumeration, etc). See command
7151 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
7152 on how to configure the way function parameter values are printed.
7154 @cindex optimized out, in backtrace
7155 @cindex function call arguments, optimized out
7156 If your program was compiled with optimizations, some compilers will
7157 optimize away arguments passed to functions if those arguments are
7158 never used after the call. Such optimizations generate code that
7159 passes arguments through registers, but doesn't store those arguments
7160 in the stack frame. @value{GDBN} has no way of displaying such
7161 arguments in stack frames other than the innermost one. Here's what
7162 such a backtrace might look like:
7166 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7168 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
7169 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
7171 (More stack frames follow...)
7176 The values of arguments that were not saved in their stack frames are
7177 shown as @samp{<optimized out>}.
7179 If you need to display the values of such optimized-out arguments,
7180 either deduce that from other variables whose values depend on the one
7181 you are interested in, or recompile without optimizations.
7183 @cindex backtrace beyond @code{main} function
7184 @cindex program entry point
7185 @cindex startup code, and backtrace
7186 Most programs have a standard user entry point---a place where system
7187 libraries and startup code transition into user code. For C this is
7188 @code{main}@footnote{
7189 Note that embedded programs (the so-called ``free-standing''
7190 environment) are not required to have a @code{main} function as the
7191 entry point. They could even have multiple entry points.}.
7192 When @value{GDBN} finds the entry function in a backtrace
7193 it will terminate the backtrace, to avoid tracing into highly
7194 system-specific (and generally uninteresting) code.
7196 If you need to examine the startup code, or limit the number of levels
7197 in a backtrace, you can change this behavior:
7200 @item set backtrace past-main
7201 @itemx set backtrace past-main on
7202 @kindex set backtrace
7203 Backtraces will continue past the user entry point.
7205 @item set backtrace past-main off
7206 Backtraces will stop when they encounter the user entry point. This is the
7209 @item show backtrace past-main
7210 @kindex show backtrace
7211 Display the current user entry point backtrace policy.
7213 @item set backtrace past-entry
7214 @itemx set backtrace past-entry on
7215 Backtraces will continue past the internal entry point of an application.
7216 This entry point is encoded by the linker when the application is built,
7217 and is likely before the user entry point @code{main} (or equivalent) is called.
7219 @item set backtrace past-entry off
7220 Backtraces will stop when they encounter the internal entry point of an
7221 application. This is the default.
7223 @item show backtrace past-entry
7224 Display the current internal entry point backtrace policy.
7226 @item set backtrace limit @var{n}
7227 @itemx set backtrace limit 0
7228 @itemx set backtrace limit unlimited
7229 @cindex backtrace limit
7230 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
7231 or zero means unlimited levels.
7233 @item show backtrace limit
7234 Display the current limit on backtrace levels.
7237 You can control how file names are displayed.
7240 @item set filename-display
7241 @itemx set filename-display relative
7242 @cindex filename-display
7243 Display file names relative to the compilation directory. This is the default.
7245 @item set filename-display basename
7246 Display only basename of a filename.
7248 @item set filename-display absolute
7249 Display an absolute filename.
7251 @item show filename-display
7252 Show the current way to display filenames.
7256 @section Selecting a Frame
7258 Most commands for examining the stack and other data in your program work on
7259 whichever stack frame is selected at the moment. Here are the commands for
7260 selecting a stack frame; all of them finish by printing a brief description
7261 of the stack frame just selected.
7264 @kindex frame@r{, selecting}
7265 @kindex f @r{(@code{frame})}
7268 Select frame number @var{n}. Recall that frame zero is the innermost
7269 (currently executing) frame, frame one is the frame that called the
7270 innermost one, and so on. The highest-numbered frame is the one for
7273 @item frame @var{stack-addr} [ @var{pc-addr} ]
7274 @itemx f @var{stack-addr} [ @var{pc-addr} ]
7275 Select the frame at address @var{stack-addr}. This is useful mainly if the
7276 chaining of stack frames has been damaged by a bug, making it
7277 impossible for @value{GDBN} to assign numbers properly to all frames. In
7278 addition, this can be useful when your program has multiple stacks and
7279 switches between them. The optional @var{pc-addr} can also be given to
7280 specify the value of PC for the stack frame.
7284 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
7285 numbers @var{n}, this advances toward the outermost frame, to higher
7286 frame numbers, to frames that have existed longer.
7289 @kindex do @r{(@code{down})}
7291 Move @var{n} frames down the stack; @var{n} defaults to 1. For
7292 positive numbers @var{n}, this advances toward the innermost frame, to
7293 lower frame numbers, to frames that were created more recently.
7294 You may abbreviate @code{down} as @code{do}.
7297 All of these commands end by printing two lines of output describing the
7298 frame. The first line shows the frame number, the function name, the
7299 arguments, and the source file and line number of execution in that
7300 frame. The second line shows the text of that source line.
7308 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
7310 10 read_input_file (argv[i]);
7314 After such a printout, the @code{list} command with no arguments
7315 prints ten lines centered on the point of execution in the frame.
7316 You can also edit the program at the point of execution with your favorite
7317 editing program by typing @code{edit}.
7318 @xref{List, ,Printing Source Lines},
7322 @kindex select-frame
7324 The @code{select-frame} command is a variant of @code{frame} that does
7325 not display the new frame after selecting it. This command is
7326 intended primarily for use in @value{GDBN} command scripts, where the
7327 output might be unnecessary and distracting.
7329 @kindex down-silently
7331 @item up-silently @var{n}
7332 @itemx down-silently @var{n}
7333 These two commands are variants of @code{up} and @code{down},
7334 respectively; they differ in that they do their work silently, without
7335 causing display of the new frame. They are intended primarily for use
7336 in @value{GDBN} command scripts, where the output might be unnecessary and
7341 @section Information About a Frame
7343 There are several other commands to print information about the selected
7349 When used without any argument, this command does not change which
7350 frame is selected, but prints a brief description of the currently
7351 selected stack frame. It can be abbreviated @code{f}. With an
7352 argument, this command is used to select a stack frame.
7353 @xref{Selection, ,Selecting a Frame}.
7356 @kindex info f @r{(@code{info frame})}
7359 This command prints a verbose description of the selected stack frame,
7364 the address of the frame
7366 the address of the next frame down (called by this frame)
7368 the address of the next frame up (caller of this frame)
7370 the language in which the source code corresponding to this frame is written
7372 the address of the frame's arguments
7374 the address of the frame's local variables
7376 the program counter saved in it (the address of execution in the caller frame)
7378 which registers were saved in the frame
7381 @noindent The verbose description is useful when
7382 something has gone wrong that has made the stack format fail to fit
7383 the usual conventions.
7385 @item info frame @var{addr}
7386 @itemx info f @var{addr}
7387 Print a verbose description of the frame at address @var{addr}, without
7388 selecting that frame. The selected frame remains unchanged by this
7389 command. This requires the same kind of address (more than one for some
7390 architectures) that you specify in the @code{frame} command.
7391 @xref{Selection, ,Selecting a Frame}.
7395 Print the arguments of the selected frame, each on a separate line.
7399 Print the local variables of the selected frame, each on a separate
7400 line. These are all variables (declared either static or automatic)
7401 accessible at the point of execution of the selected frame.
7405 @node Frame Filter Management
7406 @section Management of Frame Filters.
7407 @cindex managing frame filters
7409 Frame filters are Python based utilities to manage and decorate the
7410 output of frames. @xref{Frame Filter API}, for further information.
7412 Managing frame filters is performed by several commands available
7413 within @value{GDBN}, detailed here.
7416 @kindex info frame-filter
7417 @item info frame-filter
7418 Print a list of installed frame filters from all dictionaries, showing
7419 their name, priority and enabled status.
7421 @kindex disable frame-filter
7422 @anchor{disable frame-filter all}
7423 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
7424 Disable a frame filter in the dictionary matching
7425 @var{filter-dictionary} and @var{filter-name}. The
7426 @var{filter-dictionary} may be @code{all}, @code{global},
7427 @code{progspace}, or the name of the object file where the frame filter
7428 dictionary resides. When @code{all} is specified, all frame filters
7429 across all dictionaries are disabled. The @var{filter-name} is the name
7430 of the frame filter and is used when @code{all} is not the option for
7431 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
7432 may be enabled again later.
7434 @kindex enable frame-filter
7435 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
7436 Enable a frame filter in the dictionary matching
7437 @var{filter-dictionary} and @var{filter-name}. The
7438 @var{filter-dictionary} may be @code{all}, @code{global},
7439 @code{progspace} or the name of the object file where the frame filter
7440 dictionary resides. When @code{all} is specified, all frame filters across
7441 all dictionaries are enabled. The @var{filter-name} is the name of the frame
7442 filter and is used when @code{all} is not the option for
7443 @var{filter-dictionary}.
7448 (gdb) info frame-filter
7450 global frame-filters:
7451 Priority Enabled Name
7452 1000 No PrimaryFunctionFilter
7455 progspace /build/test frame-filters:
7456 Priority Enabled Name
7457 100 Yes ProgspaceFilter
7459 objfile /build/test frame-filters:
7460 Priority Enabled Name
7461 999 Yes BuildProgra Filter
7463 (gdb) disable frame-filter /build/test BuildProgramFilter
7464 (gdb) info frame-filter
7466 global frame-filters:
7467 Priority Enabled Name
7468 1000 No PrimaryFunctionFilter
7471 progspace /build/test frame-filters:
7472 Priority Enabled Name
7473 100 Yes ProgspaceFilter
7475 objfile /build/test frame-filters:
7476 Priority Enabled Name
7477 999 No BuildProgramFilter
7479 (gdb) enable frame-filter global PrimaryFunctionFilter
7480 (gdb) info frame-filter
7482 global frame-filters:
7483 Priority Enabled Name
7484 1000 Yes PrimaryFunctionFilter
7487 progspace /build/test frame-filters:
7488 Priority Enabled Name
7489 100 Yes ProgspaceFilter
7491 objfile /build/test frame-filters:
7492 Priority Enabled Name
7493 999 No BuildProgramFilter
7496 @kindex set frame-filter priority
7497 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
7498 Set the @var{priority} of a frame filter in the dictionary matching
7499 @var{filter-dictionary}, and the frame filter name matching
7500 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
7501 @code{progspace} or the name of the object file where the frame filter
7502 dictionary resides. The @var{priority} is an integer.
7504 @kindex show frame-filter priority
7505 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
7506 Show the @var{priority} of a frame filter in the dictionary matching
7507 @var{filter-dictionary}, and the frame filter name matching
7508 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
7509 @code{progspace} or the name of the object file where the frame filter
7515 (gdb) info frame-filter
7517 global frame-filters:
7518 Priority Enabled Name
7519 1000 Yes PrimaryFunctionFilter
7522 progspace /build/test frame-filters:
7523 Priority Enabled Name
7524 100 Yes ProgspaceFilter
7526 objfile /build/test frame-filters:
7527 Priority Enabled Name
7528 999 No BuildProgramFilter
7530 (gdb) set frame-filter priority global Reverse 50
7531 (gdb) info frame-filter
7533 global frame-filters:
7534 Priority Enabled Name
7535 1000 Yes PrimaryFunctionFilter
7538 progspace /build/test frame-filters:
7539 Priority Enabled Name
7540 100 Yes ProgspaceFilter
7542 objfile /build/test frame-filters:
7543 Priority Enabled Name
7544 999 No BuildProgramFilter
7549 @chapter Examining Source Files
7551 @value{GDBN} can print parts of your program's source, since the debugging
7552 information recorded in the program tells @value{GDBN} what source files were
7553 used to build it. When your program stops, @value{GDBN} spontaneously prints
7554 the line where it stopped. Likewise, when you select a stack frame
7555 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
7556 execution in that frame has stopped. You can print other portions of
7557 source files by explicit command.
7559 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
7560 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
7561 @value{GDBN} under @sc{gnu} Emacs}.
7564 * List:: Printing source lines
7565 * Specify Location:: How to specify code locations
7566 * Edit:: Editing source files
7567 * Search:: Searching source files
7568 * Source Path:: Specifying source directories
7569 * Machine Code:: Source and machine code
7573 @section Printing Source Lines
7576 @kindex l @r{(@code{list})}
7577 To print lines from a source file, use the @code{list} command
7578 (abbreviated @code{l}). By default, ten lines are printed.
7579 There are several ways to specify what part of the file you want to
7580 print; see @ref{Specify Location}, for the full list.
7582 Here are the forms of the @code{list} command most commonly used:
7585 @item list @var{linenum}
7586 Print lines centered around line number @var{linenum} in the
7587 current source file.
7589 @item list @var{function}
7590 Print lines centered around the beginning of function
7594 Print more lines. If the last lines printed were printed with a
7595 @code{list} command, this prints lines following the last lines
7596 printed; however, if the last line printed was a solitary line printed
7597 as part of displaying a stack frame (@pxref{Stack, ,Examining the
7598 Stack}), this prints lines centered around that line.
7601 Print lines just before the lines last printed.
7604 @cindex @code{list}, how many lines to display
7605 By default, @value{GDBN} prints ten source lines with any of these forms of
7606 the @code{list} command. You can change this using @code{set listsize}:
7609 @kindex set listsize
7610 @item set listsize @var{count}
7611 @itemx set listsize unlimited
7612 Make the @code{list} command display @var{count} source lines (unless
7613 the @code{list} argument explicitly specifies some other number).
7614 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
7616 @kindex show listsize
7618 Display the number of lines that @code{list} prints.
7621 Repeating a @code{list} command with @key{RET} discards the argument,
7622 so it is equivalent to typing just @code{list}. This is more useful
7623 than listing the same lines again. An exception is made for an
7624 argument of @samp{-}; that argument is preserved in repetition so that
7625 each repetition moves up in the source file.
7627 In general, the @code{list} command expects you to supply zero, one or two
7628 @dfn{locations}. Locations specify source lines; there are several ways
7629 of writing them (@pxref{Specify Location}), but the effect is always
7630 to specify some source line.
7632 Here is a complete description of the possible arguments for @code{list}:
7635 @item list @var{location}
7636 Print lines centered around the line specified by @var{location}.
7638 @item list @var{first},@var{last}
7639 Print lines from @var{first} to @var{last}. Both arguments are
7640 locations. When a @code{list} command has two locations, and the
7641 source file of the second location is omitted, this refers to
7642 the same source file as the first location.
7644 @item list ,@var{last}
7645 Print lines ending with @var{last}.
7647 @item list @var{first},
7648 Print lines starting with @var{first}.
7651 Print lines just after the lines last printed.
7654 Print lines just before the lines last printed.
7657 As described in the preceding table.
7660 @node Specify Location
7661 @section Specifying a Location
7662 @cindex specifying location
7664 @cindex source location
7667 * Linespec Locations:: Linespec locations
7668 * Explicit Locations:: Explicit locations
7669 * Address Locations:: Address locations
7672 Several @value{GDBN} commands accept arguments that specify a location
7673 of your program's code. Since @value{GDBN} is a source-level
7674 debugger, a location usually specifies some line in the source code.
7675 Locations may be specified using three different formats:
7676 linespec locations, explicit locations, or address locations.
7678 @node Linespec Locations
7679 @subsection Linespec Locations
7680 @cindex linespec locations
7682 A @dfn{linespec} is a colon-separated list of source location parameters such
7683 as file name, function name, etc. Here are all the different ways of
7684 specifying a linespec:
7688 Specifies the line number @var{linenum} of the current source file.
7691 @itemx +@var{offset}
7692 Specifies the line @var{offset} lines before or after the @dfn{current
7693 line}. For the @code{list} command, the current line is the last one
7694 printed; for the breakpoint commands, this is the line at which
7695 execution stopped in the currently selected @dfn{stack frame}
7696 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
7697 used as the second of the two linespecs in a @code{list} command,
7698 this specifies the line @var{offset} lines up or down from the first
7701 @item @var{filename}:@var{linenum}
7702 Specifies the line @var{linenum} in the source file @var{filename}.
7703 If @var{filename} is a relative file name, then it will match any
7704 source file name with the same trailing components. For example, if
7705 @var{filename} is @samp{gcc/expr.c}, then it will match source file
7706 name of @file{/build/trunk/gcc/expr.c}, but not
7707 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
7709 @item @var{function}
7710 Specifies the line that begins the body of the function @var{function}.
7711 For example, in C, this is the line with the open brace.
7713 @item @var{function}:@var{label}
7714 Specifies the line where @var{label} appears in @var{function}.
7716 @item @var{filename}:@var{function}
7717 Specifies the line that begins the body of the function @var{function}
7718 in the file @var{filename}. You only need the file name with a
7719 function name to avoid ambiguity when there are identically named
7720 functions in different source files.
7723 Specifies the line at which the label named @var{label} appears
7724 in the function corresponding to the currently selected stack frame.
7725 If there is no current selected stack frame (for instance, if the inferior
7726 is not running), then @value{GDBN} will not search for a label.
7728 @cindex breakpoint at static probe point
7729 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
7730 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
7731 applications to embed static probes. @xref{Static Probe Points}, for more
7732 information on finding and using static probes. This form of linespec
7733 specifies the location of such a static probe.
7735 If @var{objfile} is given, only probes coming from that shared library
7736 or executable matching @var{objfile} as a regular expression are considered.
7737 If @var{provider} is given, then only probes from that provider are considered.
7738 If several probes match the spec, @value{GDBN} will insert a breakpoint at
7739 each one of those probes.
7742 @node Explicit Locations
7743 @subsection Explicit Locations
7744 @cindex explicit locations
7746 @dfn{Explicit locations} allow the user to directly specify the source
7747 location's parameters using option-value pairs.
7749 Explicit locations are useful when several functions, labels, or
7750 file names have the same name (base name for files) in the program's
7751 sources. In these cases, explicit locations point to the source
7752 line you meant more accurately and unambiguously. Also, using
7753 explicit locations might be faster in large programs.
7755 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
7756 defined in the file named @file{foo} or the label @code{bar} in a function
7757 named @code{foo}. @value{GDBN} must search either the file system or
7758 the symbol table to know.
7760 The list of valid explicit location options is summarized in the
7764 @item -source @var{filename}
7765 The value specifies the source file name. To differentiate between
7766 files with the same base name, prepend as many directories as is necessary
7767 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
7768 @value{GDBN} will use the first file it finds with the given base
7769 name. This option requires the use of either @code{-function} or @code{-line}.
7771 @item -function @var{function}
7772 The value specifies the name of a function. Operations
7773 on function locations unmodified by other options (such as @code{-label}
7774 or @code{-line}) refer to the line that begins the body of the function.
7775 In C, for example, this is the line with the open brace.
7777 @item -label @var{label}
7778 The value specifies the name of a label. When the function
7779 name is not specified, the label is searched in the function of the currently
7780 selected stack frame.
7782 @item -line @var{number}
7783 The value specifies a line offset for the location. The offset may either
7784 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
7785 the command. When specified without any other options, the line offset is
7786 relative to the current line.
7789 Explicit location options may be abbreviated by omitting any non-unique
7790 trailing characters from the option name, e.g., @code{break -s main.c -li 3}.
7792 @node Address Locations
7793 @subsection Address Locations
7794 @cindex address locations
7796 @dfn{Address locations} indicate a specific program address. They have
7797 the generalized form *@var{address}.
7799 For line-oriented commands, such as @code{list} and @code{edit}, this
7800 specifies a source line that contains @var{address}. For @code{break} and
7801 other breakpoint-oriented commands, this can be used to set breakpoints in
7802 parts of your program which do not have debugging information or
7805 Here @var{address} may be any expression valid in the current working
7806 language (@pxref{Languages, working language}) that specifies a code
7807 address. In addition, as a convenience, @value{GDBN} extends the
7808 semantics of expressions used in locations to cover several situations
7809 that frequently occur during debugging. Here are the various forms
7813 @item @var{expression}
7814 Any expression valid in the current working language.
7816 @item @var{funcaddr}
7817 An address of a function or procedure derived from its name. In C,
7818 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
7819 simply the function's name @var{function} (and actually a special case
7820 of a valid expression). In Pascal and Modula-2, this is
7821 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
7822 (although the Pascal form also works).
7824 This form specifies the address of the function's first instruction,
7825 before the stack frame and arguments have been set up.
7827 @item '@var{filename}':@var{funcaddr}
7828 Like @var{funcaddr} above, but also specifies the name of the source
7829 file explicitly. This is useful if the name of the function does not
7830 specify the function unambiguously, e.g., if there are several
7831 functions with identical names in different source files.
7835 @section Editing Source Files
7836 @cindex editing source files
7839 @kindex e @r{(@code{edit})}
7840 To edit the lines in a source file, use the @code{edit} command.
7841 The editing program of your choice
7842 is invoked with the current line set to
7843 the active line in the program.
7844 Alternatively, there are several ways to specify what part of the file you
7845 want to print if you want to see other parts of the program:
7848 @item edit @var{location}
7849 Edit the source file specified by @code{location}. Editing starts at
7850 that @var{location}, e.g., at the specified source line of the
7851 specified file. @xref{Specify Location}, for all the possible forms
7852 of the @var{location} argument; here are the forms of the @code{edit}
7853 command most commonly used:
7856 @item edit @var{number}
7857 Edit the current source file with @var{number} as the active line number.
7859 @item edit @var{function}
7860 Edit the file containing @var{function} at the beginning of its definition.
7865 @subsection Choosing your Editor
7866 You can customize @value{GDBN} to use any editor you want
7868 The only restriction is that your editor (say @code{ex}), recognizes the
7869 following command-line syntax:
7871 ex +@var{number} file
7873 The optional numeric value +@var{number} specifies the number of the line in
7874 the file where to start editing.}.
7875 By default, it is @file{@value{EDITOR}}, but you can change this
7876 by setting the environment variable @code{EDITOR} before using
7877 @value{GDBN}. For example, to configure @value{GDBN} to use the
7878 @code{vi} editor, you could use these commands with the @code{sh} shell:
7884 or in the @code{csh} shell,
7886 setenv EDITOR /usr/bin/vi
7891 @section Searching Source Files
7892 @cindex searching source files
7894 There are two commands for searching through the current source file for a
7899 @kindex forward-search
7900 @kindex fo @r{(@code{forward-search})}
7901 @item forward-search @var{regexp}
7902 @itemx search @var{regexp}
7903 The command @samp{forward-search @var{regexp}} checks each line,
7904 starting with the one following the last line listed, for a match for
7905 @var{regexp}. It lists the line that is found. You can use the
7906 synonym @samp{search @var{regexp}} or abbreviate the command name as
7909 @kindex reverse-search
7910 @item reverse-search @var{regexp}
7911 The command @samp{reverse-search @var{regexp}} checks each line, starting
7912 with the one before the last line listed and going backward, for a match
7913 for @var{regexp}. It lists the line that is found. You can abbreviate
7914 this command as @code{rev}.
7918 @section Specifying Source Directories
7921 @cindex directories for source files
7922 Executable programs sometimes do not record the directories of the source
7923 files from which they were compiled, just the names. Even when they do,
7924 the directories could be moved between the compilation and your debugging
7925 session. @value{GDBN} has a list of directories to search for source files;
7926 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
7927 it tries all the directories in the list, in the order they are present
7928 in the list, until it finds a file with the desired name.
7930 For example, suppose an executable references the file
7931 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
7932 @file{/mnt/cross}. The file is first looked up literally; if this
7933 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
7934 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
7935 message is printed. @value{GDBN} does not look up the parts of the
7936 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
7937 Likewise, the subdirectories of the source path are not searched: if
7938 the source path is @file{/mnt/cross}, and the binary refers to
7939 @file{foo.c}, @value{GDBN} would not find it under
7940 @file{/mnt/cross/usr/src/foo-1.0/lib}.
7942 Plain file names, relative file names with leading directories, file
7943 names containing dots, etc.@: are all treated as described above; for
7944 instance, if the source path is @file{/mnt/cross}, and the source file
7945 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
7946 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
7947 that---@file{/mnt/cross/foo.c}.
7949 Note that the executable search path is @emph{not} used to locate the
7952 Whenever you reset or rearrange the source path, @value{GDBN} clears out
7953 any information it has cached about where source files are found and where
7954 each line is in the file.
7958 When you start @value{GDBN}, its source path includes only @samp{cdir}
7959 and @samp{cwd}, in that order.
7960 To add other directories, use the @code{directory} command.
7962 The search path is used to find both program source files and @value{GDBN}
7963 script files (read using the @samp{-command} option and @samp{source} command).
7965 In addition to the source path, @value{GDBN} provides a set of commands
7966 that manage a list of source path substitution rules. A @dfn{substitution
7967 rule} specifies how to rewrite source directories stored in the program's
7968 debug information in case the sources were moved to a different
7969 directory between compilation and debugging. A rule is made of
7970 two strings, the first specifying what needs to be rewritten in
7971 the path, and the second specifying how it should be rewritten.
7972 In @ref{set substitute-path}, we name these two parts @var{from} and
7973 @var{to} respectively. @value{GDBN} does a simple string replacement
7974 of @var{from} with @var{to} at the start of the directory part of the
7975 source file name, and uses that result instead of the original file
7976 name to look up the sources.
7978 Using the previous example, suppose the @file{foo-1.0} tree has been
7979 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
7980 @value{GDBN} to replace @file{/usr/src} in all source path names with
7981 @file{/mnt/cross}. The first lookup will then be
7982 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
7983 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
7984 substitution rule, use the @code{set substitute-path} command
7985 (@pxref{set substitute-path}).
7987 To avoid unexpected substitution results, a rule is applied only if the
7988 @var{from} part of the directory name ends at a directory separator.
7989 For instance, a rule substituting @file{/usr/source} into
7990 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
7991 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
7992 is applied only at the beginning of the directory name, this rule will
7993 not be applied to @file{/root/usr/source/baz.c} either.
7995 In many cases, you can achieve the same result using the @code{directory}
7996 command. However, @code{set substitute-path} can be more efficient in
7997 the case where the sources are organized in a complex tree with multiple
7998 subdirectories. With the @code{directory} command, you need to add each
7999 subdirectory of your project. If you moved the entire tree while
8000 preserving its internal organization, then @code{set substitute-path}
8001 allows you to direct the debugger to all the sources with one single
8004 @code{set substitute-path} is also more than just a shortcut command.
8005 The source path is only used if the file at the original location no
8006 longer exists. On the other hand, @code{set substitute-path} modifies
8007 the debugger behavior to look at the rewritten location instead. So, if
8008 for any reason a source file that is not relevant to your executable is
8009 located at the original location, a substitution rule is the only
8010 method available to point @value{GDBN} at the new location.
8012 @cindex @samp{--with-relocated-sources}
8013 @cindex default source path substitution
8014 You can configure a default source path substitution rule by
8015 configuring @value{GDBN} with the
8016 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
8017 should be the name of a directory under @value{GDBN}'s configured
8018 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
8019 directory names in debug information under @var{dir} will be adjusted
8020 automatically if the installed @value{GDBN} is moved to a new
8021 location. This is useful if @value{GDBN}, libraries or executables
8022 with debug information and corresponding source code are being moved
8026 @item directory @var{dirname} @dots{}
8027 @item dir @var{dirname} @dots{}
8028 Add directory @var{dirname} to the front of the source path. Several
8029 directory names may be given to this command, separated by @samp{:}
8030 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
8031 part of absolute file names) or
8032 whitespace. You may specify a directory that is already in the source
8033 path; this moves it forward, so @value{GDBN} searches it sooner.
8037 @vindex $cdir@r{, convenience variable}
8038 @vindex $cwd@r{, convenience variable}
8039 @cindex compilation directory
8040 @cindex current directory
8041 @cindex working directory
8042 @cindex directory, current
8043 @cindex directory, compilation
8044 You can use the string @samp{$cdir} to refer to the compilation
8045 directory (if one is recorded), and @samp{$cwd} to refer to the current
8046 working directory. @samp{$cwd} is not the same as @samp{.}---the former
8047 tracks the current working directory as it changes during your @value{GDBN}
8048 session, while the latter is immediately expanded to the current
8049 directory at the time you add an entry to the source path.
8052 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
8054 @c RET-repeat for @code{directory} is explicitly disabled, but since
8055 @c repeating it would be a no-op we do not say that. (thanks to RMS)
8057 @item set directories @var{path-list}
8058 @kindex set directories
8059 Set the source path to @var{path-list}.
8060 @samp{$cdir:$cwd} are added if missing.
8062 @item show directories
8063 @kindex show directories
8064 Print the source path: show which directories it contains.
8066 @anchor{set substitute-path}
8067 @item set substitute-path @var{from} @var{to}
8068 @kindex set substitute-path
8069 Define a source path substitution rule, and add it at the end of the
8070 current list of existing substitution rules. If a rule with the same
8071 @var{from} was already defined, then the old rule is also deleted.
8073 For example, if the file @file{/foo/bar/baz.c} was moved to
8074 @file{/mnt/cross/baz.c}, then the command
8077 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
8081 will tell @value{GDBN} to replace @samp{/foo/bar} with
8082 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
8083 @file{baz.c} even though it was moved.
8085 In the case when more than one substitution rule have been defined,
8086 the rules are evaluated one by one in the order where they have been
8087 defined. The first one matching, if any, is selected to perform
8090 For instance, if we had entered the following commands:
8093 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
8094 (@value{GDBP}) set substitute-path /usr/src /mnt/src
8098 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
8099 @file{/mnt/include/defs.h} by using the first rule. However, it would
8100 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
8101 @file{/mnt/src/lib/foo.c}.
8104 @item unset substitute-path [path]
8105 @kindex unset substitute-path
8106 If a path is specified, search the current list of substitution rules
8107 for a rule that would rewrite that path. Delete that rule if found.
8108 A warning is emitted by the debugger if no rule could be found.
8110 If no path is specified, then all substitution rules are deleted.
8112 @item show substitute-path [path]
8113 @kindex show substitute-path
8114 If a path is specified, then print the source path substitution rule
8115 which would rewrite that path, if any.
8117 If no path is specified, then print all existing source path substitution
8122 If your source path is cluttered with directories that are no longer of
8123 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
8124 versions of source. You can correct the situation as follows:
8128 Use @code{directory} with no argument to reset the source path to its default value.
8131 Use @code{directory} with suitable arguments to reinstall the
8132 directories you want in the source path. You can add all the
8133 directories in one command.
8137 @section Source and Machine Code
8138 @cindex source line and its code address
8140 You can use the command @code{info line} to map source lines to program
8141 addresses (and vice versa), and the command @code{disassemble} to display
8142 a range of addresses as machine instructions. You can use the command
8143 @code{set disassemble-next-line} to set whether to disassemble next
8144 source line when execution stops. When run under @sc{gnu} Emacs
8145 mode, the @code{info line} command causes the arrow to point to the
8146 line specified. Also, @code{info line} prints addresses in symbolic form as
8151 @item info line @var{location}
8152 Print the starting and ending addresses of the compiled code for
8153 source line @var{location}. You can specify source lines in any of
8154 the ways documented in @ref{Specify Location}.
8157 For example, we can use @code{info line} to discover the location of
8158 the object code for the first line of function
8159 @code{m4_changequote}:
8161 @c FIXME: I think this example should also show the addresses in
8162 @c symbolic form, as they usually would be displayed.
8164 (@value{GDBP}) info line m4_changequote
8165 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
8169 @cindex code address and its source line
8170 We can also inquire (using @code{*@var{addr}} as the form for
8171 @var{location}) what source line covers a particular address:
8173 (@value{GDBP}) info line *0x63ff
8174 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
8177 @cindex @code{$_} and @code{info line}
8178 @cindex @code{x} command, default address
8179 @kindex x@r{(examine), and} info line
8180 After @code{info line}, the default address for the @code{x} command
8181 is changed to the starting address of the line, so that @samp{x/i} is
8182 sufficient to begin examining the machine code (@pxref{Memory,
8183 ,Examining Memory}). Also, this address is saved as the value of the
8184 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
8189 @cindex assembly instructions
8190 @cindex instructions, assembly
8191 @cindex machine instructions
8192 @cindex listing machine instructions
8194 @itemx disassemble /m
8195 @itemx disassemble /s
8196 @itemx disassemble /r
8197 This specialized command dumps a range of memory as machine
8198 instructions. It can also print mixed source+disassembly by specifying
8199 the @code{/m} or @code{/s} modifier and print the raw instructions in hex
8200 as well as in symbolic form by specifying the @code{/r} modifier.
8201 The default memory range is the function surrounding the
8202 program counter of the selected frame. A single argument to this
8203 command is a program counter value; @value{GDBN} dumps the function
8204 surrounding this value. When two arguments are given, they should
8205 be separated by a comma, possibly surrounded by whitespace. The
8206 arguments specify a range of addresses to dump, in one of two forms:
8209 @item @var{start},@var{end}
8210 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
8211 @item @var{start},+@var{length}
8212 the addresses from @var{start} (inclusive) to
8213 @code{@var{start}+@var{length}} (exclusive).
8217 When 2 arguments are specified, the name of the function is also
8218 printed (since there could be several functions in the given range).
8220 The argument(s) can be any expression yielding a numeric value, such as
8221 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
8223 If the range of memory being disassembled contains current program counter,
8224 the instruction at that location is shown with a @code{=>} marker.
8227 The following example shows the disassembly of a range of addresses of
8228 HP PA-RISC 2.0 code:
8231 (@value{GDBP}) disas 0x32c4, 0x32e4
8232 Dump of assembler code from 0x32c4 to 0x32e4:
8233 0x32c4 <main+204>: addil 0,dp
8234 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
8235 0x32cc <main+212>: ldil 0x3000,r31
8236 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
8237 0x32d4 <main+220>: ldo 0(r31),rp
8238 0x32d8 <main+224>: addil -0x800,dp
8239 0x32dc <main+228>: ldo 0x588(r1),r26
8240 0x32e0 <main+232>: ldil 0x3000,r31
8241 End of assembler dump.
8244 Here is an example showing mixed source+assembly for Intel x86
8245 with @code{/m} or @code{/s}, when the program is stopped just after
8246 function prologue in a non-optimized function with no inline code.
8249 (@value{GDBP}) disas /m main
8250 Dump of assembler code for function main:
8252 0x08048330 <+0>: push %ebp
8253 0x08048331 <+1>: mov %esp,%ebp
8254 0x08048333 <+3>: sub $0x8,%esp
8255 0x08048336 <+6>: and $0xfffffff0,%esp
8256 0x08048339 <+9>: sub $0x10,%esp
8258 6 printf ("Hello.\n");
8259 => 0x0804833c <+12>: movl $0x8048440,(%esp)
8260 0x08048343 <+19>: call 0x8048284 <puts@@plt>
8264 0x08048348 <+24>: mov $0x0,%eax
8265 0x0804834d <+29>: leave
8266 0x0804834e <+30>: ret
8268 End of assembler dump.
8271 The @code{/m} option is deprecated as its output is not useful when
8272 there is either inlined code or re-ordered code.
8273 The @code{/s} option is the preferred choice.
8274 Here is an example for AMD x86-64 showing the difference between
8275 @code{/m} output and @code{/s} output.
8276 This example has one inline function defined in a header file,
8277 and the code is compiled with @samp{-O2} optimization.
8278 Note how the @code{/m} output is missing the disassembly of
8279 several instructions that are present in the @code{/s} output.
8309 (@value{GDBP}) disas /m main
8310 Dump of assembler code for function main:
8314 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
8315 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
8319 0x000000000040041d <+29>: xor %eax,%eax
8320 0x000000000040041f <+31>: retq
8321 0x0000000000400420 <+32>: add %eax,%eax
8322 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
8324 End of assembler dump.
8325 (@value{GDBP}) disas /s main
8326 Dump of assembler code for function main:
8330 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
8334 0x0000000000400406 <+6>: test %eax,%eax
8335 0x0000000000400408 <+8>: js 0x400420 <main+32>
8340 0x000000000040040a <+10>: lea 0xa(%rax),%edx
8341 0x000000000040040d <+13>: test %eax,%eax
8342 0x000000000040040f <+15>: mov $0x1,%eax
8343 0x0000000000400414 <+20>: cmovne %edx,%eax
8347 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
8351 0x000000000040041d <+29>: xor %eax,%eax
8352 0x000000000040041f <+31>: retq
8356 0x0000000000400420 <+32>: add %eax,%eax
8357 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
8358 End of assembler dump.
8361 Here is another example showing raw instructions in hex for AMD x86-64,
8364 (gdb) disas /r 0x400281,+10
8365 Dump of assembler code from 0x400281 to 0x40028b:
8366 0x0000000000400281: 38 36 cmp %dh,(%rsi)
8367 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
8368 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
8369 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
8370 End of assembler dump.
8373 Addresses cannot be specified as a location (@pxref{Specify Location}).
8374 So, for example, if you want to disassemble function @code{bar}
8375 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
8376 and not @samp{disassemble foo.c:bar}.
8378 Some architectures have more than one commonly-used set of instruction
8379 mnemonics or other syntax.
8381 For programs that were dynamically linked and use shared libraries,
8382 instructions that call functions or branch to locations in the shared
8383 libraries might show a seemingly bogus location---it's actually a
8384 location of the relocation table. On some architectures, @value{GDBN}
8385 might be able to resolve these to actual function names.
8388 @kindex set disassembly-flavor
8389 @cindex Intel disassembly flavor
8390 @cindex AT&T disassembly flavor
8391 @item set disassembly-flavor @var{instruction-set}
8392 Select the instruction set to use when disassembling the
8393 program via the @code{disassemble} or @code{x/i} commands.
8395 Currently this command is only defined for the Intel x86 family. You
8396 can set @var{instruction-set} to either @code{intel} or @code{att}.
8397 The default is @code{att}, the AT&T flavor used by default by Unix
8398 assemblers for x86-based targets.
8400 @kindex show disassembly-flavor
8401 @item show disassembly-flavor
8402 Show the current setting of the disassembly flavor.
8406 @kindex set disassemble-next-line
8407 @kindex show disassemble-next-line
8408 @item set disassemble-next-line
8409 @itemx show disassemble-next-line
8410 Control whether or not @value{GDBN} will disassemble the next source
8411 line or instruction when execution stops. If ON, @value{GDBN} will
8412 display disassembly of the next source line when execution of the
8413 program being debugged stops. This is @emph{in addition} to
8414 displaying the source line itself, which @value{GDBN} always does if
8415 possible. If the next source line cannot be displayed for some reason
8416 (e.g., if @value{GDBN} cannot find the source file, or there's no line
8417 info in the debug info), @value{GDBN} will display disassembly of the
8418 next @emph{instruction} instead of showing the next source line. If
8419 AUTO, @value{GDBN} will display disassembly of next instruction only
8420 if the source line cannot be displayed. This setting causes
8421 @value{GDBN} to display some feedback when you step through a function
8422 with no line info or whose source file is unavailable. The default is
8423 OFF, which means never display the disassembly of the next line or
8429 @chapter Examining Data
8431 @cindex printing data
8432 @cindex examining data
8435 The usual way to examine data in your program is with the @code{print}
8436 command (abbreviated @code{p}), or its synonym @code{inspect}. It
8437 evaluates and prints the value of an expression of the language your
8438 program is written in (@pxref{Languages, ,Using @value{GDBN} with
8439 Different Languages}). It may also print the expression using a
8440 Python-based pretty-printer (@pxref{Pretty Printing}).
8443 @item print @var{expr}
8444 @itemx print /@var{f} @var{expr}
8445 @var{expr} is an expression (in the source language). By default the
8446 value of @var{expr} is printed in a format appropriate to its data type;
8447 you can choose a different format by specifying @samp{/@var{f}}, where
8448 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
8452 @itemx print /@var{f}
8453 @cindex reprint the last value
8454 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
8455 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
8456 conveniently inspect the same value in an alternative format.
8459 A more low-level way of examining data is with the @code{x} command.
8460 It examines data in memory at a specified address and prints it in a
8461 specified format. @xref{Memory, ,Examining Memory}.
8463 If you are interested in information about types, or about how the
8464 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
8465 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
8468 @cindex exploring hierarchical data structures
8470 Another way of examining values of expressions and type information is
8471 through the Python extension command @code{explore} (available only if
8472 the @value{GDBN} build is configured with @code{--with-python}). It
8473 offers an interactive way to start at the highest level (or, the most
8474 abstract level) of the data type of an expression (or, the data type
8475 itself) and explore all the way down to leaf scalar values/fields
8476 embedded in the higher level data types.
8479 @item explore @var{arg}
8480 @var{arg} is either an expression (in the source language), or a type
8481 visible in the current context of the program being debugged.
8484 The working of the @code{explore} command can be illustrated with an
8485 example. If a data type @code{struct ComplexStruct} is defined in your
8495 struct ComplexStruct
8497 struct SimpleStruct *ss_p;
8503 followed by variable declarations as
8506 struct SimpleStruct ss = @{ 10, 1.11 @};
8507 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
8511 then, the value of the variable @code{cs} can be explored using the
8512 @code{explore} command as follows.
8516 The value of `cs' is a struct/class of type `struct ComplexStruct' with
8517 the following fields:
8519 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
8520 arr = <Enter 1 to explore this field of type `int [10]'>
8522 Enter the field number of choice:
8526 Since the fields of @code{cs} are not scalar values, you are being
8527 prompted to chose the field you want to explore. Let's say you choose
8528 the field @code{ss_p} by entering @code{0}. Then, since this field is a
8529 pointer, you will be asked if it is pointing to a single value. From
8530 the declaration of @code{cs} above, it is indeed pointing to a single
8531 value, hence you enter @code{y}. If you enter @code{n}, then you will
8532 be asked if it were pointing to an array of values, in which case this
8533 field will be explored as if it were an array.
8536 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
8537 Continue exploring it as a pointer to a single value [y/n]: y
8538 The value of `*(cs.ss_p)' is a struct/class of type `struct
8539 SimpleStruct' with the following fields:
8541 i = 10 .. (Value of type `int')
8542 d = 1.1100000000000001 .. (Value of type `double')
8544 Press enter to return to parent value:
8548 If the field @code{arr} of @code{cs} was chosen for exploration by
8549 entering @code{1} earlier, then since it is as array, you will be
8550 prompted to enter the index of the element in the array that you want
8554 `cs.arr' is an array of `int'.
8555 Enter the index of the element you want to explore in `cs.arr': 5
8557 `(cs.arr)[5]' is a scalar value of type `int'.
8561 Press enter to return to parent value:
8564 In general, at any stage of exploration, you can go deeper towards the
8565 leaf values by responding to the prompts appropriately, or hit the
8566 return key to return to the enclosing data structure (the @i{higher}
8567 level data structure).
8569 Similar to exploring values, you can use the @code{explore} command to
8570 explore types. Instead of specifying a value (which is typically a
8571 variable name or an expression valid in the current context of the
8572 program being debugged), you specify a type name. If you consider the
8573 same example as above, your can explore the type
8574 @code{struct ComplexStruct} by passing the argument
8575 @code{struct ComplexStruct} to the @code{explore} command.
8578 (gdb) explore struct ComplexStruct
8582 By responding to the prompts appropriately in the subsequent interactive
8583 session, you can explore the type @code{struct ComplexStruct} in a
8584 manner similar to how the value @code{cs} was explored in the above
8587 The @code{explore} command also has two sub-commands,
8588 @code{explore value} and @code{explore type}. The former sub-command is
8589 a way to explicitly specify that value exploration of the argument is
8590 being invoked, while the latter is a way to explicitly specify that type
8591 exploration of the argument is being invoked.
8594 @item explore value @var{expr}
8595 @cindex explore value
8596 This sub-command of @code{explore} explores the value of the
8597 expression @var{expr} (if @var{expr} is an expression valid in the
8598 current context of the program being debugged). The behavior of this
8599 command is identical to that of the behavior of the @code{explore}
8600 command being passed the argument @var{expr}.
8602 @item explore type @var{arg}
8603 @cindex explore type
8604 This sub-command of @code{explore} explores the type of @var{arg} (if
8605 @var{arg} is a type visible in the current context of program being
8606 debugged), or the type of the value/expression @var{arg} (if @var{arg}
8607 is an expression valid in the current context of the program being
8608 debugged). If @var{arg} is a type, then the behavior of this command is
8609 identical to that of the @code{explore} command being passed the
8610 argument @var{arg}. If @var{arg} is an expression, then the behavior of
8611 this command will be identical to that of the @code{explore} command
8612 being passed the type of @var{arg} as the argument.
8616 * Expressions:: Expressions
8617 * Ambiguous Expressions:: Ambiguous Expressions
8618 * Variables:: Program variables
8619 * Arrays:: Artificial arrays
8620 * Output Formats:: Output formats
8621 * Memory:: Examining memory
8622 * Auto Display:: Automatic display
8623 * Print Settings:: Print settings
8624 * Pretty Printing:: Python pretty printing
8625 * Value History:: Value history
8626 * Convenience Vars:: Convenience variables
8627 * Convenience Funs:: Convenience functions
8628 * Registers:: Registers
8629 * Floating Point Hardware:: Floating point hardware
8630 * Vector Unit:: Vector Unit
8631 * OS Information:: Auxiliary data provided by operating system
8632 * Memory Region Attributes:: Memory region attributes
8633 * Dump/Restore Files:: Copy between memory and a file
8634 * Core File Generation:: Cause a program dump its core
8635 * Character Sets:: Debugging programs that use a different
8636 character set than GDB does
8637 * Caching Target Data:: Data caching for targets
8638 * Searching Memory:: Searching memory for a sequence of bytes
8642 @section Expressions
8645 @code{print} and many other @value{GDBN} commands accept an expression and
8646 compute its value. Any kind of constant, variable or operator defined
8647 by the programming language you are using is valid in an expression in
8648 @value{GDBN}. This includes conditional expressions, function calls,
8649 casts, and string constants. It also includes preprocessor macros, if
8650 you compiled your program to include this information; see
8653 @cindex arrays in expressions
8654 @value{GDBN} supports array constants in expressions input by
8655 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
8656 you can use the command @code{print @{1, 2, 3@}} to create an array
8657 of three integers. If you pass an array to a function or assign it
8658 to a program variable, @value{GDBN} copies the array to memory that
8659 is @code{malloc}ed in the target program.
8661 Because C is so widespread, most of the expressions shown in examples in
8662 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
8663 Languages}, for information on how to use expressions in other
8666 In this section, we discuss operators that you can use in @value{GDBN}
8667 expressions regardless of your programming language.
8669 @cindex casts, in expressions
8670 Casts are supported in all languages, not just in C, because it is so
8671 useful to cast a number into a pointer in order to examine a structure
8672 at that address in memory.
8673 @c FIXME: casts supported---Mod2 true?
8675 @value{GDBN} supports these operators, in addition to those common
8676 to programming languages:
8680 @samp{@@} is a binary operator for treating parts of memory as arrays.
8681 @xref{Arrays, ,Artificial Arrays}, for more information.
8684 @samp{::} allows you to specify a variable in terms of the file or
8685 function where it is defined. @xref{Variables, ,Program Variables}.
8687 @cindex @{@var{type}@}
8688 @cindex type casting memory
8689 @cindex memory, viewing as typed object
8690 @cindex casts, to view memory
8691 @item @{@var{type}@} @var{addr}
8692 Refers to an object of type @var{type} stored at address @var{addr} in
8693 memory. The address @var{addr} may be any expression whose value is
8694 an integer or pointer (but parentheses are required around binary
8695 operators, just as in a cast). This construct is allowed regardless
8696 of what kind of data is normally supposed to reside at @var{addr}.
8699 @node Ambiguous Expressions
8700 @section Ambiguous Expressions
8701 @cindex ambiguous expressions
8703 Expressions can sometimes contain some ambiguous elements. For instance,
8704 some programming languages (notably Ada, C@t{++} and Objective-C) permit
8705 a single function name to be defined several times, for application in
8706 different contexts. This is called @dfn{overloading}. Another example
8707 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
8708 templates and is typically instantiated several times, resulting in
8709 the same function name being defined in different contexts.
8711 In some cases and depending on the language, it is possible to adjust
8712 the expression to remove the ambiguity. For instance in C@t{++}, you
8713 can specify the signature of the function you want to break on, as in
8714 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
8715 qualified name of your function often makes the expression unambiguous
8718 When an ambiguity that needs to be resolved is detected, the debugger
8719 has the capability to display a menu of numbered choices for each
8720 possibility, and then waits for the selection with the prompt @samp{>}.
8721 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
8722 aborts the current command. If the command in which the expression was
8723 used allows more than one choice to be selected, the next option in the
8724 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
8727 For example, the following session excerpt shows an attempt to set a
8728 breakpoint at the overloaded symbol @code{String::after}.
8729 We choose three particular definitions of that function name:
8731 @c FIXME! This is likely to change to show arg type lists, at least
8734 (@value{GDBP}) b String::after
8737 [2] file:String.cc; line number:867
8738 [3] file:String.cc; line number:860
8739 [4] file:String.cc; line number:875
8740 [5] file:String.cc; line number:853
8741 [6] file:String.cc; line number:846
8742 [7] file:String.cc; line number:735
8744 Breakpoint 1 at 0xb26c: file String.cc, line 867.
8745 Breakpoint 2 at 0xb344: file String.cc, line 875.
8746 Breakpoint 3 at 0xafcc: file String.cc, line 846.
8747 Multiple breakpoints were set.
8748 Use the "delete" command to delete unwanted
8755 @kindex set multiple-symbols
8756 @item set multiple-symbols @var{mode}
8757 @cindex multiple-symbols menu
8759 This option allows you to adjust the debugger behavior when an expression
8762 By default, @var{mode} is set to @code{all}. If the command with which
8763 the expression is used allows more than one choice, then @value{GDBN}
8764 automatically selects all possible choices. For instance, inserting
8765 a breakpoint on a function using an ambiguous name results in a breakpoint
8766 inserted on each possible match. However, if a unique choice must be made,
8767 then @value{GDBN} uses the menu to help you disambiguate the expression.
8768 For instance, printing the address of an overloaded function will result
8769 in the use of the menu.
8771 When @var{mode} is set to @code{ask}, the debugger always uses the menu
8772 when an ambiguity is detected.
8774 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
8775 an error due to the ambiguity and the command is aborted.
8777 @kindex show multiple-symbols
8778 @item show multiple-symbols
8779 Show the current value of the @code{multiple-symbols} setting.
8783 @section Program Variables
8785 The most common kind of expression to use is the name of a variable
8788 Variables in expressions are understood in the selected stack frame
8789 (@pxref{Selection, ,Selecting a Frame}); they must be either:
8793 global (or file-static)
8800 visible according to the scope rules of the
8801 programming language from the point of execution in that frame
8804 @noindent This means that in the function
8819 you can examine and use the variable @code{a} whenever your program is
8820 executing within the function @code{foo}, but you can only use or
8821 examine the variable @code{b} while your program is executing inside
8822 the block where @code{b} is declared.
8824 @cindex variable name conflict
8825 There is an exception: you can refer to a variable or function whose
8826 scope is a single source file even if the current execution point is not
8827 in this file. But it is possible to have more than one such variable or
8828 function with the same name (in different source files). If that
8829 happens, referring to that name has unpredictable effects. If you wish,
8830 you can specify a static variable in a particular function or file by
8831 using the colon-colon (@code{::}) notation:
8833 @cindex colon-colon, context for variables/functions
8835 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
8836 @cindex @code{::}, context for variables/functions
8839 @var{file}::@var{variable}
8840 @var{function}::@var{variable}
8844 Here @var{file} or @var{function} is the name of the context for the
8845 static @var{variable}. In the case of file names, you can use quotes to
8846 make sure @value{GDBN} parses the file name as a single word---for example,
8847 to print a global value of @code{x} defined in @file{f2.c}:
8850 (@value{GDBP}) p 'f2.c'::x
8853 The @code{::} notation is normally used for referring to
8854 static variables, since you typically disambiguate uses of local variables
8855 in functions by selecting the appropriate frame and using the
8856 simple name of the variable. However, you may also use this notation
8857 to refer to local variables in frames enclosing the selected frame:
8866 process (a); /* Stop here */
8877 For example, if there is a breakpoint at the commented line,
8878 here is what you might see
8879 when the program stops after executing the call @code{bar(0)}:
8884 (@value{GDBP}) p bar::a
8887 #2 0x080483d0 in foo (a=5) at foobar.c:12
8890 (@value{GDBP}) p bar::a
8894 @cindex C@t{++} scope resolution
8895 These uses of @samp{::} are very rarely in conflict with the very
8896 similar use of the same notation in C@t{++}. When they are in
8897 conflict, the C@t{++} meaning takes precedence; however, this can be
8898 overridden by quoting the file or function name with single quotes.
8900 For example, suppose the program is stopped in a method of a class
8901 that has a field named @code{includefile}, and there is also an
8902 include file named @file{includefile} that defines a variable,
8906 (@value{GDBP}) p includefile
8908 (@value{GDBP}) p includefile::some_global
8909 A syntax error in expression, near `'.
8910 (@value{GDBP}) p 'includefile'::some_global
8914 @cindex wrong values
8915 @cindex variable values, wrong
8916 @cindex function entry/exit, wrong values of variables
8917 @cindex optimized code, wrong values of variables
8919 @emph{Warning:} Occasionally, a local variable may appear to have the
8920 wrong value at certain points in a function---just after entry to a new
8921 scope, and just before exit.
8923 You may see this problem when you are stepping by machine instructions.
8924 This is because, on most machines, it takes more than one instruction to
8925 set up a stack frame (including local variable definitions); if you are
8926 stepping by machine instructions, variables may appear to have the wrong
8927 values until the stack frame is completely built. On exit, it usually
8928 also takes more than one machine instruction to destroy a stack frame;
8929 after you begin stepping through that group of instructions, local
8930 variable definitions may be gone.
8932 This may also happen when the compiler does significant optimizations.
8933 To be sure of always seeing accurate values, turn off all optimization
8936 @cindex ``No symbol "foo" in current context''
8937 Another possible effect of compiler optimizations is to optimize
8938 unused variables out of existence, or assign variables to registers (as
8939 opposed to memory addresses). Depending on the support for such cases
8940 offered by the debug info format used by the compiler, @value{GDBN}
8941 might not be able to display values for such local variables. If that
8942 happens, @value{GDBN} will print a message like this:
8945 No symbol "foo" in current context.
8948 To solve such problems, either recompile without optimizations, or use a
8949 different debug info format, if the compiler supports several such
8950 formats. @xref{Compilation}, for more information on choosing compiler
8951 options. @xref{C, ,C and C@t{++}}, for more information about debug
8952 info formats that are best suited to C@t{++} programs.
8954 If you ask to print an object whose contents are unknown to
8955 @value{GDBN}, e.g., because its data type is not completely specified
8956 by the debug information, @value{GDBN} will say @samp{<incomplete
8957 type>}. @xref{Symbols, incomplete type}, for more about this.
8959 If you append @kbd{@@entry} string to a function parameter name you get its
8960 value at the time the function got called. If the value is not available an
8961 error message is printed. Entry values are available only with some compilers.
8962 Entry values are normally also printed at the function parameter list according
8963 to @ref{set print entry-values}.
8966 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
8972 (gdb) print i@@entry
8976 Strings are identified as arrays of @code{char} values without specified
8977 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
8978 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
8979 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
8980 defines literal string type @code{"char"} as @code{char} without a sign.
8985 signed char var1[] = "A";
8988 You get during debugging
8993 $2 = @{65 'A', 0 '\0'@}
8997 @section Artificial Arrays
8999 @cindex artificial array
9001 @kindex @@@r{, referencing memory as an array}
9002 It is often useful to print out several successive objects of the
9003 same type in memory; a section of an array, or an array of
9004 dynamically determined size for which only a pointer exists in the
9007 You can do this by referring to a contiguous span of memory as an
9008 @dfn{artificial array}, using the binary operator @samp{@@}. The left
9009 operand of @samp{@@} should be the first element of the desired array
9010 and be an individual object. The right operand should be the desired length
9011 of the array. The result is an array value whose elements are all of
9012 the type of the left argument. The first element is actually the left
9013 argument; the second element comes from bytes of memory immediately
9014 following those that hold the first element, and so on. Here is an
9015 example. If a program says
9018 int *array = (int *) malloc (len * sizeof (int));
9022 you can print the contents of @code{array} with
9028 The left operand of @samp{@@} must reside in memory. Array values made
9029 with @samp{@@} in this way behave just like other arrays in terms of
9030 subscripting, and are coerced to pointers when used in expressions.
9031 Artificial arrays most often appear in expressions via the value history
9032 (@pxref{Value History, ,Value History}), after printing one out.
9034 Another way to create an artificial array is to use a cast.
9035 This re-interprets a value as if it were an array.
9036 The value need not be in memory:
9038 (@value{GDBP}) p/x (short[2])0x12345678
9039 $1 = @{0x1234, 0x5678@}
9042 As a convenience, if you leave the array length out (as in
9043 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
9044 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
9046 (@value{GDBP}) p/x (short[])0x12345678
9047 $2 = @{0x1234, 0x5678@}
9050 Sometimes the artificial array mechanism is not quite enough; in
9051 moderately complex data structures, the elements of interest may not
9052 actually be adjacent---for example, if you are interested in the values
9053 of pointers in an array. One useful work-around in this situation is
9054 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
9055 Variables}) as a counter in an expression that prints the first
9056 interesting value, and then repeat that expression via @key{RET}. For
9057 instance, suppose you have an array @code{dtab} of pointers to
9058 structures, and you are interested in the values of a field @code{fv}
9059 in each structure. Here is an example of what you might type:
9069 @node Output Formats
9070 @section Output Formats
9072 @cindex formatted output
9073 @cindex output formats
9074 By default, @value{GDBN} prints a value according to its data type. Sometimes
9075 this is not what you want. For example, you might want to print a number
9076 in hex, or a pointer in decimal. Or you might want to view data in memory
9077 at a certain address as a character string or as an instruction. To do
9078 these things, specify an @dfn{output format} when you print a value.
9080 The simplest use of output formats is to say how to print a value
9081 already computed. This is done by starting the arguments of the
9082 @code{print} command with a slash and a format letter. The format
9083 letters supported are:
9087 Regard the bits of the value as an integer, and print the integer in
9091 Print as integer in signed decimal.
9094 Print as integer in unsigned decimal.
9097 Print as integer in octal.
9100 Print as integer in binary. The letter @samp{t} stands for ``two''.
9101 @footnote{@samp{b} cannot be used because these format letters are also
9102 used with the @code{x} command, where @samp{b} stands for ``byte'';
9103 see @ref{Memory,,Examining Memory}.}
9106 @cindex unknown address, locating
9107 @cindex locate address
9108 Print as an address, both absolute in hexadecimal and as an offset from
9109 the nearest preceding symbol. You can use this format used to discover
9110 where (in what function) an unknown address is located:
9113 (@value{GDBP}) p/a 0x54320
9114 $3 = 0x54320 <_initialize_vx+396>
9118 The command @code{info symbol 0x54320} yields similar results.
9119 @xref{Symbols, info symbol}.
9122 Regard as an integer and print it as a character constant. This
9123 prints both the numerical value and its character representation. The
9124 character representation is replaced with the octal escape @samp{\nnn}
9125 for characters outside the 7-bit @sc{ascii} range.
9127 Without this format, @value{GDBN} displays @code{char},
9128 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
9129 constants. Single-byte members of vectors are displayed as integer
9133 Regard the bits of the value as a floating point number and print
9134 using typical floating point syntax.
9137 @cindex printing strings
9138 @cindex printing byte arrays
9139 Regard as a string, if possible. With this format, pointers to single-byte
9140 data are displayed as null-terminated strings and arrays of single-byte data
9141 are displayed as fixed-length strings. Other values are displayed in their
9144 Without this format, @value{GDBN} displays pointers to and arrays of
9145 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
9146 strings. Single-byte members of a vector are displayed as an integer
9150 Like @samp{x} formatting, the value is treated as an integer and
9151 printed as hexadecimal, but leading zeros are printed to pad the value
9152 to the size of the integer type.
9155 @cindex raw printing
9156 Print using the @samp{raw} formatting. By default, @value{GDBN} will
9157 use a Python-based pretty-printer, if one is available (@pxref{Pretty
9158 Printing}). This typically results in a higher-level display of the
9159 value's contents. The @samp{r} format bypasses any Python
9160 pretty-printer which might exist.
9163 For example, to print the program counter in hex (@pxref{Registers}), type
9170 Note that no space is required before the slash; this is because command
9171 names in @value{GDBN} cannot contain a slash.
9173 To reprint the last value in the value history with a different format,
9174 you can use the @code{print} command with just a format and no
9175 expression. For example, @samp{p/x} reprints the last value in hex.
9178 @section Examining Memory
9180 You can use the command @code{x} (for ``examine'') to examine memory in
9181 any of several formats, independently of your program's data types.
9183 @cindex examining memory
9185 @kindex x @r{(examine memory)}
9186 @item x/@var{nfu} @var{addr}
9189 Use the @code{x} command to examine memory.
9192 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
9193 much memory to display and how to format it; @var{addr} is an
9194 expression giving the address where you want to start displaying memory.
9195 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
9196 Several commands set convenient defaults for @var{addr}.
9199 @item @var{n}, the repeat count
9200 The repeat count is a decimal integer; the default is 1. It specifies
9201 how much memory (counting by units @var{u}) to display.
9202 @c This really is **decimal**; unaffected by 'set radix' as of GDB
9205 @item @var{f}, the display format
9206 The display format is one of the formats used by @code{print}
9207 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
9208 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
9209 The default is @samp{x} (hexadecimal) initially. The default changes
9210 each time you use either @code{x} or @code{print}.
9212 @item @var{u}, the unit size
9213 The unit size is any of
9219 Halfwords (two bytes).
9221 Words (four bytes). This is the initial default.
9223 Giant words (eight bytes).
9226 Each time you specify a unit size with @code{x}, that size becomes the
9227 default unit the next time you use @code{x}. For the @samp{i} format,
9228 the unit size is ignored and is normally not written. For the @samp{s} format,
9229 the unit size defaults to @samp{b}, unless it is explicitly given.
9230 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
9231 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
9232 Note that the results depend on the programming language of the
9233 current compilation unit. If the language is C, the @samp{s}
9234 modifier will use the UTF-16 encoding while @samp{w} will use
9235 UTF-32. The encoding is set by the programming language and cannot
9238 @item @var{addr}, starting display address
9239 @var{addr} is the address where you want @value{GDBN} to begin displaying
9240 memory. The expression need not have a pointer value (though it may);
9241 it is always interpreted as an integer address of a byte of memory.
9242 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
9243 @var{addr} is usually just after the last address examined---but several
9244 other commands also set the default address: @code{info breakpoints} (to
9245 the address of the last breakpoint listed), @code{info line} (to the
9246 starting address of a line), and @code{print} (if you use it to display
9247 a value from memory).
9250 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
9251 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
9252 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
9253 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
9254 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
9256 Since the letters indicating unit sizes are all distinct from the
9257 letters specifying output formats, you do not have to remember whether
9258 unit size or format comes first; either order works. The output
9259 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
9260 (However, the count @var{n} must come first; @samp{wx4} does not work.)
9262 Even though the unit size @var{u} is ignored for the formats @samp{s}
9263 and @samp{i}, you might still want to use a count @var{n}; for example,
9264 @samp{3i} specifies that you want to see three machine instructions,
9265 including any operands. For convenience, especially when used with
9266 the @code{display} command, the @samp{i} format also prints branch delay
9267 slot instructions, if any, beyond the count specified, which immediately
9268 follow the last instruction that is within the count. The command
9269 @code{disassemble} gives an alternative way of inspecting machine
9270 instructions; see @ref{Machine Code,,Source and Machine Code}.
9272 All the defaults for the arguments to @code{x} are designed to make it
9273 easy to continue scanning memory with minimal specifications each time
9274 you use @code{x}. For example, after you have inspected three machine
9275 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
9276 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
9277 the repeat count @var{n} is used again; the other arguments default as
9278 for successive uses of @code{x}.
9280 When examining machine instructions, the instruction at current program
9281 counter is shown with a @code{=>} marker. For example:
9284 (@value{GDBP}) x/5i $pc-6
9285 0x804837f <main+11>: mov %esp,%ebp
9286 0x8048381 <main+13>: push %ecx
9287 0x8048382 <main+14>: sub $0x4,%esp
9288 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
9289 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
9292 @cindex @code{$_}, @code{$__}, and value history
9293 The addresses and contents printed by the @code{x} command are not saved
9294 in the value history because there is often too much of them and they
9295 would get in the way. Instead, @value{GDBN} makes these values available for
9296 subsequent use in expressions as values of the convenience variables
9297 @code{$_} and @code{$__}. After an @code{x} command, the last address
9298 examined is available for use in expressions in the convenience variable
9299 @code{$_}. The contents of that address, as examined, are available in
9300 the convenience variable @code{$__}.
9302 If the @code{x} command has a repeat count, the address and contents saved
9303 are from the last memory unit printed; this is not the same as the last
9304 address printed if several units were printed on the last line of output.
9306 @anchor{addressable memory unit}
9307 @cindex addressable memory unit
9308 Most targets have an addressable memory unit size of 8 bits. This means
9309 that to each memory address are associated 8 bits of data. Some
9310 targets, however, have other addressable memory unit sizes.
9311 Within @value{GDBN} and this document, the term
9312 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
9313 when explicitly referring to a chunk of data of that size. The word
9314 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
9315 the addressable memory unit size of the target. For most systems,
9316 addressable memory unit is a synonym of byte.
9318 @cindex remote memory comparison
9319 @cindex target memory comparison
9320 @cindex verify remote memory image
9321 @cindex verify target memory image
9322 When you are debugging a program running on a remote target machine
9323 (@pxref{Remote Debugging}), you may wish to verify the program's image
9324 in the remote machine's memory against the executable file you
9325 downloaded to the target. Or, on any target, you may want to check
9326 whether the program has corrupted its own read-only sections. The
9327 @code{compare-sections} command is provided for such situations.
9330 @kindex compare-sections
9331 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
9332 Compare the data of a loadable section @var{section-name} in the
9333 executable file of the program being debugged with the same section in
9334 the target machine's memory, and report any mismatches. With no
9335 arguments, compares all loadable sections. With an argument of
9336 @code{-r}, compares all loadable read-only sections.
9338 Note: for remote targets, this command can be accelerated if the
9339 target supports computing the CRC checksum of a block of memory
9340 (@pxref{qCRC packet}).
9344 @section Automatic Display
9345 @cindex automatic display
9346 @cindex display of expressions
9348 If you find that you want to print the value of an expression frequently
9349 (to see how it changes), you might want to add it to the @dfn{automatic
9350 display list} so that @value{GDBN} prints its value each time your program stops.
9351 Each expression added to the list is given a number to identify it;
9352 to remove an expression from the list, you specify that number.
9353 The automatic display looks like this:
9357 3: bar[5] = (struct hack *) 0x3804
9361 This display shows item numbers, expressions and their current values. As with
9362 displays you request manually using @code{x} or @code{print}, you can
9363 specify the output format you prefer; in fact, @code{display} decides
9364 whether to use @code{print} or @code{x} depending your format
9365 specification---it uses @code{x} if you specify either the @samp{i}
9366 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
9370 @item display @var{expr}
9371 Add the expression @var{expr} to the list of expressions to display
9372 each time your program stops. @xref{Expressions, ,Expressions}.
9374 @code{display} does not repeat if you press @key{RET} again after using it.
9376 @item display/@var{fmt} @var{expr}
9377 For @var{fmt} specifying only a display format and not a size or
9378 count, add the expression @var{expr} to the auto-display list but
9379 arrange to display it each time in the specified format @var{fmt}.
9380 @xref{Output Formats,,Output Formats}.
9382 @item display/@var{fmt} @var{addr}
9383 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
9384 number of units, add the expression @var{addr} as a memory address to
9385 be examined each time your program stops. Examining means in effect
9386 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
9389 For example, @samp{display/i $pc} can be helpful, to see the machine
9390 instruction about to be executed each time execution stops (@samp{$pc}
9391 is a common name for the program counter; @pxref{Registers, ,Registers}).
9394 @kindex delete display
9396 @item undisplay @var{dnums}@dots{}
9397 @itemx delete display @var{dnums}@dots{}
9398 Remove items from the list of expressions to display. Specify the
9399 numbers of the displays that you want affected with the command
9400 argument @var{dnums}. It can be a single display number, one of the
9401 numbers shown in the first field of the @samp{info display} display;
9402 or it could be a range of display numbers, as in @code{2-4}.
9404 @code{undisplay} does not repeat if you press @key{RET} after using it.
9405 (Otherwise you would just get the error @samp{No display number @dots{}}.)
9407 @kindex disable display
9408 @item disable display @var{dnums}@dots{}
9409 Disable the display of item numbers @var{dnums}. A disabled display
9410 item is not printed automatically, but is not forgotten. It may be
9411 enabled again later. Specify the numbers of the displays that you
9412 want affected with the command argument @var{dnums}. It can be a
9413 single display number, one of the numbers shown in the first field of
9414 the @samp{info display} display; or it could be a range of display
9415 numbers, as in @code{2-4}.
9417 @kindex enable display
9418 @item enable display @var{dnums}@dots{}
9419 Enable display of item numbers @var{dnums}. It becomes effective once
9420 again in auto display of its expression, until you specify otherwise.
9421 Specify the numbers of the displays that you want affected with the
9422 command argument @var{dnums}. It can be a single display number, one
9423 of the numbers shown in the first field of the @samp{info display}
9424 display; or it could be a range of display numbers, as in @code{2-4}.
9427 Display the current values of the expressions on the list, just as is
9428 done when your program stops.
9430 @kindex info display
9432 Print the list of expressions previously set up to display
9433 automatically, each one with its item number, but without showing the
9434 values. This includes disabled expressions, which are marked as such.
9435 It also includes expressions which would not be displayed right now
9436 because they refer to automatic variables not currently available.
9439 @cindex display disabled out of scope
9440 If a display expression refers to local variables, then it does not make
9441 sense outside the lexical context for which it was set up. Such an
9442 expression is disabled when execution enters a context where one of its
9443 variables is not defined. For example, if you give the command
9444 @code{display last_char} while inside a function with an argument
9445 @code{last_char}, @value{GDBN} displays this argument while your program
9446 continues to stop inside that function. When it stops elsewhere---where
9447 there is no variable @code{last_char}---the display is disabled
9448 automatically. The next time your program stops where @code{last_char}
9449 is meaningful, you can enable the display expression once again.
9451 @node Print Settings
9452 @section Print Settings
9454 @cindex format options
9455 @cindex print settings
9456 @value{GDBN} provides the following ways to control how arrays, structures,
9457 and symbols are printed.
9460 These settings are useful for debugging programs in any language:
9464 @item set print address
9465 @itemx set print address on
9466 @cindex print/don't print memory addresses
9467 @value{GDBN} prints memory addresses showing the location of stack
9468 traces, structure values, pointer values, breakpoints, and so forth,
9469 even when it also displays the contents of those addresses. The default
9470 is @code{on}. For example, this is what a stack frame display looks like with
9471 @code{set print address on}:
9476 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
9478 530 if (lquote != def_lquote)
9482 @item set print address off
9483 Do not print addresses when displaying their contents. For example,
9484 this is the same stack frame displayed with @code{set print address off}:
9488 (@value{GDBP}) set print addr off
9490 #0 set_quotes (lq="<<", rq=">>") at input.c:530
9491 530 if (lquote != def_lquote)
9495 You can use @samp{set print address off} to eliminate all machine
9496 dependent displays from the @value{GDBN} interface. For example, with
9497 @code{print address off}, you should get the same text for backtraces on
9498 all machines---whether or not they involve pointer arguments.
9501 @item show print address
9502 Show whether or not addresses are to be printed.
9505 When @value{GDBN} prints a symbolic address, it normally prints the
9506 closest earlier symbol plus an offset. If that symbol does not uniquely
9507 identify the address (for example, it is a name whose scope is a single
9508 source file), you may need to clarify. One way to do this is with
9509 @code{info line}, for example @samp{info line *0x4537}. Alternately,
9510 you can set @value{GDBN} to print the source file and line number when
9511 it prints a symbolic address:
9514 @item set print symbol-filename on
9515 @cindex source file and line of a symbol
9516 @cindex symbol, source file and line
9517 Tell @value{GDBN} to print the source file name and line number of a
9518 symbol in the symbolic form of an address.
9520 @item set print symbol-filename off
9521 Do not print source file name and line number of a symbol. This is the
9524 @item show print symbol-filename
9525 Show whether or not @value{GDBN} will print the source file name and
9526 line number of a symbol in the symbolic form of an address.
9529 Another situation where it is helpful to show symbol filenames and line
9530 numbers is when disassembling code; @value{GDBN} shows you the line
9531 number and source file that corresponds to each instruction.
9533 Also, you may wish to see the symbolic form only if the address being
9534 printed is reasonably close to the closest earlier symbol:
9537 @item set print max-symbolic-offset @var{max-offset}
9538 @itemx set print max-symbolic-offset unlimited
9539 @cindex maximum value for offset of closest symbol
9540 Tell @value{GDBN} to only display the symbolic form of an address if the
9541 offset between the closest earlier symbol and the address is less than
9542 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
9543 to always print the symbolic form of an address if any symbol precedes
9544 it. Zero is equivalent to @code{unlimited}.
9546 @item show print max-symbolic-offset
9547 Ask how large the maximum offset is that @value{GDBN} prints in a
9551 @cindex wild pointer, interpreting
9552 @cindex pointer, finding referent
9553 If you have a pointer and you are not sure where it points, try
9554 @samp{set print symbol-filename on}. Then you can determine the name
9555 and source file location of the variable where it points, using
9556 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
9557 For example, here @value{GDBN} shows that a variable @code{ptt} points
9558 at another variable @code{t}, defined in @file{hi2.c}:
9561 (@value{GDBP}) set print symbol-filename on
9562 (@value{GDBP}) p/a ptt
9563 $4 = 0xe008 <t in hi2.c>
9567 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
9568 does not show the symbol name and filename of the referent, even with
9569 the appropriate @code{set print} options turned on.
9572 You can also enable @samp{/a}-like formatting all the time using
9573 @samp{set print symbol on}:
9576 @item set print symbol on
9577 Tell @value{GDBN} to print the symbol corresponding to an address, if
9580 @item set print symbol off
9581 Tell @value{GDBN} not to print the symbol corresponding to an
9582 address. In this mode, @value{GDBN} will still print the symbol
9583 corresponding to pointers to functions. This is the default.
9585 @item show print symbol
9586 Show whether @value{GDBN} will display the symbol corresponding to an
9590 Other settings control how different kinds of objects are printed:
9593 @item set print array
9594 @itemx set print array on
9595 @cindex pretty print arrays
9596 Pretty print arrays. This format is more convenient to read,
9597 but uses more space. The default is off.
9599 @item set print array off
9600 Return to compressed format for arrays.
9602 @item show print array
9603 Show whether compressed or pretty format is selected for displaying
9606 @cindex print array indexes
9607 @item set print array-indexes
9608 @itemx set print array-indexes on
9609 Print the index of each element when displaying arrays. May be more
9610 convenient to locate a given element in the array or quickly find the
9611 index of a given element in that printed array. The default is off.
9613 @item set print array-indexes off
9614 Stop printing element indexes when displaying arrays.
9616 @item show print array-indexes
9617 Show whether the index of each element is printed when displaying
9620 @item set print elements @var{number-of-elements}
9621 @itemx set print elements unlimited
9622 @cindex number of array elements to print
9623 @cindex limit on number of printed array elements
9624 Set a limit on how many elements of an array @value{GDBN} will print.
9625 If @value{GDBN} is printing a large array, it stops printing after it has
9626 printed the number of elements set by the @code{set print elements} command.
9627 This limit also applies to the display of strings.
9628 When @value{GDBN} starts, this limit is set to 200.
9629 Setting @var{number-of-elements} to @code{unlimited} or zero means
9630 that the number of elements to print is unlimited.
9632 @item show print elements
9633 Display the number of elements of a large array that @value{GDBN} will print.
9634 If the number is 0, then the printing is unlimited.
9636 @item set print frame-arguments @var{value}
9637 @kindex set print frame-arguments
9638 @cindex printing frame argument values
9639 @cindex print all frame argument values
9640 @cindex print frame argument values for scalars only
9641 @cindex do not print frame argument values
9642 This command allows to control how the values of arguments are printed
9643 when the debugger prints a frame (@pxref{Frames}). The possible
9648 The values of all arguments are printed.
9651 Print the value of an argument only if it is a scalar. The value of more
9652 complex arguments such as arrays, structures, unions, etc, is replaced
9653 by @code{@dots{}}. This is the default. Here is an example where
9654 only scalar arguments are shown:
9657 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
9662 None of the argument values are printed. Instead, the value of each argument
9663 is replaced by @code{@dots{}}. In this case, the example above now becomes:
9666 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
9671 By default, only scalar arguments are printed. This command can be used
9672 to configure the debugger to print the value of all arguments, regardless
9673 of their type. However, it is often advantageous to not print the value
9674 of more complex parameters. For instance, it reduces the amount of
9675 information printed in each frame, making the backtrace more readable.
9676 Also, it improves performance when displaying Ada frames, because
9677 the computation of large arguments can sometimes be CPU-intensive,
9678 especially in large applications. Setting @code{print frame-arguments}
9679 to @code{scalars} (the default) or @code{none} avoids this computation,
9680 thus speeding up the display of each Ada frame.
9682 @item show print frame-arguments
9683 Show how the value of arguments should be displayed when printing a frame.
9685 @item set print raw frame-arguments on
9686 Print frame arguments in raw, non pretty-printed, form.
9688 @item set print raw frame-arguments off
9689 Print frame arguments in pretty-printed form, if there is a pretty-printer
9690 for the value (@pxref{Pretty Printing}),
9691 otherwise print the value in raw form.
9692 This is the default.
9694 @item show print raw frame-arguments
9695 Show whether to print frame arguments in raw form.
9697 @anchor{set print entry-values}
9698 @item set print entry-values @var{value}
9699 @kindex set print entry-values
9700 Set printing of frame argument values at function entry. In some cases
9701 @value{GDBN} can determine the value of function argument which was passed by
9702 the function caller, even if the value was modified inside the called function
9703 and therefore is different. With optimized code, the current value could be
9704 unavailable, but the entry value may still be known.
9706 The default value is @code{default} (see below for its description). Older
9707 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
9708 this feature will behave in the @code{default} setting the same way as with the
9711 This functionality is currently supported only by DWARF 2 debugging format and
9712 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
9713 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
9716 The @var{value} parameter can be one of the following:
9720 Print only actual parameter values, never print values from function entry
9724 #0 different (val=6)
9725 #0 lost (val=<optimized out>)
9727 #0 invalid (val=<optimized out>)
9731 Print only parameter values from function entry point. The actual parameter
9732 values are never printed.
9734 #0 equal (val@@entry=5)
9735 #0 different (val@@entry=5)
9736 #0 lost (val@@entry=5)
9737 #0 born (val@@entry=<optimized out>)
9738 #0 invalid (val@@entry=<optimized out>)
9742 Print only parameter values from function entry point. If value from function
9743 entry point is not known while the actual value is known, print the actual
9744 value for such parameter.
9746 #0 equal (val@@entry=5)
9747 #0 different (val@@entry=5)
9748 #0 lost (val@@entry=5)
9750 #0 invalid (val@@entry=<optimized out>)
9754 Print actual parameter values. If actual parameter value is not known while
9755 value from function entry point is known, print the entry point value for such
9759 #0 different (val=6)
9760 #0 lost (val@@entry=5)
9762 #0 invalid (val=<optimized out>)
9766 Always print both the actual parameter value and its value from function entry
9767 point, even if values of one or both are not available due to compiler
9770 #0 equal (val=5, val@@entry=5)
9771 #0 different (val=6, val@@entry=5)
9772 #0 lost (val=<optimized out>, val@@entry=5)
9773 #0 born (val=10, val@@entry=<optimized out>)
9774 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
9778 Print the actual parameter value if it is known and also its value from
9779 function entry point if it is known. If neither is known, print for the actual
9780 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
9781 values are known and identical, print the shortened
9782 @code{param=param@@entry=VALUE} notation.
9784 #0 equal (val=val@@entry=5)
9785 #0 different (val=6, val@@entry=5)
9786 #0 lost (val@@entry=5)
9788 #0 invalid (val=<optimized out>)
9792 Always print the actual parameter value. Print also its value from function
9793 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
9794 if both values are known and identical, print the shortened
9795 @code{param=param@@entry=VALUE} notation.
9797 #0 equal (val=val@@entry=5)
9798 #0 different (val=6, val@@entry=5)
9799 #0 lost (val=<optimized out>, val@@entry=5)
9801 #0 invalid (val=<optimized out>)
9805 For analysis messages on possible failures of frame argument values at function
9806 entry resolution see @ref{set debug entry-values}.
9808 @item show print entry-values
9809 Show the method being used for printing of frame argument values at function
9812 @item set print repeats @var{number-of-repeats}
9813 @itemx set print repeats unlimited
9814 @cindex repeated array elements
9815 Set the threshold for suppressing display of repeated array
9816 elements. When the number of consecutive identical elements of an
9817 array exceeds the threshold, @value{GDBN} prints the string
9818 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
9819 identical repetitions, instead of displaying the identical elements
9820 themselves. Setting the threshold to @code{unlimited} or zero will
9821 cause all elements to be individually printed. The default threshold
9824 @item show print repeats
9825 Display the current threshold for printing repeated identical
9828 @item set print null-stop
9829 @cindex @sc{null} elements in arrays
9830 Cause @value{GDBN} to stop printing the characters of an array when the first
9831 @sc{null} is encountered. This is useful when large arrays actually
9832 contain only short strings.
9835 @item show print null-stop
9836 Show whether @value{GDBN} stops printing an array on the first
9837 @sc{null} character.
9839 @item set print pretty on
9840 @cindex print structures in indented form
9841 @cindex indentation in structure display
9842 Cause @value{GDBN} to print structures in an indented format with one member
9843 per line, like this:
9858 @item set print pretty off
9859 Cause @value{GDBN} to print structures in a compact format, like this:
9863 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
9864 meat = 0x54 "Pork"@}
9869 This is the default format.
9871 @item show print pretty
9872 Show which format @value{GDBN} is using to print structures.
9874 @item set print sevenbit-strings on
9875 @cindex eight-bit characters in strings
9876 @cindex octal escapes in strings
9877 Print using only seven-bit characters; if this option is set,
9878 @value{GDBN} displays any eight-bit characters (in strings or
9879 character values) using the notation @code{\}@var{nnn}. This setting is
9880 best if you are working in English (@sc{ascii}) and you use the
9881 high-order bit of characters as a marker or ``meta'' bit.
9883 @item set print sevenbit-strings off
9884 Print full eight-bit characters. This allows the use of more
9885 international character sets, and is the default.
9887 @item show print sevenbit-strings
9888 Show whether or not @value{GDBN} is printing only seven-bit characters.
9890 @item set print union on
9891 @cindex unions in structures, printing
9892 Tell @value{GDBN} to print unions which are contained in structures
9893 and other unions. This is the default setting.
9895 @item set print union off
9896 Tell @value{GDBN} not to print unions which are contained in
9897 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
9900 @item show print union
9901 Ask @value{GDBN} whether or not it will print unions which are contained in
9902 structures and other unions.
9904 For example, given the declarations
9907 typedef enum @{Tree, Bug@} Species;
9908 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
9909 typedef enum @{Caterpillar, Cocoon, Butterfly@}
9920 struct thing foo = @{Tree, @{Acorn@}@};
9924 with @code{set print union on} in effect @samp{p foo} would print
9927 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
9931 and with @code{set print union off} in effect it would print
9934 $1 = @{it = Tree, form = @{...@}@}
9938 @code{set print union} affects programs written in C-like languages
9944 These settings are of interest when debugging C@t{++} programs:
9947 @cindex demangling C@t{++} names
9948 @item set print demangle
9949 @itemx set print demangle on
9950 Print C@t{++} names in their source form rather than in the encoded
9951 (``mangled'') form passed to the assembler and linker for type-safe
9952 linkage. The default is on.
9954 @item show print demangle
9955 Show whether C@t{++} names are printed in mangled or demangled form.
9957 @item set print asm-demangle
9958 @itemx set print asm-demangle on
9959 Print C@t{++} names in their source form rather than their mangled form, even
9960 in assembler code printouts such as instruction disassemblies.
9963 @item show print asm-demangle
9964 Show whether C@t{++} names in assembly listings are printed in mangled
9967 @cindex C@t{++} symbol decoding style
9968 @cindex symbol decoding style, C@t{++}
9969 @kindex set demangle-style
9970 @item set demangle-style @var{style}
9971 Choose among several encoding schemes used by different compilers to
9972 represent C@t{++} names. The choices for @var{style} are currently:
9976 Allow @value{GDBN} to choose a decoding style by inspecting your program.
9977 This is the default.
9980 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
9983 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
9986 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
9989 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
9990 @strong{Warning:} this setting alone is not sufficient to allow
9991 debugging @code{cfront}-generated executables. @value{GDBN} would
9992 require further enhancement to permit that.
9995 If you omit @var{style}, you will see a list of possible formats.
9997 @item show demangle-style
9998 Display the encoding style currently in use for decoding C@t{++} symbols.
10000 @item set print object
10001 @itemx set print object on
10002 @cindex derived type of an object, printing
10003 @cindex display derived types
10004 When displaying a pointer to an object, identify the @emph{actual}
10005 (derived) type of the object rather than the @emph{declared} type, using
10006 the virtual function table. Note that the virtual function table is
10007 required---this feature can only work for objects that have run-time
10008 type identification; a single virtual method in the object's declared
10009 type is sufficient. Note that this setting is also taken into account when
10010 working with variable objects via MI (@pxref{GDB/MI}).
10012 @item set print object off
10013 Display only the declared type of objects, without reference to the
10014 virtual function table. This is the default setting.
10016 @item show print object
10017 Show whether actual, or declared, object types are displayed.
10019 @item set print static-members
10020 @itemx set print static-members on
10021 @cindex static members of C@t{++} objects
10022 Print static members when displaying a C@t{++} object. The default is on.
10024 @item set print static-members off
10025 Do not print static members when displaying a C@t{++} object.
10027 @item show print static-members
10028 Show whether C@t{++} static members are printed or not.
10030 @item set print pascal_static-members
10031 @itemx set print pascal_static-members on
10032 @cindex static members of Pascal objects
10033 @cindex Pascal objects, static members display
10034 Print static members when displaying a Pascal object. The default is on.
10036 @item set print pascal_static-members off
10037 Do not print static members when displaying a Pascal object.
10039 @item show print pascal_static-members
10040 Show whether Pascal static members are printed or not.
10042 @c These don't work with HP ANSI C++ yet.
10043 @item set print vtbl
10044 @itemx set print vtbl on
10045 @cindex pretty print C@t{++} virtual function tables
10046 @cindex virtual functions (C@t{++}) display
10047 @cindex VTBL display
10048 Pretty print C@t{++} virtual function tables. The default is off.
10049 (The @code{vtbl} commands do not work on programs compiled with the HP
10050 ANSI C@t{++} compiler (@code{aCC}).)
10052 @item set print vtbl off
10053 Do not pretty print C@t{++} virtual function tables.
10055 @item show print vtbl
10056 Show whether C@t{++} virtual function tables are pretty printed, or not.
10059 @node Pretty Printing
10060 @section Pretty Printing
10062 @value{GDBN} provides a mechanism to allow pretty-printing of values using
10063 Python code. It greatly simplifies the display of complex objects. This
10064 mechanism works for both MI and the CLI.
10067 * Pretty-Printer Introduction:: Introduction to pretty-printers
10068 * Pretty-Printer Example:: An example pretty-printer
10069 * Pretty-Printer Commands:: Pretty-printer commands
10072 @node Pretty-Printer Introduction
10073 @subsection Pretty-Printer Introduction
10075 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
10076 registered for the value. If there is then @value{GDBN} invokes the
10077 pretty-printer to print the value. Otherwise the value is printed normally.
10079 Pretty-printers are normally named. This makes them easy to manage.
10080 The @samp{info pretty-printer} command will list all the installed
10081 pretty-printers with their names.
10082 If a pretty-printer can handle multiple data types, then its
10083 @dfn{subprinters} are the printers for the individual data types.
10084 Each such subprinter has its own name.
10085 The format of the name is @var{printer-name};@var{subprinter-name}.
10087 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
10088 Typically they are automatically loaded and registered when the corresponding
10089 debug information is loaded, thus making them available without having to
10090 do anything special.
10092 There are three places where a pretty-printer can be registered.
10096 Pretty-printers registered globally are available when debugging
10100 Pretty-printers registered with a program space are available only
10101 when debugging that program.
10102 @xref{Progspaces In Python}, for more details on program spaces in Python.
10105 Pretty-printers registered with an objfile are loaded and unloaded
10106 with the corresponding objfile (e.g., shared library).
10107 @xref{Objfiles In Python}, for more details on objfiles in Python.
10110 @xref{Selecting Pretty-Printers}, for further information on how
10111 pretty-printers are selected,
10113 @xref{Writing a Pretty-Printer}, for implementing pretty printers
10116 @node Pretty-Printer Example
10117 @subsection Pretty-Printer Example
10119 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
10122 (@value{GDBP}) print s
10124 static npos = 4294967295,
10126 <std::allocator<char>> = @{
10127 <__gnu_cxx::new_allocator<char>> = @{
10128 <No data fields>@}, <No data fields>
10130 members of std::basic_string<char, std::char_traits<char>,
10131 std::allocator<char> >::_Alloc_hider:
10132 _M_p = 0x804a014 "abcd"
10137 With a pretty-printer for @code{std::string} only the contents are printed:
10140 (@value{GDBP}) print s
10144 @node Pretty-Printer Commands
10145 @subsection Pretty-Printer Commands
10146 @cindex pretty-printer commands
10149 @kindex info pretty-printer
10150 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
10151 Print the list of installed pretty-printers.
10152 This includes disabled pretty-printers, which are marked as such.
10154 @var{object-regexp} is a regular expression matching the objects
10155 whose pretty-printers to list.
10156 Objects can be @code{global}, the program space's file
10157 (@pxref{Progspaces In Python}),
10158 and the object files within that program space (@pxref{Objfiles In Python}).
10159 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
10160 looks up a printer from these three objects.
10162 @var{name-regexp} is a regular expression matching the name of the printers
10165 @kindex disable pretty-printer
10166 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
10167 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
10168 A disabled pretty-printer is not forgotten, it may be enabled again later.
10170 @kindex enable pretty-printer
10171 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
10172 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
10177 Suppose we have three pretty-printers installed: one from library1.so
10178 named @code{foo} that prints objects of type @code{foo}, and
10179 another from library2.so named @code{bar} that prints two types of objects,
10180 @code{bar1} and @code{bar2}.
10183 (gdb) info pretty-printer
10190 (gdb) info pretty-printer library2
10195 (gdb) disable pretty-printer library1
10197 2 of 3 printers enabled
10198 (gdb) info pretty-printer
10205 (gdb) disable pretty-printer library2 bar:bar1
10207 1 of 3 printers enabled
10208 (gdb) info pretty-printer library2
10215 (gdb) disable pretty-printer library2 bar
10217 0 of 3 printers enabled
10218 (gdb) info pretty-printer library2
10227 Note that for @code{bar} the entire printer can be disabled,
10228 as can each individual subprinter.
10230 @node Value History
10231 @section Value History
10233 @cindex value history
10234 @cindex history of values printed by @value{GDBN}
10235 Values printed by the @code{print} command are saved in the @value{GDBN}
10236 @dfn{value history}. This allows you to refer to them in other expressions.
10237 Values are kept until the symbol table is re-read or discarded
10238 (for example with the @code{file} or @code{symbol-file} commands).
10239 When the symbol table changes, the value history is discarded,
10240 since the values may contain pointers back to the types defined in the
10245 @cindex history number
10246 The values printed are given @dfn{history numbers} by which you can
10247 refer to them. These are successive integers starting with one.
10248 @code{print} shows you the history number assigned to a value by
10249 printing @samp{$@var{num} = } before the value; here @var{num} is the
10252 To refer to any previous value, use @samp{$} followed by the value's
10253 history number. The way @code{print} labels its output is designed to
10254 remind you of this. Just @code{$} refers to the most recent value in
10255 the history, and @code{$$} refers to the value before that.
10256 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
10257 is the value just prior to @code{$$}, @code{$$1} is equivalent to
10258 @code{$$}, and @code{$$0} is equivalent to @code{$}.
10260 For example, suppose you have just printed a pointer to a structure and
10261 want to see the contents of the structure. It suffices to type
10267 If you have a chain of structures where the component @code{next} points
10268 to the next one, you can print the contents of the next one with this:
10275 You can print successive links in the chain by repeating this
10276 command---which you can do by just typing @key{RET}.
10278 Note that the history records values, not expressions. If the value of
10279 @code{x} is 4 and you type these commands:
10287 then the value recorded in the value history by the @code{print} command
10288 remains 4 even though the value of @code{x} has changed.
10291 @kindex show values
10293 Print the last ten values in the value history, with their item numbers.
10294 This is like @samp{p@ $$9} repeated ten times, except that @code{show
10295 values} does not change the history.
10297 @item show values @var{n}
10298 Print ten history values centered on history item number @var{n}.
10300 @item show values +
10301 Print ten history values just after the values last printed. If no more
10302 values are available, @code{show values +} produces no display.
10305 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
10306 same effect as @samp{show values +}.
10308 @node Convenience Vars
10309 @section Convenience Variables
10311 @cindex convenience variables
10312 @cindex user-defined variables
10313 @value{GDBN} provides @dfn{convenience variables} that you can use within
10314 @value{GDBN} to hold on to a value and refer to it later. These variables
10315 exist entirely within @value{GDBN}; they are not part of your program, and
10316 setting a convenience variable has no direct effect on further execution
10317 of your program. That is why you can use them freely.
10319 Convenience variables are prefixed with @samp{$}. Any name preceded by
10320 @samp{$} can be used for a convenience variable, unless it is one of
10321 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
10322 (Value history references, in contrast, are @emph{numbers} preceded
10323 by @samp{$}. @xref{Value History, ,Value History}.)
10325 You can save a value in a convenience variable with an assignment
10326 expression, just as you would set a variable in your program.
10330 set $foo = *object_ptr
10334 would save in @code{$foo} the value contained in the object pointed to by
10337 Using a convenience variable for the first time creates it, but its
10338 value is @code{void} until you assign a new value. You can alter the
10339 value with another assignment at any time.
10341 Convenience variables have no fixed types. You can assign a convenience
10342 variable any type of value, including structures and arrays, even if
10343 that variable already has a value of a different type. The convenience
10344 variable, when used as an expression, has the type of its current value.
10347 @kindex show convenience
10348 @cindex show all user variables and functions
10349 @item show convenience
10350 Print a list of convenience variables used so far, and their values,
10351 as well as a list of the convenience functions.
10352 Abbreviated @code{show conv}.
10354 @kindex init-if-undefined
10355 @cindex convenience variables, initializing
10356 @item init-if-undefined $@var{variable} = @var{expression}
10357 Set a convenience variable if it has not already been set. This is useful
10358 for user-defined commands that keep some state. It is similar, in concept,
10359 to using local static variables with initializers in C (except that
10360 convenience variables are global). It can also be used to allow users to
10361 override default values used in a command script.
10363 If the variable is already defined then the expression is not evaluated so
10364 any side-effects do not occur.
10367 One of the ways to use a convenience variable is as a counter to be
10368 incremented or a pointer to be advanced. For example, to print
10369 a field from successive elements of an array of structures:
10373 print bar[$i++]->contents
10377 Repeat that command by typing @key{RET}.
10379 Some convenience variables are created automatically by @value{GDBN} and given
10380 values likely to be useful.
10383 @vindex $_@r{, convenience variable}
10385 The variable @code{$_} is automatically set by the @code{x} command to
10386 the last address examined (@pxref{Memory, ,Examining Memory}). Other
10387 commands which provide a default address for @code{x} to examine also
10388 set @code{$_} to that address; these commands include @code{info line}
10389 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
10390 except when set by the @code{x} command, in which case it is a pointer
10391 to the type of @code{$__}.
10393 @vindex $__@r{, convenience variable}
10395 The variable @code{$__} is automatically set by the @code{x} command
10396 to the value found in the last address examined. Its type is chosen
10397 to match the format in which the data was printed.
10400 @vindex $_exitcode@r{, convenience variable}
10401 When the program being debugged terminates normally, @value{GDBN}
10402 automatically sets this variable to the exit code of the program, and
10403 resets @code{$_exitsignal} to @code{void}.
10406 @vindex $_exitsignal@r{, convenience variable}
10407 When the program being debugged dies due to an uncaught signal,
10408 @value{GDBN} automatically sets this variable to that signal's number,
10409 and resets @code{$_exitcode} to @code{void}.
10411 To distinguish between whether the program being debugged has exited
10412 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
10413 @code{$_exitsignal} is not @code{void}), the convenience function
10414 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
10415 Functions}). For example, considering the following source code:
10418 #include <signal.h>
10421 main (int argc, char *argv[])
10428 A valid way of telling whether the program being debugged has exited
10429 or signalled would be:
10432 (@value{GDBP}) define has_exited_or_signalled
10433 Type commands for definition of ``has_exited_or_signalled''.
10434 End with a line saying just ``end''.
10435 >if $_isvoid ($_exitsignal)
10436 >echo The program has exited\n
10438 >echo The program has signalled\n
10444 Program terminated with signal SIGALRM, Alarm clock.
10445 The program no longer exists.
10446 (@value{GDBP}) has_exited_or_signalled
10447 The program has signalled
10450 As can be seen, @value{GDBN} correctly informs that the program being
10451 debugged has signalled, since it calls @code{raise} and raises a
10452 @code{SIGALRM} signal. If the program being debugged had not called
10453 @code{raise}, then @value{GDBN} would report a normal exit:
10456 (@value{GDBP}) has_exited_or_signalled
10457 The program has exited
10461 The variable @code{$_exception} is set to the exception object being
10462 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
10465 @itemx $_probe_arg0@dots{}$_probe_arg11
10466 Arguments to a static probe. @xref{Static Probe Points}.
10469 @vindex $_sdata@r{, inspect, convenience variable}
10470 The variable @code{$_sdata} contains extra collected static tracepoint
10471 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
10472 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
10473 if extra static tracepoint data has not been collected.
10476 @vindex $_siginfo@r{, convenience variable}
10477 The variable @code{$_siginfo} contains extra signal information
10478 (@pxref{extra signal information}). Note that @code{$_siginfo}
10479 could be empty, if the application has not yet received any signals.
10480 For example, it will be empty before you execute the @code{run} command.
10483 @vindex $_tlb@r{, convenience variable}
10484 The variable @code{$_tlb} is automatically set when debugging
10485 applications running on MS-Windows in native mode or connected to
10486 gdbserver that supports the @code{qGetTIBAddr} request.
10487 @xref{General Query Packets}.
10488 This variable contains the address of the thread information block.
10491 The number of the current inferior. @xref{Inferiors and
10492 Programs, ,Debugging Multiple Inferiors and Programs}.
10495 The thread number of the current thread. @xref{thread numbers}.
10498 The global number of the current thread. @xref{global thread numbers}.
10502 @node Convenience Funs
10503 @section Convenience Functions
10505 @cindex convenience functions
10506 @value{GDBN} also supplies some @dfn{convenience functions}. These
10507 have a syntax similar to convenience variables. A convenience
10508 function can be used in an expression just like an ordinary function;
10509 however, a convenience function is implemented internally to
10512 These functions do not require @value{GDBN} to be configured with
10513 @code{Python} support, which means that they are always available.
10517 @item $_isvoid (@var{expr})
10518 @findex $_isvoid@r{, convenience function}
10519 Return one if the expression @var{expr} is @code{void}. Otherwise it
10522 A @code{void} expression is an expression where the type of the result
10523 is @code{void}. For example, you can examine a convenience variable
10524 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
10528 (@value{GDBP}) print $_exitcode
10530 (@value{GDBP}) print $_isvoid ($_exitcode)
10533 Starting program: ./a.out
10534 [Inferior 1 (process 29572) exited normally]
10535 (@value{GDBP}) print $_exitcode
10537 (@value{GDBP}) print $_isvoid ($_exitcode)
10541 In the example above, we used @code{$_isvoid} to check whether
10542 @code{$_exitcode} is @code{void} before and after the execution of the
10543 program being debugged. Before the execution there is no exit code to
10544 be examined, therefore @code{$_exitcode} is @code{void}. After the
10545 execution the program being debugged returned zero, therefore
10546 @code{$_exitcode} is zero, which means that it is not @code{void}
10549 The @code{void} expression can also be a call of a function from the
10550 program being debugged. For example, given the following function:
10559 The result of calling it inside @value{GDBN} is @code{void}:
10562 (@value{GDBP}) print foo ()
10564 (@value{GDBP}) print $_isvoid (foo ())
10566 (@value{GDBP}) set $v = foo ()
10567 (@value{GDBP}) print $v
10569 (@value{GDBP}) print $_isvoid ($v)
10575 These functions require @value{GDBN} to be configured with
10576 @code{Python} support.
10580 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
10581 @findex $_memeq@r{, convenience function}
10582 Returns one if the @var{length} bytes at the addresses given by
10583 @var{buf1} and @var{buf2} are equal.
10584 Otherwise it returns zero.
10586 @item $_regex(@var{str}, @var{regex})
10587 @findex $_regex@r{, convenience function}
10588 Returns one if the string @var{str} matches the regular expression
10589 @var{regex}. Otherwise it returns zero.
10590 The syntax of the regular expression is that specified by @code{Python}'s
10591 regular expression support.
10593 @item $_streq(@var{str1}, @var{str2})
10594 @findex $_streq@r{, convenience function}
10595 Returns one if the strings @var{str1} and @var{str2} are equal.
10596 Otherwise it returns zero.
10598 @item $_strlen(@var{str})
10599 @findex $_strlen@r{, convenience function}
10600 Returns the length of string @var{str}.
10602 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
10603 @findex $_caller_is@r{, convenience function}
10604 Returns one if the calling function's name is equal to @var{name}.
10605 Otherwise it returns zero.
10607 If the optional argument @var{number_of_frames} is provided,
10608 it is the number of frames up in the stack to look.
10616 at testsuite/gdb.python/py-caller-is.c:21
10617 #1 0x00000000004005a0 in middle_func ()
10618 at testsuite/gdb.python/py-caller-is.c:27
10619 #2 0x00000000004005ab in top_func ()
10620 at testsuite/gdb.python/py-caller-is.c:33
10621 #3 0x00000000004005b6 in main ()
10622 at testsuite/gdb.python/py-caller-is.c:39
10623 (gdb) print $_caller_is ("middle_func")
10625 (gdb) print $_caller_is ("top_func", 2)
10629 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
10630 @findex $_caller_matches@r{, convenience function}
10631 Returns one if the calling function's name matches the regular expression
10632 @var{regexp}. Otherwise it returns zero.
10634 If the optional argument @var{number_of_frames} is provided,
10635 it is the number of frames up in the stack to look.
10638 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
10639 @findex $_any_caller_is@r{, convenience function}
10640 Returns one if any calling function's name is equal to @var{name}.
10641 Otherwise it returns zero.
10643 If the optional argument @var{number_of_frames} is provided,
10644 it is the number of frames up in the stack to look.
10647 This function differs from @code{$_caller_is} in that this function
10648 checks all stack frames from the immediate caller to the frame specified
10649 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
10650 frame specified by @var{number_of_frames}.
10652 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
10653 @findex $_any_caller_matches@r{, convenience function}
10654 Returns one if any calling function's name matches the regular expression
10655 @var{regexp}. Otherwise it returns zero.
10657 If the optional argument @var{number_of_frames} is provided,
10658 it is the number of frames up in the stack to look.
10661 This function differs from @code{$_caller_matches} in that this function
10662 checks all stack frames from the immediate caller to the frame specified
10663 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
10664 frame specified by @var{number_of_frames}.
10668 @value{GDBN} provides the ability to list and get help on
10669 convenience functions.
10672 @item help function
10673 @kindex help function
10674 @cindex show all convenience functions
10675 Print a list of all convenience functions.
10682 You can refer to machine register contents, in expressions, as variables
10683 with names starting with @samp{$}. The names of registers are different
10684 for each machine; use @code{info registers} to see the names used on
10688 @kindex info registers
10689 @item info registers
10690 Print the names and values of all registers except floating-point
10691 and vector registers (in the selected stack frame).
10693 @kindex info all-registers
10694 @cindex floating point registers
10695 @item info all-registers
10696 Print the names and values of all registers, including floating-point
10697 and vector registers (in the selected stack frame).
10699 @item info registers @var{regname} @dots{}
10700 Print the @dfn{relativized} value of each specified register @var{regname}.
10701 As discussed in detail below, register values are normally relative to
10702 the selected stack frame. The @var{regname} may be any register name valid on
10703 the machine you are using, with or without the initial @samp{$}.
10706 @anchor{standard registers}
10707 @cindex stack pointer register
10708 @cindex program counter register
10709 @cindex process status register
10710 @cindex frame pointer register
10711 @cindex standard registers
10712 @value{GDBN} has four ``standard'' register names that are available (in
10713 expressions) on most machines---whenever they do not conflict with an
10714 architecture's canonical mnemonics for registers. The register names
10715 @code{$pc} and @code{$sp} are used for the program counter register and
10716 the stack pointer. @code{$fp} is used for a register that contains a
10717 pointer to the current stack frame, and @code{$ps} is used for a
10718 register that contains the processor status. For example,
10719 you could print the program counter in hex with
10726 or print the instruction to be executed next with
10733 or add four to the stack pointer@footnote{This is a way of removing
10734 one word from the stack, on machines where stacks grow downward in
10735 memory (most machines, nowadays). This assumes that the innermost
10736 stack frame is selected; setting @code{$sp} is not allowed when other
10737 stack frames are selected. To pop entire frames off the stack,
10738 regardless of machine architecture, use @code{return};
10739 see @ref{Returning, ,Returning from a Function}.} with
10745 Whenever possible, these four standard register names are available on
10746 your machine even though the machine has different canonical mnemonics,
10747 so long as there is no conflict. The @code{info registers} command
10748 shows the canonical names. For example, on the SPARC, @code{info
10749 registers} displays the processor status register as @code{$psr} but you
10750 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
10751 is an alias for the @sc{eflags} register.
10753 @value{GDBN} always considers the contents of an ordinary register as an
10754 integer when the register is examined in this way. Some machines have
10755 special registers which can hold nothing but floating point; these
10756 registers are considered to have floating point values. There is no way
10757 to refer to the contents of an ordinary register as floating point value
10758 (although you can @emph{print} it as a floating point value with
10759 @samp{print/f $@var{regname}}).
10761 Some registers have distinct ``raw'' and ``virtual'' data formats. This
10762 means that the data format in which the register contents are saved by
10763 the operating system is not the same one that your program normally
10764 sees. For example, the registers of the 68881 floating point
10765 coprocessor are always saved in ``extended'' (raw) format, but all C
10766 programs expect to work with ``double'' (virtual) format. In such
10767 cases, @value{GDBN} normally works with the virtual format only (the format
10768 that makes sense for your program), but the @code{info registers} command
10769 prints the data in both formats.
10771 @cindex SSE registers (x86)
10772 @cindex MMX registers (x86)
10773 Some machines have special registers whose contents can be interpreted
10774 in several different ways. For example, modern x86-based machines
10775 have SSE and MMX registers that can hold several values packed
10776 together in several different formats. @value{GDBN} refers to such
10777 registers in @code{struct} notation:
10780 (@value{GDBP}) print $xmm1
10782 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
10783 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
10784 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
10785 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
10786 v4_int32 = @{0, 20657912, 11, 13@},
10787 v2_int64 = @{88725056443645952, 55834574859@},
10788 uint128 = 0x0000000d0000000b013b36f800000000
10793 To set values of such registers, you need to tell @value{GDBN} which
10794 view of the register you wish to change, as if you were assigning
10795 value to a @code{struct} member:
10798 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
10801 Normally, register values are relative to the selected stack frame
10802 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
10803 value that the register would contain if all stack frames farther in
10804 were exited and their saved registers restored. In order to see the
10805 true contents of hardware registers, you must select the innermost
10806 frame (with @samp{frame 0}).
10808 @cindex caller-saved registers
10809 @cindex call-clobbered registers
10810 @cindex volatile registers
10811 @cindex <not saved> values
10812 Usually ABIs reserve some registers as not needed to be saved by the
10813 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
10814 registers). It may therefore not be possible for @value{GDBN} to know
10815 the value a register had before the call (in other words, in the outer
10816 frame), if the register value has since been changed by the callee.
10817 @value{GDBN} tries to deduce where the inner frame saved
10818 (``callee-saved'') registers, from the debug info, unwind info, or the
10819 machine code generated by your compiler. If some register is not
10820 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
10821 its own knowledge of the ABI, or because the debug/unwind info
10822 explicitly says the register's value is undefined), @value{GDBN}
10823 displays @w{@samp{<not saved>}} as the register's value. With targets
10824 that @value{GDBN} has no knowledge of the register saving convention,
10825 if a register was not saved by the callee, then its value and location
10826 in the outer frame are assumed to be the same of the inner frame.
10827 This is usually harmless, because if the register is call-clobbered,
10828 the caller either does not care what is in the register after the
10829 call, or has code to restore the value that it does care about. Note,
10830 however, that if you change such a register in the outer frame, you
10831 may also be affecting the inner frame. Also, the more ``outer'' the
10832 frame is you're looking at, the more likely a call-clobbered
10833 register's value is to be wrong, in the sense that it doesn't actually
10834 represent the value the register had just before the call.
10836 @node Floating Point Hardware
10837 @section Floating Point Hardware
10838 @cindex floating point
10840 Depending on the configuration, @value{GDBN} may be able to give
10841 you more information about the status of the floating point hardware.
10846 Display hardware-dependent information about the floating
10847 point unit. The exact contents and layout vary depending on the
10848 floating point chip. Currently, @samp{info float} is supported on
10849 the ARM and x86 machines.
10853 @section Vector Unit
10854 @cindex vector unit
10856 Depending on the configuration, @value{GDBN} may be able to give you
10857 more information about the status of the vector unit.
10860 @kindex info vector
10862 Display information about the vector unit. The exact contents and
10863 layout vary depending on the hardware.
10866 @node OS Information
10867 @section Operating System Auxiliary Information
10868 @cindex OS information
10870 @value{GDBN} provides interfaces to useful OS facilities that can help
10871 you debug your program.
10873 @cindex auxiliary vector
10874 @cindex vector, auxiliary
10875 Some operating systems supply an @dfn{auxiliary vector} to programs at
10876 startup. This is akin to the arguments and environment that you
10877 specify for a program, but contains a system-dependent variety of
10878 binary values that tell system libraries important details about the
10879 hardware, operating system, and process. Each value's purpose is
10880 identified by an integer tag; the meanings are well-known but system-specific.
10881 Depending on the configuration and operating system facilities,
10882 @value{GDBN} may be able to show you this information. For remote
10883 targets, this functionality may further depend on the remote stub's
10884 support of the @samp{qXfer:auxv:read} packet, see
10885 @ref{qXfer auxiliary vector read}.
10890 Display the auxiliary vector of the inferior, which can be either a
10891 live process or a core dump file. @value{GDBN} prints each tag value
10892 numerically, and also shows names and text descriptions for recognized
10893 tags. Some values in the vector are numbers, some bit masks, and some
10894 pointers to strings or other data. @value{GDBN} displays each value in the
10895 most appropriate form for a recognized tag, and in hexadecimal for
10896 an unrecognized tag.
10899 On some targets, @value{GDBN} can access operating system-specific
10900 information and show it to you. The types of information available
10901 will differ depending on the type of operating system running on the
10902 target. The mechanism used to fetch the data is described in
10903 @ref{Operating System Information}. For remote targets, this
10904 functionality depends on the remote stub's support of the
10905 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
10909 @item info os @var{infotype}
10911 Display OS information of the requested type.
10913 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
10915 @anchor{linux info os infotypes}
10917 @kindex info os cpus
10919 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
10920 the available fields from /proc/cpuinfo. For each supported architecture
10921 different fields are available. Two common entries are processor which gives
10922 CPU number and bogomips; a system constant that is calculated during
10923 kernel initialization.
10925 @kindex info os files
10927 Display the list of open file descriptors on the target. For each
10928 file descriptor, @value{GDBN} prints the identifier of the process
10929 owning the descriptor, the command of the owning process, the value
10930 of the descriptor, and the target of the descriptor.
10932 @kindex info os modules
10934 Display the list of all loaded kernel modules on the target. For each
10935 module, @value{GDBN} prints the module name, the size of the module in
10936 bytes, the number of times the module is used, the dependencies of the
10937 module, the status of the module, and the address of the loaded module
10940 @kindex info os msg
10942 Display the list of all System V message queues on the target. For each
10943 message queue, @value{GDBN} prints the message queue key, the message
10944 queue identifier, the access permissions, the current number of bytes
10945 on the queue, the current number of messages on the queue, the processes
10946 that last sent and received a message on the queue, the user and group
10947 of the owner and creator of the message queue, the times at which a
10948 message was last sent and received on the queue, and the time at which
10949 the message queue was last changed.
10951 @kindex info os processes
10953 Display the list of processes on the target. For each process,
10954 @value{GDBN} prints the process identifier, the name of the user, the
10955 command corresponding to the process, and the list of processor cores
10956 that the process is currently running on. (To understand what these
10957 properties mean, for this and the following info types, please consult
10958 the general @sc{gnu}/Linux documentation.)
10960 @kindex info os procgroups
10962 Display the list of process groups on the target. For each process,
10963 @value{GDBN} prints the identifier of the process group that it belongs
10964 to, the command corresponding to the process group leader, the process
10965 identifier, and the command line of the process. The list is sorted
10966 first by the process group identifier, then by the process identifier,
10967 so that processes belonging to the same process group are grouped together
10968 and the process group leader is listed first.
10970 @kindex info os semaphores
10972 Display the list of all System V semaphore sets on the target. For each
10973 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
10974 set identifier, the access permissions, the number of semaphores in the
10975 set, the user and group of the owner and creator of the semaphore set,
10976 and the times at which the semaphore set was operated upon and changed.
10978 @kindex info os shm
10980 Display the list of all System V shared-memory regions on the target.
10981 For each shared-memory region, @value{GDBN} prints the region key,
10982 the shared-memory identifier, the access permissions, the size of the
10983 region, the process that created the region, the process that last
10984 attached to or detached from the region, the current number of live
10985 attaches to the region, and the times at which the region was last
10986 attached to, detach from, and changed.
10988 @kindex info os sockets
10990 Display the list of Internet-domain sockets on the target. For each
10991 socket, @value{GDBN} prints the address and port of the local and
10992 remote endpoints, the current state of the connection, the creator of
10993 the socket, the IP address family of the socket, and the type of the
10996 @kindex info os threads
10998 Display the list of threads running on the target. For each thread,
10999 @value{GDBN} prints the identifier of the process that the thread
11000 belongs to, the command of the process, the thread identifier, and the
11001 processor core that it is currently running on. The main thread of a
11002 process is not listed.
11006 If @var{infotype} is omitted, then list the possible values for
11007 @var{infotype} and the kind of OS information available for each
11008 @var{infotype}. If the target does not return a list of possible
11009 types, this command will report an error.
11012 @node Memory Region Attributes
11013 @section Memory Region Attributes
11014 @cindex memory region attributes
11016 @dfn{Memory region attributes} allow you to describe special handling
11017 required by regions of your target's memory. @value{GDBN} uses
11018 attributes to determine whether to allow certain types of memory
11019 accesses; whether to use specific width accesses; and whether to cache
11020 target memory. By default the description of memory regions is
11021 fetched from the target (if the current target supports this), but the
11022 user can override the fetched regions.
11024 Defined memory regions can be individually enabled and disabled. When a
11025 memory region is disabled, @value{GDBN} uses the default attributes when
11026 accessing memory in that region. Similarly, if no memory regions have
11027 been defined, @value{GDBN} uses the default attributes when accessing
11030 When a memory region is defined, it is given a number to identify it;
11031 to enable, disable, or remove a memory region, you specify that number.
11035 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
11036 Define a memory region bounded by @var{lower} and @var{upper} with
11037 attributes @var{attributes}@dots{}, and add it to the list of regions
11038 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
11039 case: it is treated as the target's maximum memory address.
11040 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
11043 Discard any user changes to the memory regions and use target-supplied
11044 regions, if available, or no regions if the target does not support.
11047 @item delete mem @var{nums}@dots{}
11048 Remove memory regions @var{nums}@dots{} from the list of regions
11049 monitored by @value{GDBN}.
11051 @kindex disable mem
11052 @item disable mem @var{nums}@dots{}
11053 Disable monitoring of memory regions @var{nums}@dots{}.
11054 A disabled memory region is not forgotten.
11055 It may be enabled again later.
11058 @item enable mem @var{nums}@dots{}
11059 Enable monitoring of memory regions @var{nums}@dots{}.
11063 Print a table of all defined memory regions, with the following columns
11067 @item Memory Region Number
11068 @item Enabled or Disabled.
11069 Enabled memory regions are marked with @samp{y}.
11070 Disabled memory regions are marked with @samp{n}.
11073 The address defining the inclusive lower bound of the memory region.
11076 The address defining the exclusive upper bound of the memory region.
11079 The list of attributes set for this memory region.
11084 @subsection Attributes
11086 @subsubsection Memory Access Mode
11087 The access mode attributes set whether @value{GDBN} may make read or
11088 write accesses to a memory region.
11090 While these attributes prevent @value{GDBN} from performing invalid
11091 memory accesses, they do nothing to prevent the target system, I/O DMA,
11092 etc.@: from accessing memory.
11096 Memory is read only.
11098 Memory is write only.
11100 Memory is read/write. This is the default.
11103 @subsubsection Memory Access Size
11104 The access size attribute tells @value{GDBN} to use specific sized
11105 accesses in the memory region. Often memory mapped device registers
11106 require specific sized accesses. If no access size attribute is
11107 specified, @value{GDBN} may use accesses of any size.
11111 Use 8 bit memory accesses.
11113 Use 16 bit memory accesses.
11115 Use 32 bit memory accesses.
11117 Use 64 bit memory accesses.
11120 @c @subsubsection Hardware/Software Breakpoints
11121 @c The hardware/software breakpoint attributes set whether @value{GDBN}
11122 @c will use hardware or software breakpoints for the internal breakpoints
11123 @c used by the step, next, finish, until, etc. commands.
11127 @c Always use hardware breakpoints
11128 @c @item swbreak (default)
11131 @subsubsection Data Cache
11132 The data cache attributes set whether @value{GDBN} will cache target
11133 memory. While this generally improves performance by reducing debug
11134 protocol overhead, it can lead to incorrect results because @value{GDBN}
11135 does not know about volatile variables or memory mapped device
11140 Enable @value{GDBN} to cache target memory.
11142 Disable @value{GDBN} from caching target memory. This is the default.
11145 @subsection Memory Access Checking
11146 @value{GDBN} can be instructed to refuse accesses to memory that is
11147 not explicitly described. This can be useful if accessing such
11148 regions has undesired effects for a specific target, or to provide
11149 better error checking. The following commands control this behaviour.
11152 @kindex set mem inaccessible-by-default
11153 @item set mem inaccessible-by-default [on|off]
11154 If @code{on} is specified, make @value{GDBN} treat memory not
11155 explicitly described by the memory ranges as non-existent and refuse accesses
11156 to such memory. The checks are only performed if there's at least one
11157 memory range defined. If @code{off} is specified, make @value{GDBN}
11158 treat the memory not explicitly described by the memory ranges as RAM.
11159 The default value is @code{on}.
11160 @kindex show mem inaccessible-by-default
11161 @item show mem inaccessible-by-default
11162 Show the current handling of accesses to unknown memory.
11166 @c @subsubsection Memory Write Verification
11167 @c The memory write verification attributes set whether @value{GDBN}
11168 @c will re-reads data after each write to verify the write was successful.
11172 @c @item noverify (default)
11175 @node Dump/Restore Files
11176 @section Copy Between Memory and a File
11177 @cindex dump/restore files
11178 @cindex append data to a file
11179 @cindex dump data to a file
11180 @cindex restore data from a file
11182 You can use the commands @code{dump}, @code{append}, and
11183 @code{restore} to copy data between target memory and a file. The
11184 @code{dump} and @code{append} commands write data to a file, and the
11185 @code{restore} command reads data from a file back into the inferior's
11186 memory. Files may be in binary, Motorola S-record, Intel hex,
11187 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
11188 append to binary files, and cannot read from Verilog Hex files.
11193 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
11194 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
11195 Dump the contents of memory from @var{start_addr} to @var{end_addr},
11196 or the value of @var{expr}, to @var{filename} in the given format.
11198 The @var{format} parameter may be any one of:
11205 Motorola S-record format.
11207 Tektronix Hex format.
11209 Verilog Hex format.
11212 @value{GDBN} uses the same definitions of these formats as the
11213 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
11214 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
11218 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
11219 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
11220 Append the contents of memory from @var{start_addr} to @var{end_addr},
11221 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
11222 (@value{GDBN} can only append data to files in raw binary form.)
11225 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
11226 Restore the contents of file @var{filename} into memory. The
11227 @code{restore} command can automatically recognize any known @sc{bfd}
11228 file format, except for raw binary. To restore a raw binary file you
11229 must specify the optional keyword @code{binary} after the filename.
11231 If @var{bias} is non-zero, its value will be added to the addresses
11232 contained in the file. Binary files always start at address zero, so
11233 they will be restored at address @var{bias}. Other bfd files have
11234 a built-in location; they will be restored at offset @var{bias}
11235 from that location.
11237 If @var{start} and/or @var{end} are non-zero, then only data between
11238 file offset @var{start} and file offset @var{end} will be restored.
11239 These offsets are relative to the addresses in the file, before
11240 the @var{bias} argument is applied.
11244 @node Core File Generation
11245 @section How to Produce a Core File from Your Program
11246 @cindex dump core from inferior
11248 A @dfn{core file} or @dfn{core dump} is a file that records the memory
11249 image of a running process and its process status (register values
11250 etc.). Its primary use is post-mortem debugging of a program that
11251 crashed while it ran outside a debugger. A program that crashes
11252 automatically produces a core file, unless this feature is disabled by
11253 the user. @xref{Files}, for information on invoking @value{GDBN} in
11254 the post-mortem debugging mode.
11256 Occasionally, you may wish to produce a core file of the program you
11257 are debugging in order to preserve a snapshot of its state.
11258 @value{GDBN} has a special command for that.
11262 @kindex generate-core-file
11263 @item generate-core-file [@var{file}]
11264 @itemx gcore [@var{file}]
11265 Produce a core dump of the inferior process. The optional argument
11266 @var{file} specifies the file name where to put the core dump. If not
11267 specified, the file name defaults to @file{core.@var{pid}}, where
11268 @var{pid} is the inferior process ID.
11270 Note that this command is implemented only for some systems (as of
11271 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
11273 On @sc{gnu}/Linux, this command can take into account the value of the
11274 file @file{/proc/@var{pid}/coredump_filter} when generating the core
11275 dump (@pxref{set use-coredump-filter}).
11277 @kindex set use-coredump-filter
11278 @anchor{set use-coredump-filter}
11279 @item set use-coredump-filter on
11280 @itemx set use-coredump-filter off
11281 Enable or disable the use of the file
11282 @file{/proc/@var{pid}/coredump_filter} when generating core dump
11283 files. This file is used by the Linux kernel to decide what types of
11284 memory mappings will be dumped or ignored when generating a core dump
11285 file. @var{pid} is the process ID of a currently running process.
11287 To make use of this feature, you have to write in the
11288 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
11289 which is a bit mask representing the memory mapping types. If a bit
11290 is set in the bit mask, then the memory mappings of the corresponding
11291 types will be dumped; otherwise, they will be ignored. This
11292 configuration is inherited by child processes. For more information
11293 about the bits that can be set in the
11294 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
11295 manpage of @code{core(5)}.
11297 By default, this option is @code{on}. If this option is turned
11298 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
11299 and instead uses the same default value as the Linux kernel in order
11300 to decide which pages will be dumped in the core dump file. This
11301 value is currently @code{0x33}, which means that bits @code{0}
11302 (anonymous private mappings), @code{1} (anonymous shared mappings),
11303 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
11304 This will cause these memory mappings to be dumped automatically.
11307 @node Character Sets
11308 @section Character Sets
11309 @cindex character sets
11311 @cindex translating between character sets
11312 @cindex host character set
11313 @cindex target character set
11315 If the program you are debugging uses a different character set to
11316 represent characters and strings than the one @value{GDBN} uses itself,
11317 @value{GDBN} can automatically translate between the character sets for
11318 you. The character set @value{GDBN} uses we call the @dfn{host
11319 character set}; the one the inferior program uses we call the
11320 @dfn{target character set}.
11322 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
11323 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
11324 remote protocol (@pxref{Remote Debugging}) to debug a program
11325 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
11326 then the host character set is Latin-1, and the target character set is
11327 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
11328 target-charset EBCDIC-US}, then @value{GDBN} translates between
11329 @sc{ebcdic} and Latin 1 as you print character or string values, or use
11330 character and string literals in expressions.
11332 @value{GDBN} has no way to automatically recognize which character set
11333 the inferior program uses; you must tell it, using the @code{set
11334 target-charset} command, described below.
11336 Here are the commands for controlling @value{GDBN}'s character set
11340 @item set target-charset @var{charset}
11341 @kindex set target-charset
11342 Set the current target character set to @var{charset}. To display the
11343 list of supported target character sets, type
11344 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
11346 @item set host-charset @var{charset}
11347 @kindex set host-charset
11348 Set the current host character set to @var{charset}.
11350 By default, @value{GDBN} uses a host character set appropriate to the
11351 system it is running on; you can override that default using the
11352 @code{set host-charset} command. On some systems, @value{GDBN} cannot
11353 automatically determine the appropriate host character set. In this
11354 case, @value{GDBN} uses @samp{UTF-8}.
11356 @value{GDBN} can only use certain character sets as its host character
11357 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
11358 @value{GDBN} will list the host character sets it supports.
11360 @item set charset @var{charset}
11361 @kindex set charset
11362 Set the current host and target character sets to @var{charset}. As
11363 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
11364 @value{GDBN} will list the names of the character sets that can be used
11365 for both host and target.
11368 @kindex show charset
11369 Show the names of the current host and target character sets.
11371 @item show host-charset
11372 @kindex show host-charset
11373 Show the name of the current host character set.
11375 @item show target-charset
11376 @kindex show target-charset
11377 Show the name of the current target character set.
11379 @item set target-wide-charset @var{charset}
11380 @kindex set target-wide-charset
11381 Set the current target's wide character set to @var{charset}. This is
11382 the character set used by the target's @code{wchar_t} type. To
11383 display the list of supported wide character sets, type
11384 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
11386 @item show target-wide-charset
11387 @kindex show target-wide-charset
11388 Show the name of the current target's wide character set.
11391 Here is an example of @value{GDBN}'s character set support in action.
11392 Assume that the following source code has been placed in the file
11393 @file{charset-test.c}:
11399 = @{72, 101, 108, 108, 111, 44, 32, 119,
11400 111, 114, 108, 100, 33, 10, 0@};
11401 char ibm1047_hello[]
11402 = @{200, 133, 147, 147, 150, 107, 64, 166,
11403 150, 153, 147, 132, 90, 37, 0@};
11407 printf ("Hello, world!\n");
11411 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
11412 containing the string @samp{Hello, world!} followed by a newline,
11413 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
11415 We compile the program, and invoke the debugger on it:
11418 $ gcc -g charset-test.c -o charset-test
11419 $ gdb -nw charset-test
11420 GNU gdb 2001-12-19-cvs
11421 Copyright 2001 Free Software Foundation, Inc.
11426 We can use the @code{show charset} command to see what character sets
11427 @value{GDBN} is currently using to interpret and display characters and
11431 (@value{GDBP}) show charset
11432 The current host and target character set is `ISO-8859-1'.
11436 For the sake of printing this manual, let's use @sc{ascii} as our
11437 initial character set:
11439 (@value{GDBP}) set charset ASCII
11440 (@value{GDBP}) show charset
11441 The current host and target character set is `ASCII'.
11445 Let's assume that @sc{ascii} is indeed the correct character set for our
11446 host system --- in other words, let's assume that if @value{GDBN} prints
11447 characters using the @sc{ascii} character set, our terminal will display
11448 them properly. Since our current target character set is also
11449 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
11452 (@value{GDBP}) print ascii_hello
11453 $1 = 0x401698 "Hello, world!\n"
11454 (@value{GDBP}) print ascii_hello[0]
11459 @value{GDBN} uses the target character set for character and string
11460 literals you use in expressions:
11463 (@value{GDBP}) print '+'
11468 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
11471 @value{GDBN} relies on the user to tell it which character set the
11472 target program uses. If we print @code{ibm1047_hello} while our target
11473 character set is still @sc{ascii}, we get jibberish:
11476 (@value{GDBP}) print ibm1047_hello
11477 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
11478 (@value{GDBP}) print ibm1047_hello[0]
11483 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
11484 @value{GDBN} tells us the character sets it supports:
11487 (@value{GDBP}) set target-charset
11488 ASCII EBCDIC-US IBM1047 ISO-8859-1
11489 (@value{GDBP}) set target-charset
11492 We can select @sc{ibm1047} as our target character set, and examine the
11493 program's strings again. Now the @sc{ascii} string is wrong, but
11494 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
11495 target character set, @sc{ibm1047}, to the host character set,
11496 @sc{ascii}, and they display correctly:
11499 (@value{GDBP}) set target-charset IBM1047
11500 (@value{GDBP}) show charset
11501 The current host character set is `ASCII'.
11502 The current target character set is `IBM1047'.
11503 (@value{GDBP}) print ascii_hello
11504 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
11505 (@value{GDBP}) print ascii_hello[0]
11507 (@value{GDBP}) print ibm1047_hello
11508 $8 = 0x4016a8 "Hello, world!\n"
11509 (@value{GDBP}) print ibm1047_hello[0]
11514 As above, @value{GDBN} uses the target character set for character and
11515 string literals you use in expressions:
11518 (@value{GDBP}) print '+'
11523 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
11526 @node Caching Target Data
11527 @section Caching Data of Targets
11528 @cindex caching data of targets
11530 @value{GDBN} caches data exchanged between the debugger and a target.
11531 Each cache is associated with the address space of the inferior.
11532 @xref{Inferiors and Programs}, about inferior and address space.
11533 Such caching generally improves performance in remote debugging
11534 (@pxref{Remote Debugging}), because it reduces the overhead of the
11535 remote protocol by bundling memory reads and writes into large chunks.
11536 Unfortunately, simply caching everything would lead to incorrect results,
11537 since @value{GDBN} does not necessarily know anything about volatile
11538 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
11539 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
11541 Therefore, by default, @value{GDBN} only caches data
11542 known to be on the stack@footnote{In non-stop mode, it is moderately
11543 rare for a running thread to modify the stack of a stopped thread
11544 in a way that would interfere with a backtrace, and caching of
11545 stack reads provides a significant speed up of remote backtraces.} or
11546 in the code segment.
11547 Other regions of memory can be explicitly marked as
11548 cacheable; @pxref{Memory Region Attributes}.
11551 @kindex set remotecache
11552 @item set remotecache on
11553 @itemx set remotecache off
11554 This option no longer does anything; it exists for compatibility
11557 @kindex show remotecache
11558 @item show remotecache
11559 Show the current state of the obsolete remotecache flag.
11561 @kindex set stack-cache
11562 @item set stack-cache on
11563 @itemx set stack-cache off
11564 Enable or disable caching of stack accesses. When @code{on}, use
11565 caching. By default, this option is @code{on}.
11567 @kindex show stack-cache
11568 @item show stack-cache
11569 Show the current state of data caching for memory accesses.
11571 @kindex set code-cache
11572 @item set code-cache on
11573 @itemx set code-cache off
11574 Enable or disable caching of code segment accesses. When @code{on},
11575 use caching. By default, this option is @code{on}. This improves
11576 performance of disassembly in remote debugging.
11578 @kindex show code-cache
11579 @item show code-cache
11580 Show the current state of target memory cache for code segment
11583 @kindex info dcache
11584 @item info dcache @r{[}line@r{]}
11585 Print the information about the performance of data cache of the
11586 current inferior's address space. The information displayed
11587 includes the dcache width and depth, and for each cache line, its
11588 number, address, and how many times it was referenced. This
11589 command is useful for debugging the data cache operation.
11591 If a line number is specified, the contents of that line will be
11594 @item set dcache size @var{size}
11595 @cindex dcache size
11596 @kindex set dcache size
11597 Set maximum number of entries in dcache (dcache depth above).
11599 @item set dcache line-size @var{line-size}
11600 @cindex dcache line-size
11601 @kindex set dcache line-size
11602 Set number of bytes each dcache entry caches (dcache width above).
11603 Must be a power of 2.
11605 @item show dcache size
11606 @kindex show dcache size
11607 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
11609 @item show dcache line-size
11610 @kindex show dcache line-size
11611 Show default size of dcache lines.
11615 @node Searching Memory
11616 @section Search Memory
11617 @cindex searching memory
11619 Memory can be searched for a particular sequence of bytes with the
11620 @code{find} command.
11624 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
11625 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
11626 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
11627 etc. The search begins at address @var{start_addr} and continues for either
11628 @var{len} bytes or through to @var{end_addr} inclusive.
11631 @var{s} and @var{n} are optional parameters.
11632 They may be specified in either order, apart or together.
11635 @item @var{s}, search query size
11636 The size of each search query value.
11642 halfwords (two bytes)
11646 giant words (eight bytes)
11649 All values are interpreted in the current language.
11650 This means, for example, that if the current source language is C/C@t{++}
11651 then searching for the string ``hello'' includes the trailing '\0'.
11653 If the value size is not specified, it is taken from the
11654 value's type in the current language.
11655 This is useful when one wants to specify the search
11656 pattern as a mixture of types.
11657 Note that this means, for example, that in the case of C-like languages
11658 a search for an untyped 0x42 will search for @samp{(int) 0x42}
11659 which is typically four bytes.
11661 @item @var{n}, maximum number of finds
11662 The maximum number of matches to print. The default is to print all finds.
11665 You can use strings as search values. Quote them with double-quotes
11667 The string value is copied into the search pattern byte by byte,
11668 regardless of the endianness of the target and the size specification.
11670 The address of each match found is printed as well as a count of the
11671 number of matches found.
11673 The address of the last value found is stored in convenience variable
11675 A count of the number of matches is stored in @samp{$numfound}.
11677 For example, if stopped at the @code{printf} in this function:
11683 static char hello[] = "hello-hello";
11684 static struct @{ char c; short s; int i; @}
11685 __attribute__ ((packed)) mixed
11686 = @{ 'c', 0x1234, 0x87654321 @};
11687 printf ("%s\n", hello);
11692 you get during debugging:
11695 (gdb) find &hello[0], +sizeof(hello), "hello"
11696 0x804956d <hello.1620+6>
11698 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
11699 0x8049567 <hello.1620>
11700 0x804956d <hello.1620+6>
11702 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
11703 0x8049567 <hello.1620>
11705 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
11706 0x8049560 <mixed.1625>
11708 (gdb) print $numfound
11711 $2 = (void *) 0x8049560
11714 @node Optimized Code
11715 @chapter Debugging Optimized Code
11716 @cindex optimized code, debugging
11717 @cindex debugging optimized code
11719 Almost all compilers support optimization. With optimization
11720 disabled, the compiler generates assembly code that corresponds
11721 directly to your source code, in a simplistic way. As the compiler
11722 applies more powerful optimizations, the generated assembly code
11723 diverges from your original source code. With help from debugging
11724 information generated by the compiler, @value{GDBN} can map from
11725 the running program back to constructs from your original source.
11727 @value{GDBN} is more accurate with optimization disabled. If you
11728 can recompile without optimization, it is easier to follow the
11729 progress of your program during debugging. But, there are many cases
11730 where you may need to debug an optimized version.
11732 When you debug a program compiled with @samp{-g -O}, remember that the
11733 optimizer has rearranged your code; the debugger shows you what is
11734 really there. Do not be too surprised when the execution path does not
11735 exactly match your source file! An extreme example: if you define a
11736 variable, but never use it, @value{GDBN} never sees that
11737 variable---because the compiler optimizes it out of existence.
11739 Some things do not work as well with @samp{-g -O} as with just
11740 @samp{-g}, particularly on machines with instruction scheduling. If in
11741 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
11742 please report it to us as a bug (including a test case!).
11743 @xref{Variables}, for more information about debugging optimized code.
11746 * Inline Functions:: How @value{GDBN} presents inlining
11747 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
11750 @node Inline Functions
11751 @section Inline Functions
11752 @cindex inline functions, debugging
11754 @dfn{Inlining} is an optimization that inserts a copy of the function
11755 body directly at each call site, instead of jumping to a shared
11756 routine. @value{GDBN} displays inlined functions just like
11757 non-inlined functions. They appear in backtraces. You can view their
11758 arguments and local variables, step into them with @code{step}, skip
11759 them with @code{next}, and escape from them with @code{finish}.
11760 You can check whether a function was inlined by using the
11761 @code{info frame} command.
11763 For @value{GDBN} to support inlined functions, the compiler must
11764 record information about inlining in the debug information ---
11765 @value{NGCC} using the @sc{dwarf 2} format does this, and several
11766 other compilers do also. @value{GDBN} only supports inlined functions
11767 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
11768 do not emit two required attributes (@samp{DW_AT_call_file} and
11769 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
11770 function calls with earlier versions of @value{NGCC}. It instead
11771 displays the arguments and local variables of inlined functions as
11772 local variables in the caller.
11774 The body of an inlined function is directly included at its call site;
11775 unlike a non-inlined function, there are no instructions devoted to
11776 the call. @value{GDBN} still pretends that the call site and the
11777 start of the inlined function are different instructions. Stepping to
11778 the call site shows the call site, and then stepping again shows
11779 the first line of the inlined function, even though no additional
11780 instructions are executed.
11782 This makes source-level debugging much clearer; you can see both the
11783 context of the call and then the effect of the call. Only stepping by
11784 a single instruction using @code{stepi} or @code{nexti} does not do
11785 this; single instruction steps always show the inlined body.
11787 There are some ways that @value{GDBN} does not pretend that inlined
11788 function calls are the same as normal calls:
11792 Setting breakpoints at the call site of an inlined function may not
11793 work, because the call site does not contain any code. @value{GDBN}
11794 may incorrectly move the breakpoint to the next line of the enclosing
11795 function, after the call. This limitation will be removed in a future
11796 version of @value{GDBN}; until then, set a breakpoint on an earlier line
11797 or inside the inlined function instead.
11800 @value{GDBN} cannot locate the return value of inlined calls after
11801 using the @code{finish} command. This is a limitation of compiler-generated
11802 debugging information; after @code{finish}, you can step to the next line
11803 and print a variable where your program stored the return value.
11807 @node Tail Call Frames
11808 @section Tail Call Frames
11809 @cindex tail call frames, debugging
11811 Function @code{B} can call function @code{C} in its very last statement. In
11812 unoptimized compilation the call of @code{C} is immediately followed by return
11813 instruction at the end of @code{B} code. Optimizing compiler may replace the
11814 call and return in function @code{B} into one jump to function @code{C}
11815 instead. Such use of a jump instruction is called @dfn{tail call}.
11817 During execution of function @code{C}, there will be no indication in the
11818 function call stack frames that it was tail-called from @code{B}. If function
11819 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
11820 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
11821 some cases @value{GDBN} can determine that @code{C} was tail-called from
11822 @code{B}, and it will then create fictitious call frame for that, with the
11823 return address set up as if @code{B} called @code{C} normally.
11825 This functionality is currently supported only by DWARF 2 debugging format and
11826 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
11827 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
11830 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
11831 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
11835 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
11837 Stack level 1, frame at 0x7fffffffda30:
11838 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
11839 tail call frame, caller of frame at 0x7fffffffda30
11840 source language c++.
11841 Arglist at unknown address.
11842 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
11845 The detection of all the possible code path executions can find them ambiguous.
11846 There is no execution history stored (possible @ref{Reverse Execution} is never
11847 used for this purpose) and the last known caller could have reached the known
11848 callee by multiple different jump sequences. In such case @value{GDBN} still
11849 tries to show at least all the unambiguous top tail callers and all the
11850 unambiguous bottom tail calees, if any.
11853 @anchor{set debug entry-values}
11854 @item set debug entry-values
11855 @kindex set debug entry-values
11856 When set to on, enables printing of analysis messages for both frame argument
11857 values at function entry and tail calls. It will show all the possible valid
11858 tail calls code paths it has considered. It will also print the intersection
11859 of them with the final unambiguous (possibly partial or even empty) code path
11862 @item show debug entry-values
11863 @kindex show debug entry-values
11864 Show the current state of analysis messages printing for both frame argument
11865 values at function entry and tail calls.
11868 The analysis messages for tail calls can for example show why the virtual tail
11869 call frame for function @code{c} has not been recognized (due to the indirect
11870 reference by variable @code{x}):
11873 static void __attribute__((noinline, noclone)) c (void);
11874 void (*x) (void) = c;
11875 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
11876 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
11877 int main (void) @{ x (); return 0; @}
11879 Breakpoint 1, DW_OP_GNU_entry_value resolving cannot find
11880 DW_TAG_GNU_call_site 0x40039a in main
11882 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
11885 #1 0x000000000040039a in main () at t.c:5
11888 Another possibility is an ambiguous virtual tail call frames resolution:
11892 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
11893 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
11894 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
11895 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
11896 static void __attribute__((noinline, noclone)) b (void)
11897 @{ if (i) c (); else e (); @}
11898 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
11899 int main (void) @{ a (); return 0; @}
11901 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
11902 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
11903 tailcall: reduced: 0x4004d2(a) |
11906 #1 0x00000000004004d2 in a () at t.c:8
11907 #2 0x0000000000400395 in main () at t.c:9
11910 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
11911 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
11913 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
11914 @ifset HAVE_MAKEINFO_CLICK
11915 @set ARROW @click{}
11916 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
11917 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
11919 @ifclear HAVE_MAKEINFO_CLICK
11921 @set CALLSEQ1B @value{CALLSEQ1A}
11922 @set CALLSEQ2B @value{CALLSEQ2A}
11925 Frames #0 and #2 are real, #1 is a virtual tail call frame.
11926 The code can have possible execution paths @value{CALLSEQ1B} or
11927 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
11929 @code{initial:} state shows some random possible calling sequence @value{GDBN}
11930 has found. It then finds another possible calling sequcen - that one is
11931 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
11932 printed as the @code{reduced:} calling sequence. That one could have many
11933 futher @code{compare:} and @code{reduced:} statements as long as there remain
11934 any non-ambiguous sequence entries.
11936 For the frame of function @code{b} in both cases there are different possible
11937 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
11938 also ambigous. The only non-ambiguous frame is the one for function @code{a},
11939 therefore this one is displayed to the user while the ambiguous frames are
11942 There can be also reasons why printing of frame argument values at function
11947 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
11948 static void __attribute__((noinline, noclone)) a (int i);
11949 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
11950 static void __attribute__((noinline, noclone)) a (int i)
11951 @{ if (i) b (i - 1); else c (0); @}
11952 int main (void) @{ a (5); return 0; @}
11955 #0 c (i=i@@entry=0) at t.c:2
11956 #1 0x0000000000400428 in a (DW_OP_GNU_entry_value resolving has found
11957 function "a" at 0x400420 can call itself via tail calls
11958 i=<optimized out>) at t.c:6
11959 #2 0x000000000040036e in main () at t.c:7
11962 @value{GDBN} cannot find out from the inferior state if and how many times did
11963 function @code{a} call itself (via function @code{b}) as these calls would be
11964 tail calls. Such tail calls would modify thue @code{i} variable, therefore
11965 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
11966 prints @code{<optimized out>} instead.
11969 @chapter C Preprocessor Macros
11971 Some languages, such as C and C@t{++}, provide a way to define and invoke
11972 ``preprocessor macros'' which expand into strings of tokens.
11973 @value{GDBN} can evaluate expressions containing macro invocations, show
11974 the result of macro expansion, and show a macro's definition, including
11975 where it was defined.
11977 You may need to compile your program specially to provide @value{GDBN}
11978 with information about preprocessor macros. Most compilers do not
11979 include macros in their debugging information, even when you compile
11980 with the @option{-g} flag. @xref{Compilation}.
11982 A program may define a macro at one point, remove that definition later,
11983 and then provide a different definition after that. Thus, at different
11984 points in the program, a macro may have different definitions, or have
11985 no definition at all. If there is a current stack frame, @value{GDBN}
11986 uses the macros in scope at that frame's source code line. Otherwise,
11987 @value{GDBN} uses the macros in scope at the current listing location;
11990 Whenever @value{GDBN} evaluates an expression, it always expands any
11991 macro invocations present in the expression. @value{GDBN} also provides
11992 the following commands for working with macros explicitly.
11996 @kindex macro expand
11997 @cindex macro expansion, showing the results of preprocessor
11998 @cindex preprocessor macro expansion, showing the results of
11999 @cindex expanding preprocessor macros
12000 @item macro expand @var{expression}
12001 @itemx macro exp @var{expression}
12002 Show the results of expanding all preprocessor macro invocations in
12003 @var{expression}. Since @value{GDBN} simply expands macros, but does
12004 not parse the result, @var{expression} need not be a valid expression;
12005 it can be any string of tokens.
12008 @item macro expand-once @var{expression}
12009 @itemx macro exp1 @var{expression}
12010 @cindex expand macro once
12011 @i{(This command is not yet implemented.)} Show the results of
12012 expanding those preprocessor macro invocations that appear explicitly in
12013 @var{expression}. Macro invocations appearing in that expansion are
12014 left unchanged. This command allows you to see the effect of a
12015 particular macro more clearly, without being confused by further
12016 expansions. Since @value{GDBN} simply expands macros, but does not
12017 parse the result, @var{expression} need not be a valid expression; it
12018 can be any string of tokens.
12021 @cindex macro definition, showing
12022 @cindex definition of a macro, showing
12023 @cindex macros, from debug info
12024 @item info macro [-a|-all] [--] @var{macro}
12025 Show the current definition or all definitions of the named @var{macro},
12026 and describe the source location or compiler command-line where that
12027 definition was established. The optional double dash is to signify the end of
12028 argument processing and the beginning of @var{macro} for non C-like macros where
12029 the macro may begin with a hyphen.
12031 @kindex info macros
12032 @item info macros @var{location}
12033 Show all macro definitions that are in effect at the location specified
12034 by @var{location}, and describe the source location or compiler
12035 command-line where those definitions were established.
12037 @kindex macro define
12038 @cindex user-defined macros
12039 @cindex defining macros interactively
12040 @cindex macros, user-defined
12041 @item macro define @var{macro} @var{replacement-list}
12042 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
12043 Introduce a definition for a preprocessor macro named @var{macro},
12044 invocations of which are replaced by the tokens given in
12045 @var{replacement-list}. The first form of this command defines an
12046 ``object-like'' macro, which takes no arguments; the second form
12047 defines a ``function-like'' macro, which takes the arguments given in
12050 A definition introduced by this command is in scope in every
12051 expression evaluated in @value{GDBN}, until it is removed with the
12052 @code{macro undef} command, described below. The definition overrides
12053 all definitions for @var{macro} present in the program being debugged,
12054 as well as any previous user-supplied definition.
12056 @kindex macro undef
12057 @item macro undef @var{macro}
12058 Remove any user-supplied definition for the macro named @var{macro}.
12059 This command only affects definitions provided with the @code{macro
12060 define} command, described above; it cannot remove definitions present
12061 in the program being debugged.
12065 List all the macros defined using the @code{macro define} command.
12068 @cindex macros, example of debugging with
12069 Here is a transcript showing the above commands in action. First, we
12070 show our source files:
12075 #include "sample.h"
12078 #define ADD(x) (M + x)
12083 printf ("Hello, world!\n");
12085 printf ("We're so creative.\n");
12087 printf ("Goodbye, world!\n");
12094 Now, we compile the program using the @sc{gnu} C compiler,
12095 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
12096 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
12097 and @option{-gdwarf-4}; we recommend always choosing the most recent
12098 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
12099 includes information about preprocessor macros in the debugging
12103 $ gcc -gdwarf-2 -g3 sample.c -o sample
12107 Now, we start @value{GDBN} on our sample program:
12111 GNU gdb 2002-05-06-cvs
12112 Copyright 2002 Free Software Foundation, Inc.
12113 GDB is free software, @dots{}
12117 We can expand macros and examine their definitions, even when the
12118 program is not running. @value{GDBN} uses the current listing position
12119 to decide which macro definitions are in scope:
12122 (@value{GDBP}) list main
12125 5 #define ADD(x) (M + x)
12130 10 printf ("Hello, world!\n");
12132 12 printf ("We're so creative.\n");
12133 (@value{GDBP}) info macro ADD
12134 Defined at /home/jimb/gdb/macros/play/sample.c:5
12135 #define ADD(x) (M + x)
12136 (@value{GDBP}) info macro Q
12137 Defined at /home/jimb/gdb/macros/play/sample.h:1
12138 included at /home/jimb/gdb/macros/play/sample.c:2
12140 (@value{GDBP}) macro expand ADD(1)
12141 expands to: (42 + 1)
12142 (@value{GDBP}) macro expand-once ADD(1)
12143 expands to: once (M + 1)
12147 In the example above, note that @code{macro expand-once} expands only
12148 the macro invocation explicit in the original text --- the invocation of
12149 @code{ADD} --- but does not expand the invocation of the macro @code{M},
12150 which was introduced by @code{ADD}.
12152 Once the program is running, @value{GDBN} uses the macro definitions in
12153 force at the source line of the current stack frame:
12156 (@value{GDBP}) break main
12157 Breakpoint 1 at 0x8048370: file sample.c, line 10.
12159 Starting program: /home/jimb/gdb/macros/play/sample
12161 Breakpoint 1, main () at sample.c:10
12162 10 printf ("Hello, world!\n");
12166 At line 10, the definition of the macro @code{N} at line 9 is in force:
12169 (@value{GDBP}) info macro N
12170 Defined at /home/jimb/gdb/macros/play/sample.c:9
12172 (@value{GDBP}) macro expand N Q M
12173 expands to: 28 < 42
12174 (@value{GDBP}) print N Q M
12179 As we step over directives that remove @code{N}'s definition, and then
12180 give it a new definition, @value{GDBN} finds the definition (or lack
12181 thereof) in force at each point:
12184 (@value{GDBP}) next
12186 12 printf ("We're so creative.\n");
12187 (@value{GDBP}) info macro N
12188 The symbol `N' has no definition as a C/C++ preprocessor macro
12189 at /home/jimb/gdb/macros/play/sample.c:12
12190 (@value{GDBP}) next
12192 14 printf ("Goodbye, world!\n");
12193 (@value{GDBP}) info macro N
12194 Defined at /home/jimb/gdb/macros/play/sample.c:13
12196 (@value{GDBP}) macro expand N Q M
12197 expands to: 1729 < 42
12198 (@value{GDBP}) print N Q M
12203 In addition to source files, macros can be defined on the compilation command
12204 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
12205 such a way, @value{GDBN} displays the location of their definition as line zero
12206 of the source file submitted to the compiler.
12209 (@value{GDBP}) info macro __STDC__
12210 Defined at /home/jimb/gdb/macros/play/sample.c:0
12217 @chapter Tracepoints
12218 @c This chapter is based on the documentation written by Michael
12219 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
12221 @cindex tracepoints
12222 In some applications, it is not feasible for the debugger to interrupt
12223 the program's execution long enough for the developer to learn
12224 anything helpful about its behavior. If the program's correctness
12225 depends on its real-time behavior, delays introduced by a debugger
12226 might cause the program to change its behavior drastically, or perhaps
12227 fail, even when the code itself is correct. It is useful to be able
12228 to observe the program's behavior without interrupting it.
12230 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
12231 specify locations in the program, called @dfn{tracepoints}, and
12232 arbitrary expressions to evaluate when those tracepoints are reached.
12233 Later, using the @code{tfind} command, you can examine the values
12234 those expressions had when the program hit the tracepoints. The
12235 expressions may also denote objects in memory---structures or arrays,
12236 for example---whose values @value{GDBN} should record; while visiting
12237 a particular tracepoint, you may inspect those objects as if they were
12238 in memory at that moment. However, because @value{GDBN} records these
12239 values without interacting with you, it can do so quickly and
12240 unobtrusively, hopefully not disturbing the program's behavior.
12242 The tracepoint facility is currently available only for remote
12243 targets. @xref{Targets}. In addition, your remote target must know
12244 how to collect trace data. This functionality is implemented in the
12245 remote stub; however, none of the stubs distributed with @value{GDBN}
12246 support tracepoints as of this writing. The format of the remote
12247 packets used to implement tracepoints are described in @ref{Tracepoint
12250 It is also possible to get trace data from a file, in a manner reminiscent
12251 of corefiles; you specify the filename, and use @code{tfind} to search
12252 through the file. @xref{Trace Files}, for more details.
12254 This chapter describes the tracepoint commands and features.
12257 * Set Tracepoints::
12258 * Analyze Collected Data::
12259 * Tracepoint Variables::
12263 @node Set Tracepoints
12264 @section Commands to Set Tracepoints
12266 Before running such a @dfn{trace experiment}, an arbitrary number of
12267 tracepoints can be set. A tracepoint is actually a special type of
12268 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
12269 standard breakpoint commands. For instance, as with breakpoints,
12270 tracepoint numbers are successive integers starting from one, and many
12271 of the commands associated with tracepoints take the tracepoint number
12272 as their argument, to identify which tracepoint to work on.
12274 For each tracepoint, you can specify, in advance, some arbitrary set
12275 of data that you want the target to collect in the trace buffer when
12276 it hits that tracepoint. The collected data can include registers,
12277 local variables, or global data. Later, you can use @value{GDBN}
12278 commands to examine the values these data had at the time the
12279 tracepoint was hit.
12281 Tracepoints do not support every breakpoint feature. Ignore counts on
12282 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
12283 commands when they are hit. Tracepoints may not be thread-specific
12286 @cindex fast tracepoints
12287 Some targets may support @dfn{fast tracepoints}, which are inserted in
12288 a different way (such as with a jump instead of a trap), that is
12289 faster but possibly restricted in where they may be installed.
12291 @cindex static tracepoints
12292 @cindex markers, static tracepoints
12293 @cindex probing markers, static tracepoints
12294 Regular and fast tracepoints are dynamic tracing facilities, meaning
12295 that they can be used to insert tracepoints at (almost) any location
12296 in the target. Some targets may also support controlling @dfn{static
12297 tracepoints} from @value{GDBN}. With static tracing, a set of
12298 instrumentation points, also known as @dfn{markers}, are embedded in
12299 the target program, and can be activated or deactivated by name or
12300 address. These are usually placed at locations which facilitate
12301 investigating what the target is actually doing. @value{GDBN}'s
12302 support for static tracing includes being able to list instrumentation
12303 points, and attach them with @value{GDBN} defined high level
12304 tracepoints that expose the whole range of convenience of
12305 @value{GDBN}'s tracepoints support. Namely, support for collecting
12306 registers values and values of global or local (to the instrumentation
12307 point) variables; tracepoint conditions and trace state variables.
12308 The act of installing a @value{GDBN} static tracepoint on an
12309 instrumentation point, or marker, is referred to as @dfn{probing} a
12310 static tracepoint marker.
12312 @code{gdbserver} supports tracepoints on some target systems.
12313 @xref{Server,,Tracepoints support in @code{gdbserver}}.
12315 This section describes commands to set tracepoints and associated
12316 conditions and actions.
12319 * Create and Delete Tracepoints::
12320 * Enable and Disable Tracepoints::
12321 * Tracepoint Passcounts::
12322 * Tracepoint Conditions::
12323 * Trace State Variables::
12324 * Tracepoint Actions::
12325 * Listing Tracepoints::
12326 * Listing Static Tracepoint Markers::
12327 * Starting and Stopping Trace Experiments::
12328 * Tracepoint Restrictions::
12331 @node Create and Delete Tracepoints
12332 @subsection Create and Delete Tracepoints
12335 @cindex set tracepoint
12337 @item trace @var{location}
12338 The @code{trace} command is very similar to the @code{break} command.
12339 Its argument @var{location} can be any valid location.
12340 @xref{Specify Location}. The @code{trace} command defines a tracepoint,
12341 which is a point in the target program where the debugger will briefly stop,
12342 collect some data, and then allow the program to continue. Setting a tracepoint
12343 or changing its actions takes effect immediately if the remote stub
12344 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
12346 If remote stub doesn't support the @samp{InstallInTrace} feature, all
12347 these changes don't take effect until the next @code{tstart}
12348 command, and once a trace experiment is running, further changes will
12349 not have any effect until the next trace experiment starts. In addition,
12350 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
12351 address is not yet resolved. (This is similar to pending breakpoints.)
12352 Pending tracepoints are not downloaded to the target and not installed
12353 until they are resolved. The resolution of pending tracepoints requires
12354 @value{GDBN} support---when debugging with the remote target, and
12355 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
12356 tracing}), pending tracepoints can not be resolved (and downloaded to
12357 the remote stub) while @value{GDBN} is disconnected.
12359 Here are some examples of using the @code{trace} command:
12362 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
12364 (@value{GDBP}) @b{trace +2} // 2 lines forward
12366 (@value{GDBP}) @b{trace my_function} // first source line of function
12368 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
12370 (@value{GDBP}) @b{trace *0x2117c4} // an address
12374 You can abbreviate @code{trace} as @code{tr}.
12376 @item trace @var{location} if @var{cond}
12377 Set a tracepoint with condition @var{cond}; evaluate the expression
12378 @var{cond} each time the tracepoint is reached, and collect data only
12379 if the value is nonzero---that is, if @var{cond} evaluates as true.
12380 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
12381 information on tracepoint conditions.
12383 @item ftrace @var{location} [ if @var{cond} ]
12384 @cindex set fast tracepoint
12385 @cindex fast tracepoints, setting
12387 The @code{ftrace} command sets a fast tracepoint. For targets that
12388 support them, fast tracepoints will use a more efficient but possibly
12389 less general technique to trigger data collection, such as a jump
12390 instruction instead of a trap, or some sort of hardware support. It
12391 may not be possible to create a fast tracepoint at the desired
12392 location, in which case the command will exit with an explanatory
12395 @value{GDBN} handles arguments to @code{ftrace} exactly as for
12398 On 32-bit x86-architecture systems, fast tracepoints normally need to
12399 be placed at an instruction that is 5 bytes or longer, but can be
12400 placed at 4-byte instructions if the low 64K of memory of the target
12401 program is available to install trampolines. Some Unix-type systems,
12402 such as @sc{gnu}/Linux, exclude low addresses from the program's
12403 address space; but for instance with the Linux kernel it is possible
12404 to let @value{GDBN} use this area by doing a @command{sysctl} command
12405 to set the @code{mmap_min_addr} kernel parameter, as in
12408 sudo sysctl -w vm.mmap_min_addr=32768
12412 which sets the low address to 32K, which leaves plenty of room for
12413 trampolines. The minimum address should be set to a page boundary.
12415 @item strace @var{location} [ if @var{cond} ]
12416 @cindex set static tracepoint
12417 @cindex static tracepoints, setting
12418 @cindex probe static tracepoint marker
12420 The @code{strace} command sets a static tracepoint. For targets that
12421 support it, setting a static tracepoint probes a static
12422 instrumentation point, or marker, found at @var{location}. It may not
12423 be possible to set a static tracepoint at the desired location, in
12424 which case the command will exit with an explanatory message.
12426 @value{GDBN} handles arguments to @code{strace} exactly as for
12427 @code{trace}, with the addition that the user can also specify
12428 @code{-m @var{marker}} as @var{location}. This probes the marker
12429 identified by the @var{marker} string identifier. This identifier
12430 depends on the static tracepoint backend library your program is
12431 using. You can find all the marker identifiers in the @samp{ID} field
12432 of the @code{info static-tracepoint-markers} command output.
12433 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
12434 Markers}. For example, in the following small program using the UST
12440 trace_mark(ust, bar33, "str %s", "FOOBAZ");
12445 the marker id is composed of joining the first two arguments to the
12446 @code{trace_mark} call with a slash, which translates to:
12449 (@value{GDBP}) info static-tracepoint-markers
12450 Cnt Enb ID Address What
12451 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
12457 so you may probe the marker above with:
12460 (@value{GDBP}) strace -m ust/bar33
12463 Static tracepoints accept an extra collect action --- @code{collect
12464 $_sdata}. This collects arbitrary user data passed in the probe point
12465 call to the tracing library. In the UST example above, you'll see
12466 that the third argument to @code{trace_mark} is a printf-like format
12467 string. The user data is then the result of running that formating
12468 string against the following arguments. Note that @code{info
12469 static-tracepoint-markers} command output lists that format string in
12470 the @samp{Data:} field.
12472 You can inspect this data when analyzing the trace buffer, by printing
12473 the $_sdata variable like any other variable available to
12474 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
12477 @cindex last tracepoint number
12478 @cindex recent tracepoint number
12479 @cindex tracepoint number
12480 The convenience variable @code{$tpnum} records the tracepoint number
12481 of the most recently set tracepoint.
12483 @kindex delete tracepoint
12484 @cindex tracepoint deletion
12485 @item delete tracepoint @r{[}@var{num}@r{]}
12486 Permanently delete one or more tracepoints. With no argument, the
12487 default is to delete all tracepoints. Note that the regular
12488 @code{delete} command can remove tracepoints also.
12493 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
12495 (@value{GDBP}) @b{delete trace} // remove all tracepoints
12499 You can abbreviate this command as @code{del tr}.
12502 @node Enable and Disable Tracepoints
12503 @subsection Enable and Disable Tracepoints
12505 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
12508 @kindex disable tracepoint
12509 @item disable tracepoint @r{[}@var{num}@r{]}
12510 Disable tracepoint @var{num}, or all tracepoints if no argument
12511 @var{num} is given. A disabled tracepoint will have no effect during
12512 a trace experiment, but it is not forgotten. You can re-enable
12513 a disabled tracepoint using the @code{enable tracepoint} command.
12514 If the command is issued during a trace experiment and the debug target
12515 has support for disabling tracepoints during a trace experiment, then the
12516 change will be effective immediately. Otherwise, it will be applied to the
12517 next trace experiment.
12519 @kindex enable tracepoint
12520 @item enable tracepoint @r{[}@var{num}@r{]}
12521 Enable tracepoint @var{num}, or all tracepoints. If this command is
12522 issued during a trace experiment and the debug target supports enabling
12523 tracepoints during a trace experiment, then the enabled tracepoints will
12524 become effective immediately. Otherwise, they will become effective the
12525 next time a trace experiment is run.
12528 @node Tracepoint Passcounts
12529 @subsection Tracepoint Passcounts
12533 @cindex tracepoint pass count
12534 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
12535 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
12536 automatically stop a trace experiment. If a tracepoint's passcount is
12537 @var{n}, then the trace experiment will be automatically stopped on
12538 the @var{n}'th time that tracepoint is hit. If the tracepoint number
12539 @var{num} is not specified, the @code{passcount} command sets the
12540 passcount of the most recently defined tracepoint. If no passcount is
12541 given, the trace experiment will run until stopped explicitly by the
12547 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
12548 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
12550 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
12551 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
12552 (@value{GDBP}) @b{trace foo}
12553 (@value{GDBP}) @b{pass 3}
12554 (@value{GDBP}) @b{trace bar}
12555 (@value{GDBP}) @b{pass 2}
12556 (@value{GDBP}) @b{trace baz}
12557 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
12558 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
12559 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
12560 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
12564 @node Tracepoint Conditions
12565 @subsection Tracepoint Conditions
12566 @cindex conditional tracepoints
12567 @cindex tracepoint conditions
12569 The simplest sort of tracepoint collects data every time your program
12570 reaches a specified place. You can also specify a @dfn{condition} for
12571 a tracepoint. A condition is just a Boolean expression in your
12572 programming language (@pxref{Expressions, ,Expressions}). A
12573 tracepoint with a condition evaluates the expression each time your
12574 program reaches it, and data collection happens only if the condition
12577 Tracepoint conditions can be specified when a tracepoint is set, by
12578 using @samp{if} in the arguments to the @code{trace} command.
12579 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
12580 also be set or changed at any time with the @code{condition} command,
12581 just as with breakpoints.
12583 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
12584 the conditional expression itself. Instead, @value{GDBN} encodes the
12585 expression into an agent expression (@pxref{Agent Expressions})
12586 suitable for execution on the target, independently of @value{GDBN}.
12587 Global variables become raw memory locations, locals become stack
12588 accesses, and so forth.
12590 For instance, suppose you have a function that is usually called
12591 frequently, but should not be called after an error has occurred. You
12592 could use the following tracepoint command to collect data about calls
12593 of that function that happen while the error code is propagating
12594 through the program; an unconditional tracepoint could end up
12595 collecting thousands of useless trace frames that you would have to
12599 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
12602 @node Trace State Variables
12603 @subsection Trace State Variables
12604 @cindex trace state variables
12606 A @dfn{trace state variable} is a special type of variable that is
12607 created and managed by target-side code. The syntax is the same as
12608 that for GDB's convenience variables (a string prefixed with ``$''),
12609 but they are stored on the target. They must be created explicitly,
12610 using a @code{tvariable} command. They are always 64-bit signed
12613 Trace state variables are remembered by @value{GDBN}, and downloaded
12614 to the target along with tracepoint information when the trace
12615 experiment starts. There are no intrinsic limits on the number of
12616 trace state variables, beyond memory limitations of the target.
12618 @cindex convenience variables, and trace state variables
12619 Although trace state variables are managed by the target, you can use
12620 them in print commands and expressions as if they were convenience
12621 variables; @value{GDBN} will get the current value from the target
12622 while the trace experiment is running. Trace state variables share
12623 the same namespace as other ``$'' variables, which means that you
12624 cannot have trace state variables with names like @code{$23} or
12625 @code{$pc}, nor can you have a trace state variable and a convenience
12626 variable with the same name.
12630 @item tvariable $@var{name} [ = @var{expression} ]
12632 The @code{tvariable} command creates a new trace state variable named
12633 @code{$@var{name}}, and optionally gives it an initial value of
12634 @var{expression}. The @var{expression} is evaluated when this command is
12635 entered; the result will be converted to an integer if possible,
12636 otherwise @value{GDBN} will report an error. A subsequent
12637 @code{tvariable} command specifying the same name does not create a
12638 variable, but instead assigns the supplied initial value to the
12639 existing variable of that name, overwriting any previous initial
12640 value. The default initial value is 0.
12642 @item info tvariables
12643 @kindex info tvariables
12644 List all the trace state variables along with their initial values.
12645 Their current values may also be displayed, if the trace experiment is
12648 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
12649 @kindex delete tvariable
12650 Delete the given trace state variables, or all of them if no arguments
12655 @node Tracepoint Actions
12656 @subsection Tracepoint Action Lists
12660 @cindex tracepoint actions
12661 @item actions @r{[}@var{num}@r{]}
12662 This command will prompt for a list of actions to be taken when the
12663 tracepoint is hit. If the tracepoint number @var{num} is not
12664 specified, this command sets the actions for the one that was most
12665 recently defined (so that you can define a tracepoint and then say
12666 @code{actions} without bothering about its number). You specify the
12667 actions themselves on the following lines, one action at a time, and
12668 terminate the actions list with a line containing just @code{end}. So
12669 far, the only defined actions are @code{collect}, @code{teval}, and
12670 @code{while-stepping}.
12672 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
12673 Commands, ,Breakpoint Command Lists}), except that only the defined
12674 actions are allowed; any other @value{GDBN} command is rejected.
12676 @cindex remove actions from a tracepoint
12677 To remove all actions from a tracepoint, type @samp{actions @var{num}}
12678 and follow it immediately with @samp{end}.
12681 (@value{GDBP}) @b{collect @var{data}} // collect some data
12683 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
12685 (@value{GDBP}) @b{end} // signals the end of actions.
12688 In the following example, the action list begins with @code{collect}
12689 commands indicating the things to be collected when the tracepoint is
12690 hit. Then, in order to single-step and collect additional data
12691 following the tracepoint, a @code{while-stepping} command is used,
12692 followed by the list of things to be collected after each step in a
12693 sequence of single steps. The @code{while-stepping} command is
12694 terminated by its own separate @code{end} command. Lastly, the action
12695 list is terminated by an @code{end} command.
12698 (@value{GDBP}) @b{trace foo}
12699 (@value{GDBP}) @b{actions}
12700 Enter actions for tracepoint 1, one per line:
12703 > while-stepping 12
12704 > collect $pc, arr[i]
12709 @kindex collect @r{(tracepoints)}
12710 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
12711 Collect values of the given expressions when the tracepoint is hit.
12712 This command accepts a comma-separated list of any valid expressions.
12713 In addition to global, static, or local variables, the following
12714 special arguments are supported:
12718 Collect all registers.
12721 Collect all function arguments.
12724 Collect all local variables.
12727 Collect the return address. This is helpful if you want to see more
12731 Collects the number of arguments from the static probe at which the
12732 tracepoint is located.
12733 @xref{Static Probe Points}.
12735 @item $_probe_arg@var{n}
12736 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
12737 from the static probe at which the tracepoint is located.
12738 @xref{Static Probe Points}.
12741 @vindex $_sdata@r{, collect}
12742 Collect static tracepoint marker specific data. Only available for
12743 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
12744 Lists}. On the UST static tracepoints library backend, an
12745 instrumentation point resembles a @code{printf} function call. The
12746 tracing library is able to collect user specified data formatted to a
12747 character string using the format provided by the programmer that
12748 instrumented the program. Other backends have similar mechanisms.
12749 Here's an example of a UST marker call:
12752 const char master_name[] = "$your_name";
12753 trace_mark(channel1, marker1, "hello %s", master_name)
12756 In this case, collecting @code{$_sdata} collects the string
12757 @samp{hello $yourname}. When analyzing the trace buffer, you can
12758 inspect @samp{$_sdata} like any other variable available to
12762 You can give several consecutive @code{collect} commands, each one
12763 with a single argument, or one @code{collect} command with several
12764 arguments separated by commas; the effect is the same.
12766 The optional @var{mods} changes the usual handling of the arguments.
12767 @code{s} requests that pointers to chars be handled as strings, in
12768 particular collecting the contents of the memory being pointed at, up
12769 to the first zero. The upper bound is by default the value of the
12770 @code{print elements} variable; if @code{s} is followed by a decimal
12771 number, that is the upper bound instead. So for instance
12772 @samp{collect/s25 mystr} collects as many as 25 characters at
12775 The command @code{info scope} (@pxref{Symbols, info scope}) is
12776 particularly useful for figuring out what data to collect.
12778 @kindex teval @r{(tracepoints)}
12779 @item teval @var{expr1}, @var{expr2}, @dots{}
12780 Evaluate the given expressions when the tracepoint is hit. This
12781 command accepts a comma-separated list of expressions. The results
12782 are discarded, so this is mainly useful for assigning values to trace
12783 state variables (@pxref{Trace State Variables}) without adding those
12784 values to the trace buffer, as would be the case if the @code{collect}
12787 @kindex while-stepping @r{(tracepoints)}
12788 @item while-stepping @var{n}
12789 Perform @var{n} single-step instruction traces after the tracepoint,
12790 collecting new data after each step. The @code{while-stepping}
12791 command is followed by the list of what to collect while stepping
12792 (followed by its own @code{end} command):
12795 > while-stepping 12
12796 > collect $regs, myglobal
12802 Note that @code{$pc} is not automatically collected by
12803 @code{while-stepping}; you need to explicitly collect that register if
12804 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
12807 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
12808 @kindex set default-collect
12809 @cindex default collection action
12810 This variable is a list of expressions to collect at each tracepoint
12811 hit. It is effectively an additional @code{collect} action prepended
12812 to every tracepoint action list. The expressions are parsed
12813 individually for each tracepoint, so for instance a variable named
12814 @code{xyz} may be interpreted as a global for one tracepoint, and a
12815 local for another, as appropriate to the tracepoint's location.
12817 @item show default-collect
12818 @kindex show default-collect
12819 Show the list of expressions that are collected by default at each
12824 @node Listing Tracepoints
12825 @subsection Listing Tracepoints
12828 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
12829 @kindex info tp @r{[}@var{n}@dots{}@r{]}
12830 @cindex information about tracepoints
12831 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
12832 Display information about the tracepoint @var{num}. If you don't
12833 specify a tracepoint number, displays information about all the
12834 tracepoints defined so far. The format is similar to that used for
12835 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
12836 command, simply restricting itself to tracepoints.
12838 A tracepoint's listing may include additional information specific to
12843 its passcount as given by the @code{passcount @var{n}} command
12846 the state about installed on target of each location
12850 (@value{GDBP}) @b{info trace}
12851 Num Type Disp Enb Address What
12852 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
12854 collect globfoo, $regs
12859 2 tracepoint keep y <MULTIPLE>
12861 2.1 y 0x0804859c in func4 at change-loc.h:35
12862 installed on target
12863 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
12864 installed on target
12865 2.3 y <PENDING> set_tracepoint
12866 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
12867 not installed on target
12872 This command can be abbreviated @code{info tp}.
12875 @node Listing Static Tracepoint Markers
12876 @subsection Listing Static Tracepoint Markers
12879 @kindex info static-tracepoint-markers
12880 @cindex information about static tracepoint markers
12881 @item info static-tracepoint-markers
12882 Display information about all static tracepoint markers defined in the
12885 For each marker, the following columns are printed:
12889 An incrementing counter, output to help readability. This is not a
12892 The marker ID, as reported by the target.
12893 @item Enabled or Disabled
12894 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
12895 that are not enabled.
12897 Where the marker is in your program, as a memory address.
12899 Where the marker is in the source for your program, as a file and line
12900 number. If the debug information included in the program does not
12901 allow @value{GDBN} to locate the source of the marker, this column
12902 will be left blank.
12906 In addition, the following information may be printed for each marker:
12910 User data passed to the tracing library by the marker call. In the
12911 UST backend, this is the format string passed as argument to the
12913 @item Static tracepoints probing the marker
12914 The list of static tracepoints attached to the marker.
12918 (@value{GDBP}) info static-tracepoint-markers
12919 Cnt ID Enb Address What
12920 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
12921 Data: number1 %d number2 %d
12922 Probed by static tracepoints: #2
12923 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
12929 @node Starting and Stopping Trace Experiments
12930 @subsection Starting and Stopping Trace Experiments
12933 @kindex tstart [ @var{notes} ]
12934 @cindex start a new trace experiment
12935 @cindex collected data discarded
12937 This command starts the trace experiment, and begins collecting data.
12938 It has the side effect of discarding all the data collected in the
12939 trace buffer during the previous trace experiment. If any arguments
12940 are supplied, they are taken as a note and stored with the trace
12941 experiment's state. The notes may be arbitrary text, and are
12942 especially useful with disconnected tracing in a multi-user context;
12943 the notes can explain what the trace is doing, supply user contact
12944 information, and so forth.
12946 @kindex tstop [ @var{notes} ]
12947 @cindex stop a running trace experiment
12949 This command stops the trace experiment. If any arguments are
12950 supplied, they are recorded with the experiment as a note. This is
12951 useful if you are stopping a trace started by someone else, for
12952 instance if the trace is interfering with the system's behavior and
12953 needs to be stopped quickly.
12955 @strong{Note}: a trace experiment and data collection may stop
12956 automatically if any tracepoint's passcount is reached
12957 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
12960 @cindex status of trace data collection
12961 @cindex trace experiment, status of
12963 This command displays the status of the current trace data
12967 Here is an example of the commands we described so far:
12970 (@value{GDBP}) @b{trace gdb_c_test}
12971 (@value{GDBP}) @b{actions}
12972 Enter actions for tracepoint #1, one per line.
12973 > collect $regs,$locals,$args
12974 > while-stepping 11
12978 (@value{GDBP}) @b{tstart}
12979 [time passes @dots{}]
12980 (@value{GDBP}) @b{tstop}
12983 @anchor{disconnected tracing}
12984 @cindex disconnected tracing
12985 You can choose to continue running the trace experiment even if
12986 @value{GDBN} disconnects from the target, voluntarily or
12987 involuntarily. For commands such as @code{detach}, the debugger will
12988 ask what you want to do with the trace. But for unexpected
12989 terminations (@value{GDBN} crash, network outage), it would be
12990 unfortunate to lose hard-won trace data, so the variable
12991 @code{disconnected-tracing} lets you decide whether the trace should
12992 continue running without @value{GDBN}.
12995 @item set disconnected-tracing on
12996 @itemx set disconnected-tracing off
12997 @kindex set disconnected-tracing
12998 Choose whether a tracing run should continue to run if @value{GDBN}
12999 has disconnected from the target. Note that @code{detach} or
13000 @code{quit} will ask you directly what to do about a running trace no
13001 matter what this variable's setting, so the variable is mainly useful
13002 for handling unexpected situations, such as loss of the network.
13004 @item show disconnected-tracing
13005 @kindex show disconnected-tracing
13006 Show the current choice for disconnected tracing.
13010 When you reconnect to the target, the trace experiment may or may not
13011 still be running; it might have filled the trace buffer in the
13012 meantime, or stopped for one of the other reasons. If it is running,
13013 it will continue after reconnection.
13015 Upon reconnection, the target will upload information about the
13016 tracepoints in effect. @value{GDBN} will then compare that
13017 information to the set of tracepoints currently defined, and attempt
13018 to match them up, allowing for the possibility that the numbers may
13019 have changed due to creation and deletion in the meantime. If one of
13020 the target's tracepoints does not match any in @value{GDBN}, the
13021 debugger will create a new tracepoint, so that you have a number with
13022 which to specify that tracepoint. This matching-up process is
13023 necessarily heuristic, and it may result in useless tracepoints being
13024 created; you may simply delete them if they are of no use.
13026 @cindex circular trace buffer
13027 If your target agent supports a @dfn{circular trace buffer}, then you
13028 can run a trace experiment indefinitely without filling the trace
13029 buffer; when space runs out, the agent deletes already-collected trace
13030 frames, oldest first, until there is enough room to continue
13031 collecting. This is especially useful if your tracepoints are being
13032 hit too often, and your trace gets terminated prematurely because the
13033 buffer is full. To ask for a circular trace buffer, simply set
13034 @samp{circular-trace-buffer} to on. You can set this at any time,
13035 including during tracing; if the agent can do it, it will change
13036 buffer handling on the fly, otherwise it will not take effect until
13040 @item set circular-trace-buffer on
13041 @itemx set circular-trace-buffer off
13042 @kindex set circular-trace-buffer
13043 Choose whether a tracing run should use a linear or circular buffer
13044 for trace data. A linear buffer will not lose any trace data, but may
13045 fill up prematurely, while a circular buffer will discard old trace
13046 data, but it will have always room for the latest tracepoint hits.
13048 @item show circular-trace-buffer
13049 @kindex show circular-trace-buffer
13050 Show the current choice for the trace buffer. Note that this may not
13051 match the agent's current buffer handling, nor is it guaranteed to
13052 match the setting that might have been in effect during a past run,
13053 for instance if you are looking at frames from a trace file.
13058 @item set trace-buffer-size @var{n}
13059 @itemx set trace-buffer-size unlimited
13060 @kindex set trace-buffer-size
13061 Request that the target use a trace buffer of @var{n} bytes. Not all
13062 targets will honor the request; they may have a compiled-in size for
13063 the trace buffer, or some other limitation. Set to a value of
13064 @code{unlimited} or @code{-1} to let the target use whatever size it
13065 likes. This is also the default.
13067 @item show trace-buffer-size
13068 @kindex show trace-buffer-size
13069 Show the current requested size for the trace buffer. Note that this
13070 will only match the actual size if the target supports size-setting,
13071 and was able to handle the requested size. For instance, if the
13072 target can only change buffer size between runs, this variable will
13073 not reflect the change until the next run starts. Use @code{tstatus}
13074 to get a report of the actual buffer size.
13078 @item set trace-user @var{text}
13079 @kindex set trace-user
13081 @item show trace-user
13082 @kindex show trace-user
13084 @item set trace-notes @var{text}
13085 @kindex set trace-notes
13086 Set the trace run's notes.
13088 @item show trace-notes
13089 @kindex show trace-notes
13090 Show the trace run's notes.
13092 @item set trace-stop-notes @var{text}
13093 @kindex set trace-stop-notes
13094 Set the trace run's stop notes. The handling of the note is as for
13095 @code{tstop} arguments; the set command is convenient way to fix a
13096 stop note that is mistaken or incomplete.
13098 @item show trace-stop-notes
13099 @kindex show trace-stop-notes
13100 Show the trace run's stop notes.
13104 @node Tracepoint Restrictions
13105 @subsection Tracepoint Restrictions
13107 @cindex tracepoint restrictions
13108 There are a number of restrictions on the use of tracepoints. As
13109 described above, tracepoint data gathering occurs on the target
13110 without interaction from @value{GDBN}. Thus the full capabilities of
13111 the debugger are not available during data gathering, and then at data
13112 examination time, you will be limited by only having what was
13113 collected. The following items describe some common problems, but it
13114 is not exhaustive, and you may run into additional difficulties not
13120 Tracepoint expressions are intended to gather objects (lvalues). Thus
13121 the full flexibility of GDB's expression evaluator is not available.
13122 You cannot call functions, cast objects to aggregate types, access
13123 convenience variables or modify values (except by assignment to trace
13124 state variables). Some language features may implicitly call
13125 functions (for instance Objective-C fields with accessors), and therefore
13126 cannot be collected either.
13129 Collection of local variables, either individually or in bulk with
13130 @code{$locals} or @code{$args}, during @code{while-stepping} may
13131 behave erratically. The stepping action may enter a new scope (for
13132 instance by stepping into a function), or the location of the variable
13133 may change (for instance it is loaded into a register). The
13134 tracepoint data recorded uses the location information for the
13135 variables that is correct for the tracepoint location. When the
13136 tracepoint is created, it is not possible, in general, to determine
13137 where the steps of a @code{while-stepping} sequence will advance the
13138 program---particularly if a conditional branch is stepped.
13141 Collection of an incompletely-initialized or partially-destroyed object
13142 may result in something that @value{GDBN} cannot display, or displays
13143 in a misleading way.
13146 When @value{GDBN} displays a pointer to character it automatically
13147 dereferences the pointer to also display characters of the string
13148 being pointed to. However, collecting the pointer during tracing does
13149 not automatically collect the string. You need to explicitly
13150 dereference the pointer and provide size information if you want to
13151 collect not only the pointer, but the memory pointed to. For example,
13152 @code{*ptr@@50} can be used to collect the 50 element array pointed to
13156 It is not possible to collect a complete stack backtrace at a
13157 tracepoint. Instead, you may collect the registers and a few hundred
13158 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
13159 (adjust to use the name of the actual stack pointer register on your
13160 target architecture, and the amount of stack you wish to capture).
13161 Then the @code{backtrace} command will show a partial backtrace when
13162 using a trace frame. The number of stack frames that can be examined
13163 depends on the sizes of the frames in the collected stack. Note that
13164 if you ask for a block so large that it goes past the bottom of the
13165 stack, the target agent may report an error trying to read from an
13169 If you do not collect registers at a tracepoint, @value{GDBN} can
13170 infer that the value of @code{$pc} must be the same as the address of
13171 the tracepoint and use that when you are looking at a trace frame
13172 for that tracepoint. However, this cannot work if the tracepoint has
13173 multiple locations (for instance if it was set in a function that was
13174 inlined), or if it has a @code{while-stepping} loop. In those cases
13175 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
13180 @node Analyze Collected Data
13181 @section Using the Collected Data
13183 After the tracepoint experiment ends, you use @value{GDBN} commands
13184 for examining the trace data. The basic idea is that each tracepoint
13185 collects a trace @dfn{snapshot} every time it is hit and another
13186 snapshot every time it single-steps. All these snapshots are
13187 consecutively numbered from zero and go into a buffer, and you can
13188 examine them later. The way you examine them is to @dfn{focus} on a
13189 specific trace snapshot. When the remote stub is focused on a trace
13190 snapshot, it will respond to all @value{GDBN} requests for memory and
13191 registers by reading from the buffer which belongs to that snapshot,
13192 rather than from @emph{real} memory or registers of the program being
13193 debugged. This means that @strong{all} @value{GDBN} commands
13194 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
13195 behave as if we were currently debugging the program state as it was
13196 when the tracepoint occurred. Any requests for data that are not in
13197 the buffer will fail.
13200 * tfind:: How to select a trace snapshot
13201 * tdump:: How to display all data for a snapshot
13202 * save tracepoints:: How to save tracepoints for a future run
13206 @subsection @code{tfind @var{n}}
13209 @cindex select trace snapshot
13210 @cindex find trace snapshot
13211 The basic command for selecting a trace snapshot from the buffer is
13212 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
13213 counting from zero. If no argument @var{n} is given, the next
13214 snapshot is selected.
13216 Here are the various forms of using the @code{tfind} command.
13220 Find the first snapshot in the buffer. This is a synonym for
13221 @code{tfind 0} (since 0 is the number of the first snapshot).
13224 Stop debugging trace snapshots, resume @emph{live} debugging.
13227 Same as @samp{tfind none}.
13230 No argument means find the next trace snapshot.
13233 Find the previous trace snapshot before the current one. This permits
13234 retracing earlier steps.
13236 @item tfind tracepoint @var{num}
13237 Find the next snapshot associated with tracepoint @var{num}. Search
13238 proceeds forward from the last examined trace snapshot. If no
13239 argument @var{num} is given, it means find the next snapshot collected
13240 for the same tracepoint as the current snapshot.
13242 @item tfind pc @var{addr}
13243 Find the next snapshot associated with the value @var{addr} of the
13244 program counter. Search proceeds forward from the last examined trace
13245 snapshot. If no argument @var{addr} is given, it means find the next
13246 snapshot with the same value of PC as the current snapshot.
13248 @item tfind outside @var{addr1}, @var{addr2}
13249 Find the next snapshot whose PC is outside the given range of
13250 addresses (exclusive).
13252 @item tfind range @var{addr1}, @var{addr2}
13253 Find the next snapshot whose PC is between @var{addr1} and
13254 @var{addr2} (inclusive).
13256 @item tfind line @r{[}@var{file}:@r{]}@var{n}
13257 Find the next snapshot associated with the source line @var{n}. If
13258 the optional argument @var{file} is given, refer to line @var{n} in
13259 that source file. Search proceeds forward from the last examined
13260 trace snapshot. If no argument @var{n} is given, it means find the
13261 next line other than the one currently being examined; thus saying
13262 @code{tfind line} repeatedly can appear to have the same effect as
13263 stepping from line to line in a @emph{live} debugging session.
13266 The default arguments for the @code{tfind} commands are specifically
13267 designed to make it easy to scan through the trace buffer. For
13268 instance, @code{tfind} with no argument selects the next trace
13269 snapshot, and @code{tfind -} with no argument selects the previous
13270 trace snapshot. So, by giving one @code{tfind} command, and then
13271 simply hitting @key{RET} repeatedly you can examine all the trace
13272 snapshots in order. Or, by saying @code{tfind -} and then hitting
13273 @key{RET} repeatedly you can examine the snapshots in reverse order.
13274 The @code{tfind line} command with no argument selects the snapshot
13275 for the next source line executed. The @code{tfind pc} command with
13276 no argument selects the next snapshot with the same program counter
13277 (PC) as the current frame. The @code{tfind tracepoint} command with
13278 no argument selects the next trace snapshot collected by the same
13279 tracepoint as the current one.
13281 In addition to letting you scan through the trace buffer manually,
13282 these commands make it easy to construct @value{GDBN} scripts that
13283 scan through the trace buffer and print out whatever collected data
13284 you are interested in. Thus, if we want to examine the PC, FP, and SP
13285 registers from each trace frame in the buffer, we can say this:
13288 (@value{GDBP}) @b{tfind start}
13289 (@value{GDBP}) @b{while ($trace_frame != -1)}
13290 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
13291 $trace_frame, $pc, $sp, $fp
13295 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
13296 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
13297 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
13298 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
13299 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
13300 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
13301 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
13302 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
13303 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
13304 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
13305 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
13308 Or, if we want to examine the variable @code{X} at each source line in
13312 (@value{GDBP}) @b{tfind start}
13313 (@value{GDBP}) @b{while ($trace_frame != -1)}
13314 > printf "Frame %d, X == %d\n", $trace_frame, X
13324 @subsection @code{tdump}
13326 @cindex dump all data collected at tracepoint
13327 @cindex tracepoint data, display
13329 This command takes no arguments. It prints all the data collected at
13330 the current trace snapshot.
13333 (@value{GDBP}) @b{trace 444}
13334 (@value{GDBP}) @b{actions}
13335 Enter actions for tracepoint #2, one per line:
13336 > collect $regs, $locals, $args, gdb_long_test
13339 (@value{GDBP}) @b{tstart}
13341 (@value{GDBP}) @b{tfind line 444}
13342 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
13344 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
13346 (@value{GDBP}) @b{tdump}
13347 Data collected at tracepoint 2, trace frame 1:
13348 d0 0xc4aa0085 -995491707
13352 d4 0x71aea3d 119204413
13355 d7 0x380035 3670069
13356 a0 0x19e24a 1696330
13357 a1 0x3000668 50333288
13359 a3 0x322000 3284992
13360 a4 0x3000698 50333336
13361 a5 0x1ad3cc 1758156
13362 fp 0x30bf3c 0x30bf3c
13363 sp 0x30bf34 0x30bf34
13365 pc 0x20b2c8 0x20b2c8
13369 p = 0x20e5b4 "gdb-test"
13376 gdb_long_test = 17 '\021'
13381 @code{tdump} works by scanning the tracepoint's current collection
13382 actions and printing the value of each expression listed. So
13383 @code{tdump} can fail, if after a run, you change the tracepoint's
13384 actions to mention variables that were not collected during the run.
13386 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
13387 uses the collected value of @code{$pc} to distinguish between trace
13388 frames that were collected at the tracepoint hit, and frames that were
13389 collected while stepping. This allows it to correctly choose whether
13390 to display the basic list of collections, or the collections from the
13391 body of the while-stepping loop. However, if @code{$pc} was not collected,
13392 then @code{tdump} will always attempt to dump using the basic collection
13393 list, and may fail if a while-stepping frame does not include all the
13394 same data that is collected at the tracepoint hit.
13395 @c This is getting pretty arcane, example would be good.
13397 @node save tracepoints
13398 @subsection @code{save tracepoints @var{filename}}
13399 @kindex save tracepoints
13400 @kindex save-tracepoints
13401 @cindex save tracepoints for future sessions
13403 This command saves all current tracepoint definitions together with
13404 their actions and passcounts, into a file @file{@var{filename}}
13405 suitable for use in a later debugging session. To read the saved
13406 tracepoint definitions, use the @code{source} command (@pxref{Command
13407 Files}). The @w{@code{save-tracepoints}} command is a deprecated
13408 alias for @w{@code{save tracepoints}}
13410 @node Tracepoint Variables
13411 @section Convenience Variables for Tracepoints
13412 @cindex tracepoint variables
13413 @cindex convenience variables for tracepoints
13416 @vindex $trace_frame
13417 @item (int) $trace_frame
13418 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
13419 snapshot is selected.
13421 @vindex $tracepoint
13422 @item (int) $tracepoint
13423 The tracepoint for the current trace snapshot.
13425 @vindex $trace_line
13426 @item (int) $trace_line
13427 The line number for the current trace snapshot.
13429 @vindex $trace_file
13430 @item (char []) $trace_file
13431 The source file for the current trace snapshot.
13433 @vindex $trace_func
13434 @item (char []) $trace_func
13435 The name of the function containing @code{$tracepoint}.
13438 Note: @code{$trace_file} is not suitable for use in @code{printf},
13439 use @code{output} instead.
13441 Here's a simple example of using these convenience variables for
13442 stepping through all the trace snapshots and printing some of their
13443 data. Note that these are not the same as trace state variables,
13444 which are managed by the target.
13447 (@value{GDBP}) @b{tfind start}
13449 (@value{GDBP}) @b{while $trace_frame != -1}
13450 > output $trace_file
13451 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
13457 @section Using Trace Files
13458 @cindex trace files
13460 In some situations, the target running a trace experiment may no
13461 longer be available; perhaps it crashed, or the hardware was needed
13462 for a different activity. To handle these cases, you can arrange to
13463 dump the trace data into a file, and later use that file as a source
13464 of trace data, via the @code{target tfile} command.
13469 @item tsave [ -r ] @var{filename}
13470 @itemx tsave [-ctf] @var{dirname}
13471 Save the trace data to @var{filename}. By default, this command
13472 assumes that @var{filename} refers to the host filesystem, so if
13473 necessary @value{GDBN} will copy raw trace data up from the target and
13474 then save it. If the target supports it, you can also supply the
13475 optional argument @code{-r} (``remote'') to direct the target to save
13476 the data directly into @var{filename} in its own filesystem, which may be
13477 more efficient if the trace buffer is very large. (Note, however, that
13478 @code{target tfile} can only read from files accessible to the host.)
13479 By default, this command will save trace frame in tfile format.
13480 You can supply the optional argument @code{-ctf} to save date in CTF
13481 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
13482 that can be shared by multiple debugging and tracing tools. Please go to
13483 @indicateurl{http://www.efficios.com/ctf} to get more information.
13485 @kindex target tfile
13489 @item target tfile @var{filename}
13490 @itemx target ctf @var{dirname}
13491 Use the file named @var{filename} or directory named @var{dirname} as
13492 a source of trace data. Commands that examine data work as they do with
13493 a live target, but it is not possible to run any new trace experiments.
13494 @code{tstatus} will report the state of the trace run at the moment
13495 the data was saved, as well as the current trace frame you are examining.
13496 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
13500 (@value{GDBP}) target ctf ctf.ctf
13501 (@value{GDBP}) tfind
13502 Found trace frame 0, tracepoint 2
13503 39 ++a; /* set tracepoint 1 here */
13504 (@value{GDBP}) tdump
13505 Data collected at tracepoint 2, trace frame 0:
13509 c = @{"123", "456", "789", "123", "456", "789"@}
13510 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
13518 @chapter Debugging Programs That Use Overlays
13521 If your program is too large to fit completely in your target system's
13522 memory, you can sometimes use @dfn{overlays} to work around this
13523 problem. @value{GDBN} provides some support for debugging programs that
13527 * How Overlays Work:: A general explanation of overlays.
13528 * Overlay Commands:: Managing overlays in @value{GDBN}.
13529 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
13530 mapped by asking the inferior.
13531 * Overlay Sample Program:: A sample program using overlays.
13534 @node How Overlays Work
13535 @section How Overlays Work
13536 @cindex mapped overlays
13537 @cindex unmapped overlays
13538 @cindex load address, overlay's
13539 @cindex mapped address
13540 @cindex overlay area
13542 Suppose you have a computer whose instruction address space is only 64
13543 kilobytes long, but which has much more memory which can be accessed by
13544 other means: special instructions, segment registers, or memory
13545 management hardware, for example. Suppose further that you want to
13546 adapt a program which is larger than 64 kilobytes to run on this system.
13548 One solution is to identify modules of your program which are relatively
13549 independent, and need not call each other directly; call these modules
13550 @dfn{overlays}. Separate the overlays from the main program, and place
13551 their machine code in the larger memory. Place your main program in
13552 instruction memory, but leave at least enough space there to hold the
13553 largest overlay as well.
13555 Now, to call a function located in an overlay, you must first copy that
13556 overlay's machine code from the large memory into the space set aside
13557 for it in the instruction memory, and then jump to its entry point
13560 @c NB: In the below the mapped area's size is greater or equal to the
13561 @c size of all overlays. This is intentional to remind the developer
13562 @c that overlays don't necessarily need to be the same size.
13566 Data Instruction Larger
13567 Address Space Address Space Address Space
13568 +-----------+ +-----------+ +-----------+
13570 +-----------+ +-----------+ +-----------+<-- overlay 1
13571 | program | | main | .----| overlay 1 | load address
13572 | variables | | program | | +-----------+
13573 | and heap | | | | | |
13574 +-----------+ | | | +-----------+<-- overlay 2
13575 | | +-----------+ | | | load address
13576 +-----------+ | | | .-| overlay 2 |
13578 mapped --->+-----------+ | | +-----------+
13579 address | | | | | |
13580 | overlay | <-' | | |
13581 | area | <---' +-----------+<-- overlay 3
13582 | | <---. | | load address
13583 +-----------+ `--| overlay 3 |
13590 @anchor{A code overlay}A code overlay
13594 The diagram (@pxref{A code overlay}) shows a system with separate data
13595 and instruction address spaces. To map an overlay, the program copies
13596 its code from the larger address space to the instruction address space.
13597 Since the overlays shown here all use the same mapped address, only one
13598 may be mapped at a time. For a system with a single address space for
13599 data and instructions, the diagram would be similar, except that the
13600 program variables and heap would share an address space with the main
13601 program and the overlay area.
13603 An overlay loaded into instruction memory and ready for use is called a
13604 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
13605 instruction memory. An overlay not present (or only partially present)
13606 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
13607 is its address in the larger memory. The mapped address is also called
13608 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
13609 called the @dfn{load memory address}, or @dfn{LMA}.
13611 Unfortunately, overlays are not a completely transparent way to adapt a
13612 program to limited instruction memory. They introduce a new set of
13613 global constraints you must keep in mind as you design your program:
13618 Before calling or returning to a function in an overlay, your program
13619 must make sure that overlay is actually mapped. Otherwise, the call or
13620 return will transfer control to the right address, but in the wrong
13621 overlay, and your program will probably crash.
13624 If the process of mapping an overlay is expensive on your system, you
13625 will need to choose your overlays carefully to minimize their effect on
13626 your program's performance.
13629 The executable file you load onto your system must contain each
13630 overlay's instructions, appearing at the overlay's load address, not its
13631 mapped address. However, each overlay's instructions must be relocated
13632 and its symbols defined as if the overlay were at its mapped address.
13633 You can use GNU linker scripts to specify different load and relocation
13634 addresses for pieces of your program; see @ref{Overlay Description,,,
13635 ld.info, Using ld: the GNU linker}.
13638 The procedure for loading executable files onto your system must be able
13639 to load their contents into the larger address space as well as the
13640 instruction and data spaces.
13644 The overlay system described above is rather simple, and could be
13645 improved in many ways:
13650 If your system has suitable bank switch registers or memory management
13651 hardware, you could use those facilities to make an overlay's load area
13652 contents simply appear at their mapped address in instruction space.
13653 This would probably be faster than copying the overlay to its mapped
13654 area in the usual way.
13657 If your overlays are small enough, you could set aside more than one
13658 overlay area, and have more than one overlay mapped at a time.
13661 You can use overlays to manage data, as well as instructions. In
13662 general, data overlays are even less transparent to your design than
13663 code overlays: whereas code overlays only require care when you call or
13664 return to functions, data overlays require care every time you access
13665 the data. Also, if you change the contents of a data overlay, you
13666 must copy its contents back out to its load address before you can copy a
13667 different data overlay into the same mapped area.
13672 @node Overlay Commands
13673 @section Overlay Commands
13675 To use @value{GDBN}'s overlay support, each overlay in your program must
13676 correspond to a separate section of the executable file. The section's
13677 virtual memory address and load memory address must be the overlay's
13678 mapped and load addresses. Identifying overlays with sections allows
13679 @value{GDBN} to determine the appropriate address of a function or
13680 variable, depending on whether the overlay is mapped or not.
13682 @value{GDBN}'s overlay commands all start with the word @code{overlay};
13683 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
13688 Disable @value{GDBN}'s overlay support. When overlay support is
13689 disabled, @value{GDBN} assumes that all functions and variables are
13690 always present at their mapped addresses. By default, @value{GDBN}'s
13691 overlay support is disabled.
13693 @item overlay manual
13694 @cindex manual overlay debugging
13695 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
13696 relies on you to tell it which overlays are mapped, and which are not,
13697 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
13698 commands described below.
13700 @item overlay map-overlay @var{overlay}
13701 @itemx overlay map @var{overlay}
13702 @cindex map an overlay
13703 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
13704 be the name of the object file section containing the overlay. When an
13705 overlay is mapped, @value{GDBN} assumes it can find the overlay's
13706 functions and variables at their mapped addresses. @value{GDBN} assumes
13707 that any other overlays whose mapped ranges overlap that of
13708 @var{overlay} are now unmapped.
13710 @item overlay unmap-overlay @var{overlay}
13711 @itemx overlay unmap @var{overlay}
13712 @cindex unmap an overlay
13713 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
13714 must be the name of the object file section containing the overlay.
13715 When an overlay is unmapped, @value{GDBN} assumes it can find the
13716 overlay's functions and variables at their load addresses.
13719 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
13720 consults a data structure the overlay manager maintains in the inferior
13721 to see which overlays are mapped. For details, see @ref{Automatic
13722 Overlay Debugging}.
13724 @item overlay load-target
13725 @itemx overlay load
13726 @cindex reloading the overlay table
13727 Re-read the overlay table from the inferior. Normally, @value{GDBN}
13728 re-reads the table @value{GDBN} automatically each time the inferior
13729 stops, so this command should only be necessary if you have changed the
13730 overlay mapping yourself using @value{GDBN}. This command is only
13731 useful when using automatic overlay debugging.
13733 @item overlay list-overlays
13734 @itemx overlay list
13735 @cindex listing mapped overlays
13736 Display a list of the overlays currently mapped, along with their mapped
13737 addresses, load addresses, and sizes.
13741 Normally, when @value{GDBN} prints a code address, it includes the name
13742 of the function the address falls in:
13745 (@value{GDBP}) print main
13746 $3 = @{int ()@} 0x11a0 <main>
13749 When overlay debugging is enabled, @value{GDBN} recognizes code in
13750 unmapped overlays, and prints the names of unmapped functions with
13751 asterisks around them. For example, if @code{foo} is a function in an
13752 unmapped overlay, @value{GDBN} prints it this way:
13755 (@value{GDBP}) overlay list
13756 No sections are mapped.
13757 (@value{GDBP}) print foo
13758 $5 = @{int (int)@} 0x100000 <*foo*>
13761 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
13765 (@value{GDBP}) overlay list
13766 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
13767 mapped at 0x1016 - 0x104a
13768 (@value{GDBP}) print foo
13769 $6 = @{int (int)@} 0x1016 <foo>
13772 When overlay debugging is enabled, @value{GDBN} can find the correct
13773 address for functions and variables in an overlay, whether or not the
13774 overlay is mapped. This allows most @value{GDBN} commands, like
13775 @code{break} and @code{disassemble}, to work normally, even on unmapped
13776 code. However, @value{GDBN}'s breakpoint support has some limitations:
13780 @cindex breakpoints in overlays
13781 @cindex overlays, setting breakpoints in
13782 You can set breakpoints in functions in unmapped overlays, as long as
13783 @value{GDBN} can write to the overlay at its load address.
13785 @value{GDBN} can not set hardware or simulator-based breakpoints in
13786 unmapped overlays. However, if you set a breakpoint at the end of your
13787 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
13788 you are using manual overlay management), @value{GDBN} will re-set its
13789 breakpoints properly.
13793 @node Automatic Overlay Debugging
13794 @section Automatic Overlay Debugging
13795 @cindex automatic overlay debugging
13797 @value{GDBN} can automatically track which overlays are mapped and which
13798 are not, given some simple co-operation from the overlay manager in the
13799 inferior. If you enable automatic overlay debugging with the
13800 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
13801 looks in the inferior's memory for certain variables describing the
13802 current state of the overlays.
13804 Here are the variables your overlay manager must define to support
13805 @value{GDBN}'s automatic overlay debugging:
13809 @item @code{_ovly_table}:
13810 This variable must be an array of the following structures:
13815 /* The overlay's mapped address. */
13818 /* The size of the overlay, in bytes. */
13819 unsigned long size;
13821 /* The overlay's load address. */
13824 /* Non-zero if the overlay is currently mapped;
13826 unsigned long mapped;
13830 @item @code{_novlys}:
13831 This variable must be a four-byte signed integer, holding the total
13832 number of elements in @code{_ovly_table}.
13836 To decide whether a particular overlay is mapped or not, @value{GDBN}
13837 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
13838 @code{lma} members equal the VMA and LMA of the overlay's section in the
13839 executable file. When @value{GDBN} finds a matching entry, it consults
13840 the entry's @code{mapped} member to determine whether the overlay is
13843 In addition, your overlay manager may define a function called
13844 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
13845 will silently set a breakpoint there. If the overlay manager then
13846 calls this function whenever it has changed the overlay table, this
13847 will enable @value{GDBN} to accurately keep track of which overlays
13848 are in program memory, and update any breakpoints that may be set
13849 in overlays. This will allow breakpoints to work even if the
13850 overlays are kept in ROM or other non-writable memory while they
13851 are not being executed.
13853 @node Overlay Sample Program
13854 @section Overlay Sample Program
13855 @cindex overlay example program
13857 When linking a program which uses overlays, you must place the overlays
13858 at their load addresses, while relocating them to run at their mapped
13859 addresses. To do this, you must write a linker script (@pxref{Overlay
13860 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
13861 since linker scripts are specific to a particular host system, target
13862 architecture, and target memory layout, this manual cannot provide
13863 portable sample code demonstrating @value{GDBN}'s overlay support.
13865 However, the @value{GDBN} source distribution does contain an overlaid
13866 program, with linker scripts for a few systems, as part of its test
13867 suite. The program consists of the following files from
13868 @file{gdb/testsuite/gdb.base}:
13872 The main program file.
13874 A simple overlay manager, used by @file{overlays.c}.
13879 Overlay modules, loaded and used by @file{overlays.c}.
13882 Linker scripts for linking the test program on the @code{d10v-elf}
13883 and @code{m32r-elf} targets.
13886 You can build the test program using the @code{d10v-elf} GCC
13887 cross-compiler like this:
13890 $ d10v-elf-gcc -g -c overlays.c
13891 $ d10v-elf-gcc -g -c ovlymgr.c
13892 $ d10v-elf-gcc -g -c foo.c
13893 $ d10v-elf-gcc -g -c bar.c
13894 $ d10v-elf-gcc -g -c baz.c
13895 $ d10v-elf-gcc -g -c grbx.c
13896 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
13897 baz.o grbx.o -Wl,-Td10v.ld -o overlays
13900 The build process is identical for any other architecture, except that
13901 you must substitute the appropriate compiler and linker script for the
13902 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
13906 @chapter Using @value{GDBN} with Different Languages
13909 Although programming languages generally have common aspects, they are
13910 rarely expressed in the same manner. For instance, in ANSI C,
13911 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
13912 Modula-2, it is accomplished by @code{p^}. Values can also be
13913 represented (and displayed) differently. Hex numbers in C appear as
13914 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
13916 @cindex working language
13917 Language-specific information is built into @value{GDBN} for some languages,
13918 allowing you to express operations like the above in your program's
13919 native language, and allowing @value{GDBN} to output values in a manner
13920 consistent with the syntax of your program's native language. The
13921 language you use to build expressions is called the @dfn{working
13925 * Setting:: Switching between source languages
13926 * Show:: Displaying the language
13927 * Checks:: Type and range checks
13928 * Supported Languages:: Supported languages
13929 * Unsupported Languages:: Unsupported languages
13933 @section Switching Between Source Languages
13935 There are two ways to control the working language---either have @value{GDBN}
13936 set it automatically, or select it manually yourself. You can use the
13937 @code{set language} command for either purpose. On startup, @value{GDBN}
13938 defaults to setting the language automatically. The working language is
13939 used to determine how expressions you type are interpreted, how values
13942 In addition to the working language, every source file that
13943 @value{GDBN} knows about has its own working language. For some object
13944 file formats, the compiler might indicate which language a particular
13945 source file is in. However, most of the time @value{GDBN} infers the
13946 language from the name of the file. The language of a source file
13947 controls whether C@t{++} names are demangled---this way @code{backtrace} can
13948 show each frame appropriately for its own language. There is no way to
13949 set the language of a source file from within @value{GDBN}, but you can
13950 set the language associated with a filename extension. @xref{Show, ,
13951 Displaying the Language}.
13953 This is most commonly a problem when you use a program, such
13954 as @code{cfront} or @code{f2c}, that generates C but is written in
13955 another language. In that case, make the
13956 program use @code{#line} directives in its C output; that way
13957 @value{GDBN} will know the correct language of the source code of the original
13958 program, and will display that source code, not the generated C code.
13961 * Filenames:: Filename extensions and languages.
13962 * Manually:: Setting the working language manually
13963 * Automatically:: Having @value{GDBN} infer the source language
13967 @subsection List of Filename Extensions and Languages
13969 If a source file name ends in one of the following extensions, then
13970 @value{GDBN} infers that its language is the one indicated.
13988 C@t{++} source file
13994 Objective-C source file
13998 Fortran source file
14001 Modula-2 source file
14005 Assembler source file. This actually behaves almost like C, but
14006 @value{GDBN} does not skip over function prologues when stepping.
14009 In addition, you may set the language associated with a filename
14010 extension. @xref{Show, , Displaying the Language}.
14013 @subsection Setting the Working Language
14015 If you allow @value{GDBN} to set the language automatically,
14016 expressions are interpreted the same way in your debugging session and
14019 @kindex set language
14020 If you wish, you may set the language manually. To do this, issue the
14021 command @samp{set language @var{lang}}, where @var{lang} is the name of
14022 a language, such as
14023 @code{c} or @code{modula-2}.
14024 For a list of the supported languages, type @samp{set language}.
14026 Setting the language manually prevents @value{GDBN} from updating the working
14027 language automatically. This can lead to confusion if you try
14028 to debug a program when the working language is not the same as the
14029 source language, when an expression is acceptable to both
14030 languages---but means different things. For instance, if the current
14031 source file were written in C, and @value{GDBN} was parsing Modula-2, a
14039 might not have the effect you intended. In C, this means to add
14040 @code{b} and @code{c} and place the result in @code{a}. The result
14041 printed would be the value of @code{a}. In Modula-2, this means to compare
14042 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
14044 @node Automatically
14045 @subsection Having @value{GDBN} Infer the Source Language
14047 To have @value{GDBN} set the working language automatically, use
14048 @samp{set language local} or @samp{set language auto}. @value{GDBN}
14049 then infers the working language. That is, when your program stops in a
14050 frame (usually by encountering a breakpoint), @value{GDBN} sets the
14051 working language to the language recorded for the function in that
14052 frame. If the language for a frame is unknown (that is, if the function
14053 or block corresponding to the frame was defined in a source file that
14054 does not have a recognized extension), the current working language is
14055 not changed, and @value{GDBN} issues a warning.
14057 This may not seem necessary for most programs, which are written
14058 entirely in one source language. However, program modules and libraries
14059 written in one source language can be used by a main program written in
14060 a different source language. Using @samp{set language auto} in this
14061 case frees you from having to set the working language manually.
14064 @section Displaying the Language
14066 The following commands help you find out which language is the
14067 working language, and also what language source files were written in.
14070 @item show language
14071 @anchor{show language}
14072 @kindex show language
14073 Display the current working language. This is the
14074 language you can use with commands such as @code{print} to
14075 build and compute expressions that may involve variables in your program.
14078 @kindex info frame@r{, show the source language}
14079 Display the source language for this frame. This language becomes the
14080 working language if you use an identifier from this frame.
14081 @xref{Frame Info, ,Information about a Frame}, to identify the other
14082 information listed here.
14085 @kindex info source@r{, show the source language}
14086 Display the source language of this source file.
14087 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
14088 information listed here.
14091 In unusual circumstances, you may have source files with extensions
14092 not in the standard list. You can then set the extension associated
14093 with a language explicitly:
14096 @item set extension-language @var{ext} @var{language}
14097 @kindex set extension-language
14098 Tell @value{GDBN} that source files with extension @var{ext} are to be
14099 assumed as written in the source language @var{language}.
14101 @item info extensions
14102 @kindex info extensions
14103 List all the filename extensions and the associated languages.
14107 @section Type and Range Checking
14109 Some languages are designed to guard you against making seemingly common
14110 errors through a series of compile- and run-time checks. These include
14111 checking the type of arguments to functions and operators and making
14112 sure mathematical overflows are caught at run time. Checks such as
14113 these help to ensure a program's correctness once it has been compiled
14114 by eliminating type mismatches and providing active checks for range
14115 errors when your program is running.
14117 By default @value{GDBN} checks for these errors according to the
14118 rules of the current source language. Although @value{GDBN} does not check
14119 the statements in your program, it can check expressions entered directly
14120 into @value{GDBN} for evaluation via the @code{print} command, for example.
14123 * Type Checking:: An overview of type checking
14124 * Range Checking:: An overview of range checking
14127 @cindex type checking
14128 @cindex checks, type
14129 @node Type Checking
14130 @subsection An Overview of Type Checking
14132 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
14133 arguments to operators and functions have to be of the correct type,
14134 otherwise an error occurs. These checks prevent type mismatch
14135 errors from ever causing any run-time problems. For example,
14138 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
14140 (@value{GDBP}) print obj.my_method (0)
14143 (@value{GDBP}) print obj.my_method (0x1234)
14144 Cannot resolve method klass::my_method to any overloaded instance
14147 The second example fails because in C@t{++} the integer constant
14148 @samp{0x1234} is not type-compatible with the pointer parameter type.
14150 For the expressions you use in @value{GDBN} commands, you can tell
14151 @value{GDBN} to not enforce strict type checking or
14152 to treat any mismatches as errors and abandon the expression;
14153 When type checking is disabled, @value{GDBN} successfully evaluates
14154 expressions like the second example above.
14156 Even if type checking is off, there may be other reasons
14157 related to type that prevent @value{GDBN} from evaluating an expression.
14158 For instance, @value{GDBN} does not know how to add an @code{int} and
14159 a @code{struct foo}. These particular type errors have nothing to do
14160 with the language in use and usually arise from expressions which make
14161 little sense to evaluate anyway.
14163 @value{GDBN} provides some additional commands for controlling type checking:
14165 @kindex set check type
14166 @kindex show check type
14168 @item set check type on
14169 @itemx set check type off
14170 Set strict type checking on or off. If any type mismatches occur in
14171 evaluating an expression while type checking is on, @value{GDBN} prints a
14172 message and aborts evaluation of the expression.
14174 @item show check type
14175 Show the current setting of type checking and whether @value{GDBN}
14176 is enforcing strict type checking rules.
14179 @cindex range checking
14180 @cindex checks, range
14181 @node Range Checking
14182 @subsection An Overview of Range Checking
14184 In some languages (such as Modula-2), it is an error to exceed the
14185 bounds of a type; this is enforced with run-time checks. Such range
14186 checking is meant to ensure program correctness by making sure
14187 computations do not overflow, or indices on an array element access do
14188 not exceed the bounds of the array.
14190 For expressions you use in @value{GDBN} commands, you can tell
14191 @value{GDBN} to treat range errors in one of three ways: ignore them,
14192 always treat them as errors and abandon the expression, or issue
14193 warnings but evaluate the expression anyway.
14195 A range error can result from numerical overflow, from exceeding an
14196 array index bound, or when you type a constant that is not a member
14197 of any type. Some languages, however, do not treat overflows as an
14198 error. In many implementations of C, mathematical overflow causes the
14199 result to ``wrap around'' to lower values---for example, if @var{m} is
14200 the largest integer value, and @var{s} is the smallest, then
14203 @var{m} + 1 @result{} @var{s}
14206 This, too, is specific to individual languages, and in some cases
14207 specific to individual compilers or machines. @xref{Supported Languages, ,
14208 Supported Languages}, for further details on specific languages.
14210 @value{GDBN} provides some additional commands for controlling the range checker:
14212 @kindex set check range
14213 @kindex show check range
14215 @item set check range auto
14216 Set range checking on or off based on the current working language.
14217 @xref{Supported Languages, ,Supported Languages}, for the default settings for
14220 @item set check range on
14221 @itemx set check range off
14222 Set range checking on or off, overriding the default setting for the
14223 current working language. A warning is issued if the setting does not
14224 match the language default. If a range error occurs and range checking is on,
14225 then a message is printed and evaluation of the expression is aborted.
14227 @item set check range warn
14228 Output messages when the @value{GDBN} range checker detects a range error,
14229 but attempt to evaluate the expression anyway. Evaluating the
14230 expression may still be impossible for other reasons, such as accessing
14231 memory that the process does not own (a typical example from many Unix
14235 Show the current setting of the range checker, and whether or not it is
14236 being set automatically by @value{GDBN}.
14239 @node Supported Languages
14240 @section Supported Languages
14242 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran, Java,
14243 OpenCL C, Pascal, assembly, Modula-2, and Ada.
14244 @c This is false ...
14245 Some @value{GDBN} features may be used in expressions regardless of the
14246 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
14247 and the @samp{@{type@}addr} construct (@pxref{Expressions,
14248 ,Expressions}) can be used with the constructs of any supported
14251 The following sections detail to what degree each source language is
14252 supported by @value{GDBN}. These sections are not meant to be language
14253 tutorials or references, but serve only as a reference guide to what the
14254 @value{GDBN} expression parser accepts, and what input and output
14255 formats should look like for different languages. There are many good
14256 books written on each of these languages; please look to these for a
14257 language reference or tutorial.
14260 * C:: C and C@t{++}
14263 * Objective-C:: Objective-C
14264 * OpenCL C:: OpenCL C
14265 * Fortran:: Fortran
14267 * Modula-2:: Modula-2
14272 @subsection C and C@t{++}
14274 @cindex C and C@t{++}
14275 @cindex expressions in C or C@t{++}
14277 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
14278 to both languages. Whenever this is the case, we discuss those languages
14282 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
14283 @cindex @sc{gnu} C@t{++}
14284 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
14285 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
14286 effectively, you must compile your C@t{++} programs with a supported
14287 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
14288 compiler (@code{aCC}).
14291 * C Operators:: C and C@t{++} operators
14292 * C Constants:: C and C@t{++} constants
14293 * C Plus Plus Expressions:: C@t{++} expressions
14294 * C Defaults:: Default settings for C and C@t{++}
14295 * C Checks:: C and C@t{++} type and range checks
14296 * Debugging C:: @value{GDBN} and C
14297 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
14298 * Decimal Floating Point:: Numbers in Decimal Floating Point format
14302 @subsubsection C and C@t{++} Operators
14304 @cindex C and C@t{++} operators
14306 Operators must be defined on values of specific types. For instance,
14307 @code{+} is defined on numbers, but not on structures. Operators are
14308 often defined on groups of types.
14310 For the purposes of C and C@t{++}, the following definitions hold:
14315 @emph{Integral types} include @code{int} with any of its storage-class
14316 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
14319 @emph{Floating-point types} include @code{float}, @code{double}, and
14320 @code{long double} (if supported by the target platform).
14323 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
14326 @emph{Scalar types} include all of the above.
14331 The following operators are supported. They are listed here
14332 in order of increasing precedence:
14336 The comma or sequencing operator. Expressions in a comma-separated list
14337 are evaluated from left to right, with the result of the entire
14338 expression being the last expression evaluated.
14341 Assignment. The value of an assignment expression is the value
14342 assigned. Defined on scalar types.
14345 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
14346 and translated to @w{@code{@var{a} = @var{a op b}}}.
14347 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
14348 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
14349 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
14352 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
14353 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
14354 should be of an integral type.
14357 Logical @sc{or}. Defined on integral types.
14360 Logical @sc{and}. Defined on integral types.
14363 Bitwise @sc{or}. Defined on integral types.
14366 Bitwise exclusive-@sc{or}. Defined on integral types.
14369 Bitwise @sc{and}. Defined on integral types.
14372 Equality and inequality. Defined on scalar types. The value of these
14373 expressions is 0 for false and non-zero for true.
14375 @item <@r{, }>@r{, }<=@r{, }>=
14376 Less than, greater than, less than or equal, greater than or equal.
14377 Defined on scalar types. The value of these expressions is 0 for false
14378 and non-zero for true.
14381 left shift, and right shift. Defined on integral types.
14384 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
14387 Addition and subtraction. Defined on integral types, floating-point types and
14390 @item *@r{, }/@r{, }%
14391 Multiplication, division, and modulus. Multiplication and division are
14392 defined on integral and floating-point types. Modulus is defined on
14396 Increment and decrement. When appearing before a variable, the
14397 operation is performed before the variable is used in an expression;
14398 when appearing after it, the variable's value is used before the
14399 operation takes place.
14402 Pointer dereferencing. Defined on pointer types. Same precedence as
14406 Address operator. Defined on variables. Same precedence as @code{++}.
14408 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
14409 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
14410 to examine the address
14411 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
14415 Negative. Defined on integral and floating-point types. Same
14416 precedence as @code{++}.
14419 Logical negation. Defined on integral types. Same precedence as
14423 Bitwise complement operator. Defined on integral types. Same precedence as
14428 Structure member, and pointer-to-structure member. For convenience,
14429 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
14430 pointer based on the stored type information.
14431 Defined on @code{struct} and @code{union} data.
14434 Dereferences of pointers to members.
14437 Array indexing. @code{@var{a}[@var{i}]} is defined as
14438 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
14441 Function parameter list. Same precedence as @code{->}.
14444 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
14445 and @code{class} types.
14448 Doubled colons also represent the @value{GDBN} scope operator
14449 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
14453 If an operator is redefined in the user code, @value{GDBN} usually
14454 attempts to invoke the redefined version instead of using the operator's
14455 predefined meaning.
14458 @subsubsection C and C@t{++} Constants
14460 @cindex C and C@t{++} constants
14462 @value{GDBN} allows you to express the constants of C and C@t{++} in the
14467 Integer constants are a sequence of digits. Octal constants are
14468 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
14469 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
14470 @samp{l}, specifying that the constant should be treated as a
14474 Floating point constants are a sequence of digits, followed by a decimal
14475 point, followed by a sequence of digits, and optionally followed by an
14476 exponent. An exponent is of the form:
14477 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
14478 sequence of digits. The @samp{+} is optional for positive exponents.
14479 A floating-point constant may also end with a letter @samp{f} or
14480 @samp{F}, specifying that the constant should be treated as being of
14481 the @code{float} (as opposed to the default @code{double}) type; or with
14482 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
14486 Enumerated constants consist of enumerated identifiers, or their
14487 integral equivalents.
14490 Character constants are a single character surrounded by single quotes
14491 (@code{'}), or a number---the ordinal value of the corresponding character
14492 (usually its @sc{ascii} value). Within quotes, the single character may
14493 be represented by a letter or by @dfn{escape sequences}, which are of
14494 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
14495 of the character's ordinal value; or of the form @samp{\@var{x}}, where
14496 @samp{@var{x}} is a predefined special character---for example,
14497 @samp{\n} for newline.
14499 Wide character constants can be written by prefixing a character
14500 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
14501 form of @samp{x}. The target wide character set is used when
14502 computing the value of this constant (@pxref{Character Sets}).
14505 String constants are a sequence of character constants surrounded by
14506 double quotes (@code{"}). Any valid character constant (as described
14507 above) may appear. Double quotes within the string must be preceded by
14508 a backslash, so for instance @samp{"a\"b'c"} is a string of five
14511 Wide string constants can be written by prefixing a string constant
14512 with @samp{L}, as in C. The target wide character set is used when
14513 computing the value of this constant (@pxref{Character Sets}).
14516 Pointer constants are an integral value. You can also write pointers
14517 to constants using the C operator @samp{&}.
14520 Array constants are comma-separated lists surrounded by braces @samp{@{}
14521 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
14522 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
14523 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
14526 @node C Plus Plus Expressions
14527 @subsubsection C@t{++} Expressions
14529 @cindex expressions in C@t{++}
14530 @value{GDBN} expression handling can interpret most C@t{++} expressions.
14532 @cindex debugging C@t{++} programs
14533 @cindex C@t{++} compilers
14534 @cindex debug formats and C@t{++}
14535 @cindex @value{NGCC} and C@t{++}
14537 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
14538 the proper compiler and the proper debug format. Currently,
14539 @value{GDBN} works best when debugging C@t{++} code that is compiled
14540 with the most recent version of @value{NGCC} possible. The DWARF
14541 debugging format is preferred; @value{NGCC} defaults to this on most
14542 popular platforms. Other compilers and/or debug formats are likely to
14543 work badly or not at all when using @value{GDBN} to debug C@t{++}
14544 code. @xref{Compilation}.
14549 @cindex member functions
14551 Member function calls are allowed; you can use expressions like
14554 count = aml->GetOriginal(x, y)
14557 @vindex this@r{, inside C@t{++} member functions}
14558 @cindex namespace in C@t{++}
14560 While a member function is active (in the selected stack frame), your
14561 expressions have the same namespace available as the member function;
14562 that is, @value{GDBN} allows implicit references to the class instance
14563 pointer @code{this} following the same rules as C@t{++}. @code{using}
14564 declarations in the current scope are also respected by @value{GDBN}.
14566 @cindex call overloaded functions
14567 @cindex overloaded functions, calling
14568 @cindex type conversions in C@t{++}
14570 You can call overloaded functions; @value{GDBN} resolves the function
14571 call to the right definition, with some restrictions. @value{GDBN} does not
14572 perform overload resolution involving user-defined type conversions,
14573 calls to constructors, or instantiations of templates that do not exist
14574 in the program. It also cannot handle ellipsis argument lists or
14577 It does perform integral conversions and promotions, floating-point
14578 promotions, arithmetic conversions, pointer conversions, conversions of
14579 class objects to base classes, and standard conversions such as those of
14580 functions or arrays to pointers; it requires an exact match on the
14581 number of function arguments.
14583 Overload resolution is always performed, unless you have specified
14584 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
14585 ,@value{GDBN} Features for C@t{++}}.
14587 You must specify @code{set overload-resolution off} in order to use an
14588 explicit function signature to call an overloaded function, as in
14590 p 'foo(char,int)'('x', 13)
14593 The @value{GDBN} command-completion facility can simplify this;
14594 see @ref{Completion, ,Command Completion}.
14596 @cindex reference declarations
14598 @value{GDBN} understands variables declared as C@t{++} references; you can use
14599 them in expressions just as you do in C@t{++} source---they are automatically
14602 In the parameter list shown when @value{GDBN} displays a frame, the values of
14603 reference variables are not displayed (unlike other variables); this
14604 avoids clutter, since references are often used for large structures.
14605 The @emph{address} of a reference variable is always shown, unless
14606 you have specified @samp{set print address off}.
14609 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
14610 expressions can use it just as expressions in your program do. Since
14611 one scope may be defined in another, you can use @code{::} repeatedly if
14612 necessary, for example in an expression like
14613 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
14614 resolving name scope by reference to source files, in both C and C@t{++}
14615 debugging (@pxref{Variables, ,Program Variables}).
14618 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
14623 @subsubsection C and C@t{++} Defaults
14625 @cindex C and C@t{++} defaults
14627 If you allow @value{GDBN} to set range checking automatically, it
14628 defaults to @code{off} whenever the working language changes to
14629 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
14630 selects the working language.
14632 If you allow @value{GDBN} to set the language automatically, it
14633 recognizes source files whose names end with @file{.c}, @file{.C}, or
14634 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
14635 these files, it sets the working language to C or C@t{++}.
14636 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
14637 for further details.
14640 @subsubsection C and C@t{++} Type and Range Checks
14642 @cindex C and C@t{++} checks
14644 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
14645 checking is used. However, if you turn type checking off, @value{GDBN}
14646 will allow certain non-standard conversions, such as promoting integer
14647 constants to pointers.
14649 Range checking, if turned on, is done on mathematical operations. Array
14650 indices are not checked, since they are often used to index a pointer
14651 that is not itself an array.
14654 @subsubsection @value{GDBN} and C
14656 The @code{set print union} and @code{show print union} commands apply to
14657 the @code{union} type. When set to @samp{on}, any @code{union} that is
14658 inside a @code{struct} or @code{class} is also printed. Otherwise, it
14659 appears as @samp{@{...@}}.
14661 The @code{@@} operator aids in the debugging of dynamic arrays, formed
14662 with pointers and a memory allocation function. @xref{Expressions,
14665 @node Debugging C Plus Plus
14666 @subsubsection @value{GDBN} Features for C@t{++}
14668 @cindex commands for C@t{++}
14670 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
14671 designed specifically for use with C@t{++}. Here is a summary:
14674 @cindex break in overloaded functions
14675 @item @r{breakpoint menus}
14676 When you want a breakpoint in a function whose name is overloaded,
14677 @value{GDBN} has the capability to display a menu of possible breakpoint
14678 locations to help you specify which function definition you want.
14679 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
14681 @cindex overloading in C@t{++}
14682 @item rbreak @var{regex}
14683 Setting breakpoints using regular expressions is helpful for setting
14684 breakpoints on overloaded functions that are not members of any special
14686 @xref{Set Breaks, ,Setting Breakpoints}.
14688 @cindex C@t{++} exception handling
14690 @itemx catch rethrow
14692 Debug C@t{++} exception handling using these commands. @xref{Set
14693 Catchpoints, , Setting Catchpoints}.
14695 @cindex inheritance
14696 @item ptype @var{typename}
14697 Print inheritance relationships as well as other information for type
14699 @xref{Symbols, ,Examining the Symbol Table}.
14701 @item info vtbl @var{expression}.
14702 The @code{info vtbl} command can be used to display the virtual
14703 method tables of the object computed by @var{expression}. This shows
14704 one entry per virtual table; there may be multiple virtual tables when
14705 multiple inheritance is in use.
14707 @cindex C@t{++} demangling
14708 @item demangle @var{name}
14709 Demangle @var{name}.
14710 @xref{Symbols}, for a more complete description of the @code{demangle} command.
14712 @cindex C@t{++} symbol display
14713 @item set print demangle
14714 @itemx show print demangle
14715 @itemx set print asm-demangle
14716 @itemx show print asm-demangle
14717 Control whether C@t{++} symbols display in their source form, both when
14718 displaying code as C@t{++} source and when displaying disassemblies.
14719 @xref{Print Settings, ,Print Settings}.
14721 @item set print object
14722 @itemx show print object
14723 Choose whether to print derived (actual) or declared types of objects.
14724 @xref{Print Settings, ,Print Settings}.
14726 @item set print vtbl
14727 @itemx show print vtbl
14728 Control the format for printing virtual function tables.
14729 @xref{Print Settings, ,Print Settings}.
14730 (The @code{vtbl} commands do not work on programs compiled with the HP
14731 ANSI C@t{++} compiler (@code{aCC}).)
14733 @kindex set overload-resolution
14734 @cindex overloaded functions, overload resolution
14735 @item set overload-resolution on
14736 Enable overload resolution for C@t{++} expression evaluation. The default
14737 is on. For overloaded functions, @value{GDBN} evaluates the arguments
14738 and searches for a function whose signature matches the argument types,
14739 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
14740 Expressions, ,C@t{++} Expressions}, for details).
14741 If it cannot find a match, it emits a message.
14743 @item set overload-resolution off
14744 Disable overload resolution for C@t{++} expression evaluation. For
14745 overloaded functions that are not class member functions, @value{GDBN}
14746 chooses the first function of the specified name that it finds in the
14747 symbol table, whether or not its arguments are of the correct type. For
14748 overloaded functions that are class member functions, @value{GDBN}
14749 searches for a function whose signature @emph{exactly} matches the
14752 @kindex show overload-resolution
14753 @item show overload-resolution
14754 Show the current setting of overload resolution.
14756 @item @r{Overloaded symbol names}
14757 You can specify a particular definition of an overloaded symbol, using
14758 the same notation that is used to declare such symbols in C@t{++}: type
14759 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
14760 also use the @value{GDBN} command-line word completion facilities to list the
14761 available choices, or to finish the type list for you.
14762 @xref{Completion,, Command Completion}, for details on how to do this.
14765 @node Decimal Floating Point
14766 @subsubsection Decimal Floating Point format
14767 @cindex decimal floating point format
14769 @value{GDBN} can examine, set and perform computations with numbers in
14770 decimal floating point format, which in the C language correspond to the
14771 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
14772 specified by the extension to support decimal floating-point arithmetic.
14774 There are two encodings in use, depending on the architecture: BID (Binary
14775 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
14776 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
14779 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
14780 to manipulate decimal floating point numbers, it is not possible to convert
14781 (using a cast, for example) integers wider than 32-bit to decimal float.
14783 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
14784 point computations, error checking in decimal float operations ignores
14785 underflow, overflow and divide by zero exceptions.
14787 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
14788 to inspect @code{_Decimal128} values stored in floating point registers.
14789 See @ref{PowerPC,,PowerPC} for more details.
14795 @value{GDBN} can be used to debug programs written in D and compiled with
14796 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
14797 specific feature --- dynamic arrays.
14802 @cindex Go (programming language)
14803 @value{GDBN} can be used to debug programs written in Go and compiled with
14804 @file{gccgo} or @file{6g} compilers.
14806 Here is a summary of the Go-specific features and restrictions:
14809 @cindex current Go package
14810 @item The current Go package
14811 The name of the current package does not need to be specified when
14812 specifying global variables and functions.
14814 For example, given the program:
14818 var myglob = "Shall we?"
14824 When stopped inside @code{main} either of these work:
14828 (gdb) p main.myglob
14831 @cindex builtin Go types
14832 @item Builtin Go types
14833 The @code{string} type is recognized by @value{GDBN} and is printed
14836 @cindex builtin Go functions
14837 @item Builtin Go functions
14838 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
14839 function and handles it internally.
14841 @cindex restrictions on Go expressions
14842 @item Restrictions on Go expressions
14843 All Go operators are supported except @code{&^}.
14844 The Go @code{_} ``blank identifier'' is not supported.
14845 Automatic dereferencing of pointers is not supported.
14849 @subsection Objective-C
14851 @cindex Objective-C
14852 This section provides information about some commands and command
14853 options that are useful for debugging Objective-C code. See also
14854 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
14855 few more commands specific to Objective-C support.
14858 * Method Names in Commands::
14859 * The Print Command with Objective-C::
14862 @node Method Names in Commands
14863 @subsubsection Method Names in Commands
14865 The following commands have been extended to accept Objective-C method
14866 names as line specifications:
14868 @kindex clear@r{, and Objective-C}
14869 @kindex break@r{, and Objective-C}
14870 @kindex info line@r{, and Objective-C}
14871 @kindex jump@r{, and Objective-C}
14872 @kindex list@r{, and Objective-C}
14876 @item @code{info line}
14881 A fully qualified Objective-C method name is specified as
14884 -[@var{Class} @var{methodName}]
14887 where the minus sign is used to indicate an instance method and a
14888 plus sign (not shown) is used to indicate a class method. The class
14889 name @var{Class} and method name @var{methodName} are enclosed in
14890 brackets, similar to the way messages are specified in Objective-C
14891 source code. For example, to set a breakpoint at the @code{create}
14892 instance method of class @code{Fruit} in the program currently being
14896 break -[Fruit create]
14899 To list ten program lines around the @code{initialize} class method,
14903 list +[NSText initialize]
14906 In the current version of @value{GDBN}, the plus or minus sign is
14907 required. In future versions of @value{GDBN}, the plus or minus
14908 sign will be optional, but you can use it to narrow the search. It
14909 is also possible to specify just a method name:
14915 You must specify the complete method name, including any colons. If
14916 your program's source files contain more than one @code{create} method,
14917 you'll be presented with a numbered list of classes that implement that
14918 method. Indicate your choice by number, or type @samp{0} to exit if
14921 As another example, to clear a breakpoint established at the
14922 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
14925 clear -[NSWindow makeKeyAndOrderFront:]
14928 @node The Print Command with Objective-C
14929 @subsubsection The Print Command With Objective-C
14930 @cindex Objective-C, print objects
14931 @kindex print-object
14932 @kindex po @r{(@code{print-object})}
14934 The print command has also been extended to accept methods. For example:
14937 print -[@var{object} hash]
14940 @cindex print an Objective-C object description
14941 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
14943 will tell @value{GDBN} to send the @code{hash} message to @var{object}
14944 and print the result. Also, an additional command has been added,
14945 @code{print-object} or @code{po} for short, which is meant to print
14946 the description of an object. However, this command may only work
14947 with certain Objective-C libraries that have a particular hook
14948 function, @code{_NSPrintForDebugger}, defined.
14951 @subsection OpenCL C
14954 This section provides information about @value{GDBN}s OpenCL C support.
14957 * OpenCL C Datatypes::
14958 * OpenCL C Expressions::
14959 * OpenCL C Operators::
14962 @node OpenCL C Datatypes
14963 @subsubsection OpenCL C Datatypes
14965 @cindex OpenCL C Datatypes
14966 @value{GDBN} supports the builtin scalar and vector datatypes specified
14967 by OpenCL 1.1. In addition the half- and double-precision floating point
14968 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
14969 extensions are also known to @value{GDBN}.
14971 @node OpenCL C Expressions
14972 @subsubsection OpenCL C Expressions
14974 @cindex OpenCL C Expressions
14975 @value{GDBN} supports accesses to vector components including the access as
14976 lvalue where possible. Since OpenCL C is based on C99 most C expressions
14977 supported by @value{GDBN} can be used as well.
14979 @node OpenCL C Operators
14980 @subsubsection OpenCL C Operators
14982 @cindex OpenCL C Operators
14983 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
14987 @subsection Fortran
14988 @cindex Fortran-specific support in @value{GDBN}
14990 @value{GDBN} can be used to debug programs written in Fortran, but it
14991 currently supports only the features of Fortran 77 language.
14993 @cindex trailing underscore, in Fortran symbols
14994 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
14995 among them) append an underscore to the names of variables and
14996 functions. When you debug programs compiled by those compilers, you
14997 will need to refer to variables and functions with a trailing
15001 * Fortran Operators:: Fortran operators and expressions
15002 * Fortran Defaults:: Default settings for Fortran
15003 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
15006 @node Fortran Operators
15007 @subsubsection Fortran Operators and Expressions
15009 @cindex Fortran operators and expressions
15011 Operators must be defined on values of specific types. For instance,
15012 @code{+} is defined on numbers, but not on characters or other non-
15013 arithmetic types. Operators are often defined on groups of types.
15017 The exponentiation operator. It raises the first operand to the power
15021 The range operator. Normally used in the form of array(low:high) to
15022 represent a section of array.
15025 The access component operator. Normally used to access elements in derived
15026 types. Also suitable for unions. As unions aren't part of regular Fortran,
15027 this can only happen when accessing a register that uses a gdbarch-defined
15031 @node Fortran Defaults
15032 @subsubsection Fortran Defaults
15034 @cindex Fortran Defaults
15036 Fortran symbols are usually case-insensitive, so @value{GDBN} by
15037 default uses case-insensitive matches for Fortran symbols. You can
15038 change that with the @samp{set case-insensitive} command, see
15039 @ref{Symbols}, for the details.
15041 @node Special Fortran Commands
15042 @subsubsection Special Fortran Commands
15044 @cindex Special Fortran commands
15046 @value{GDBN} has some commands to support Fortran-specific features,
15047 such as displaying common blocks.
15050 @cindex @code{COMMON} blocks, Fortran
15051 @kindex info common
15052 @item info common @r{[}@var{common-name}@r{]}
15053 This command prints the values contained in the Fortran @code{COMMON}
15054 block whose name is @var{common-name}. With no argument, the names of
15055 all @code{COMMON} blocks visible at the current program location are
15062 @cindex Pascal support in @value{GDBN}, limitations
15063 Debugging Pascal programs which use sets, subranges, file variables, or
15064 nested functions does not currently work. @value{GDBN} does not support
15065 entering expressions, printing values, or similar features using Pascal
15068 The Pascal-specific command @code{set print pascal_static-members}
15069 controls whether static members of Pascal objects are displayed.
15070 @xref{Print Settings, pascal_static-members}.
15073 @subsection Modula-2
15075 @cindex Modula-2, @value{GDBN} support
15077 The extensions made to @value{GDBN} to support Modula-2 only support
15078 output from the @sc{gnu} Modula-2 compiler (which is currently being
15079 developed). Other Modula-2 compilers are not currently supported, and
15080 attempting to debug executables produced by them is most likely
15081 to give an error as @value{GDBN} reads in the executable's symbol
15084 @cindex expressions in Modula-2
15086 * M2 Operators:: Built-in operators
15087 * Built-In Func/Proc:: Built-in functions and procedures
15088 * M2 Constants:: Modula-2 constants
15089 * M2 Types:: Modula-2 types
15090 * M2 Defaults:: Default settings for Modula-2
15091 * Deviations:: Deviations from standard Modula-2
15092 * M2 Checks:: Modula-2 type and range checks
15093 * M2 Scope:: The scope operators @code{::} and @code{.}
15094 * GDB/M2:: @value{GDBN} and Modula-2
15098 @subsubsection Operators
15099 @cindex Modula-2 operators
15101 Operators must be defined on values of specific types. For instance,
15102 @code{+} is defined on numbers, but not on structures. Operators are
15103 often defined on groups of types. For the purposes of Modula-2, the
15104 following definitions hold:
15109 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
15113 @emph{Character types} consist of @code{CHAR} and its subranges.
15116 @emph{Floating-point types} consist of @code{REAL}.
15119 @emph{Pointer types} consist of anything declared as @code{POINTER TO
15123 @emph{Scalar types} consist of all of the above.
15126 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
15129 @emph{Boolean types} consist of @code{BOOLEAN}.
15133 The following operators are supported, and appear in order of
15134 increasing precedence:
15138 Function argument or array index separator.
15141 Assignment. The value of @var{var} @code{:=} @var{value} is
15145 Less than, greater than on integral, floating-point, or enumerated
15149 Less than or equal to, greater than or equal to
15150 on integral, floating-point and enumerated types, or set inclusion on
15151 set types. Same precedence as @code{<}.
15153 @item =@r{, }<>@r{, }#
15154 Equality and two ways of expressing inequality, valid on scalar types.
15155 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
15156 available for inequality, since @code{#} conflicts with the script
15160 Set membership. Defined on set types and the types of their members.
15161 Same precedence as @code{<}.
15164 Boolean disjunction. Defined on boolean types.
15167 Boolean conjunction. Defined on boolean types.
15170 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
15173 Addition and subtraction on integral and floating-point types, or union
15174 and difference on set types.
15177 Multiplication on integral and floating-point types, or set intersection
15181 Division on floating-point types, or symmetric set difference on set
15182 types. Same precedence as @code{*}.
15185 Integer division and remainder. Defined on integral types. Same
15186 precedence as @code{*}.
15189 Negative. Defined on @code{INTEGER} and @code{REAL} data.
15192 Pointer dereferencing. Defined on pointer types.
15195 Boolean negation. Defined on boolean types. Same precedence as
15199 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
15200 precedence as @code{^}.
15203 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
15206 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
15210 @value{GDBN} and Modula-2 scope operators.
15214 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
15215 treats the use of the operator @code{IN}, or the use of operators
15216 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
15217 @code{<=}, and @code{>=} on sets as an error.
15221 @node Built-In Func/Proc
15222 @subsubsection Built-in Functions and Procedures
15223 @cindex Modula-2 built-ins
15225 Modula-2 also makes available several built-in procedures and functions.
15226 In describing these, the following metavariables are used:
15231 represents an @code{ARRAY} variable.
15234 represents a @code{CHAR} constant or variable.
15237 represents a variable or constant of integral type.
15240 represents an identifier that belongs to a set. Generally used in the
15241 same function with the metavariable @var{s}. The type of @var{s} should
15242 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
15245 represents a variable or constant of integral or floating-point type.
15248 represents a variable or constant of floating-point type.
15254 represents a variable.
15257 represents a variable or constant of one of many types. See the
15258 explanation of the function for details.
15261 All Modula-2 built-in procedures also return a result, described below.
15265 Returns the absolute value of @var{n}.
15268 If @var{c} is a lower case letter, it returns its upper case
15269 equivalent, otherwise it returns its argument.
15272 Returns the character whose ordinal value is @var{i}.
15275 Decrements the value in the variable @var{v} by one. Returns the new value.
15277 @item DEC(@var{v},@var{i})
15278 Decrements the value in the variable @var{v} by @var{i}. Returns the
15281 @item EXCL(@var{m},@var{s})
15282 Removes the element @var{m} from the set @var{s}. Returns the new
15285 @item FLOAT(@var{i})
15286 Returns the floating point equivalent of the integer @var{i}.
15288 @item HIGH(@var{a})
15289 Returns the index of the last member of @var{a}.
15292 Increments the value in the variable @var{v} by one. Returns the new value.
15294 @item INC(@var{v},@var{i})
15295 Increments the value in the variable @var{v} by @var{i}. Returns the
15298 @item INCL(@var{m},@var{s})
15299 Adds the element @var{m} to the set @var{s} if it is not already
15300 there. Returns the new set.
15303 Returns the maximum value of the type @var{t}.
15306 Returns the minimum value of the type @var{t}.
15309 Returns boolean TRUE if @var{i} is an odd number.
15312 Returns the ordinal value of its argument. For example, the ordinal
15313 value of a character is its @sc{ascii} value (on machines supporting
15314 the @sc{ascii} character set). The argument @var{x} must be of an
15315 ordered type, which include integral, character and enumerated types.
15317 @item SIZE(@var{x})
15318 Returns the size of its argument. The argument @var{x} can be a
15319 variable or a type.
15321 @item TRUNC(@var{r})
15322 Returns the integral part of @var{r}.
15324 @item TSIZE(@var{x})
15325 Returns the size of its argument. The argument @var{x} can be a
15326 variable or a type.
15328 @item VAL(@var{t},@var{i})
15329 Returns the member of the type @var{t} whose ordinal value is @var{i}.
15333 @emph{Warning:} Sets and their operations are not yet supported, so
15334 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
15338 @cindex Modula-2 constants
15340 @subsubsection Constants
15342 @value{GDBN} allows you to express the constants of Modula-2 in the following
15348 Integer constants are simply a sequence of digits. When used in an
15349 expression, a constant is interpreted to be type-compatible with the
15350 rest of the expression. Hexadecimal integers are specified by a
15351 trailing @samp{H}, and octal integers by a trailing @samp{B}.
15354 Floating point constants appear as a sequence of digits, followed by a
15355 decimal point and another sequence of digits. An optional exponent can
15356 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
15357 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
15358 digits of the floating point constant must be valid decimal (base 10)
15362 Character constants consist of a single character enclosed by a pair of
15363 like quotes, either single (@code{'}) or double (@code{"}). They may
15364 also be expressed by their ordinal value (their @sc{ascii} value, usually)
15365 followed by a @samp{C}.
15368 String constants consist of a sequence of characters enclosed by a
15369 pair of like quotes, either single (@code{'}) or double (@code{"}).
15370 Escape sequences in the style of C are also allowed. @xref{C
15371 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
15375 Enumerated constants consist of an enumerated identifier.
15378 Boolean constants consist of the identifiers @code{TRUE} and
15382 Pointer constants consist of integral values only.
15385 Set constants are not yet supported.
15389 @subsubsection Modula-2 Types
15390 @cindex Modula-2 types
15392 Currently @value{GDBN} can print the following data types in Modula-2
15393 syntax: array types, record types, set types, pointer types, procedure
15394 types, enumerated types, subrange types and base types. You can also
15395 print the contents of variables declared using these type.
15396 This section gives a number of simple source code examples together with
15397 sample @value{GDBN} sessions.
15399 The first example contains the following section of code:
15408 and you can request @value{GDBN} to interrogate the type and value of
15409 @code{r} and @code{s}.
15412 (@value{GDBP}) print s
15414 (@value{GDBP}) ptype s
15416 (@value{GDBP}) print r
15418 (@value{GDBP}) ptype r
15423 Likewise if your source code declares @code{s} as:
15427 s: SET ['A'..'Z'] ;
15431 then you may query the type of @code{s} by:
15434 (@value{GDBP}) ptype s
15435 type = SET ['A'..'Z']
15439 Note that at present you cannot interactively manipulate set
15440 expressions using the debugger.
15442 The following example shows how you might declare an array in Modula-2
15443 and how you can interact with @value{GDBN} to print its type and contents:
15447 s: ARRAY [-10..10] OF CHAR ;
15451 (@value{GDBP}) ptype s
15452 ARRAY [-10..10] OF CHAR
15455 Note that the array handling is not yet complete and although the type
15456 is printed correctly, expression handling still assumes that all
15457 arrays have a lower bound of zero and not @code{-10} as in the example
15460 Here are some more type related Modula-2 examples:
15464 colour = (blue, red, yellow, green) ;
15465 t = [blue..yellow] ;
15473 The @value{GDBN} interaction shows how you can query the data type
15474 and value of a variable.
15477 (@value{GDBP}) print s
15479 (@value{GDBP}) ptype t
15480 type = [blue..yellow]
15484 In this example a Modula-2 array is declared and its contents
15485 displayed. Observe that the contents are written in the same way as
15486 their @code{C} counterparts.
15490 s: ARRAY [1..5] OF CARDINAL ;
15496 (@value{GDBP}) print s
15497 $1 = @{1, 0, 0, 0, 0@}
15498 (@value{GDBP}) ptype s
15499 type = ARRAY [1..5] OF CARDINAL
15502 The Modula-2 language interface to @value{GDBN} also understands
15503 pointer types as shown in this example:
15507 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
15514 and you can request that @value{GDBN} describes the type of @code{s}.
15517 (@value{GDBP}) ptype s
15518 type = POINTER TO ARRAY [1..5] OF CARDINAL
15521 @value{GDBN} handles compound types as we can see in this example.
15522 Here we combine array types, record types, pointer types and subrange
15533 myarray = ARRAY myrange OF CARDINAL ;
15534 myrange = [-2..2] ;
15536 s: POINTER TO ARRAY myrange OF foo ;
15540 and you can ask @value{GDBN} to describe the type of @code{s} as shown
15544 (@value{GDBP}) ptype s
15545 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
15548 f3 : ARRAY [-2..2] OF CARDINAL;
15553 @subsubsection Modula-2 Defaults
15554 @cindex Modula-2 defaults
15556 If type and range checking are set automatically by @value{GDBN}, they
15557 both default to @code{on} whenever the working language changes to
15558 Modula-2. This happens regardless of whether you or @value{GDBN}
15559 selected the working language.
15561 If you allow @value{GDBN} to set the language automatically, then entering
15562 code compiled from a file whose name ends with @file{.mod} sets the
15563 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
15564 Infer the Source Language}, for further details.
15567 @subsubsection Deviations from Standard Modula-2
15568 @cindex Modula-2, deviations from
15570 A few changes have been made to make Modula-2 programs easier to debug.
15571 This is done primarily via loosening its type strictness:
15575 Unlike in standard Modula-2, pointer constants can be formed by
15576 integers. This allows you to modify pointer variables during
15577 debugging. (In standard Modula-2, the actual address contained in a
15578 pointer variable is hidden from you; it can only be modified
15579 through direct assignment to another pointer variable or expression that
15580 returned a pointer.)
15583 C escape sequences can be used in strings and characters to represent
15584 non-printable characters. @value{GDBN} prints out strings with these
15585 escape sequences embedded. Single non-printable characters are
15586 printed using the @samp{CHR(@var{nnn})} format.
15589 The assignment operator (@code{:=}) returns the value of its right-hand
15593 All built-in procedures both modify @emph{and} return their argument.
15597 @subsubsection Modula-2 Type and Range Checks
15598 @cindex Modula-2 checks
15601 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
15604 @c FIXME remove warning when type/range checks added
15606 @value{GDBN} considers two Modula-2 variables type equivalent if:
15610 They are of types that have been declared equivalent via a @code{TYPE
15611 @var{t1} = @var{t2}} statement
15614 They have been declared on the same line. (Note: This is true of the
15615 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
15618 As long as type checking is enabled, any attempt to combine variables
15619 whose types are not equivalent is an error.
15621 Range checking is done on all mathematical operations, assignment, array
15622 index bounds, and all built-in functions and procedures.
15625 @subsubsection The Scope Operators @code{::} and @code{.}
15627 @cindex @code{.}, Modula-2 scope operator
15628 @cindex colon, doubled as scope operator
15630 @vindex colon-colon@r{, in Modula-2}
15631 @c Info cannot handle :: but TeX can.
15634 @vindex ::@r{, in Modula-2}
15637 There are a few subtle differences between the Modula-2 scope operator
15638 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
15643 @var{module} . @var{id}
15644 @var{scope} :: @var{id}
15648 where @var{scope} is the name of a module or a procedure,
15649 @var{module} the name of a module, and @var{id} is any declared
15650 identifier within your program, except another module.
15652 Using the @code{::} operator makes @value{GDBN} search the scope
15653 specified by @var{scope} for the identifier @var{id}. If it is not
15654 found in the specified scope, then @value{GDBN} searches all scopes
15655 enclosing the one specified by @var{scope}.
15657 Using the @code{.} operator makes @value{GDBN} search the current scope for
15658 the identifier specified by @var{id} that was imported from the
15659 definition module specified by @var{module}. With this operator, it is
15660 an error if the identifier @var{id} was not imported from definition
15661 module @var{module}, or if @var{id} is not an identifier in
15665 @subsubsection @value{GDBN} and Modula-2
15667 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
15668 Five subcommands of @code{set print} and @code{show print} apply
15669 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
15670 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
15671 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
15672 analogue in Modula-2.
15674 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
15675 with any language, is not useful with Modula-2. Its
15676 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
15677 created in Modula-2 as they can in C or C@t{++}. However, because an
15678 address can be specified by an integral constant, the construct
15679 @samp{@{@var{type}@}@var{adrexp}} is still useful.
15681 @cindex @code{#} in Modula-2
15682 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
15683 interpreted as the beginning of a comment. Use @code{<>} instead.
15689 The extensions made to @value{GDBN} for Ada only support
15690 output from the @sc{gnu} Ada (GNAT) compiler.
15691 Other Ada compilers are not currently supported, and
15692 attempting to debug executables produced by them is most likely
15696 @cindex expressions in Ada
15698 * Ada Mode Intro:: General remarks on the Ada syntax
15699 and semantics supported by Ada mode
15701 * Omissions from Ada:: Restrictions on the Ada expression syntax.
15702 * Additions to Ada:: Extensions of the Ada expression syntax.
15703 * Overloading support for Ada:: Support for expressions involving overloaded
15705 * Stopping Before Main Program:: Debugging the program during elaboration.
15706 * Ada Exceptions:: Ada Exceptions
15707 * Ada Tasks:: Listing and setting breakpoints in tasks.
15708 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
15709 * Ravenscar Profile:: Tasking Support when using the Ravenscar
15711 * Ada Glitches:: Known peculiarities of Ada mode.
15714 @node Ada Mode Intro
15715 @subsubsection Introduction
15716 @cindex Ada mode, general
15718 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
15719 syntax, with some extensions.
15720 The philosophy behind the design of this subset is
15724 That @value{GDBN} should provide basic literals and access to operations for
15725 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
15726 leaving more sophisticated computations to subprograms written into the
15727 program (which therefore may be called from @value{GDBN}).
15730 That type safety and strict adherence to Ada language restrictions
15731 are not particularly important to the @value{GDBN} user.
15734 That brevity is important to the @value{GDBN} user.
15737 Thus, for brevity, the debugger acts as if all names declared in
15738 user-written packages are directly visible, even if they are not visible
15739 according to Ada rules, thus making it unnecessary to fully qualify most
15740 names with their packages, regardless of context. Where this causes
15741 ambiguity, @value{GDBN} asks the user's intent.
15743 The debugger will start in Ada mode if it detects an Ada main program.
15744 As for other languages, it will enter Ada mode when stopped in a program that
15745 was translated from an Ada source file.
15747 While in Ada mode, you may use `@t{--}' for comments. This is useful
15748 mostly for documenting command files. The standard @value{GDBN} comment
15749 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
15750 middle (to allow based literals).
15752 @node Omissions from Ada
15753 @subsubsection Omissions from Ada
15754 @cindex Ada, omissions from
15756 Here are the notable omissions from the subset:
15760 Only a subset of the attributes are supported:
15764 @t{'First}, @t{'Last}, and @t{'Length}
15765 on array objects (not on types and subtypes).
15768 @t{'Min} and @t{'Max}.
15771 @t{'Pos} and @t{'Val}.
15777 @t{'Range} on array objects (not subtypes), but only as the right
15778 operand of the membership (@code{in}) operator.
15781 @t{'Access}, @t{'Unchecked_Access}, and
15782 @t{'Unrestricted_Access} (a GNAT extension).
15790 @code{Characters.Latin_1} are not available and
15791 concatenation is not implemented. Thus, escape characters in strings are
15792 not currently available.
15795 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
15796 equality of representations. They will generally work correctly
15797 for strings and arrays whose elements have integer or enumeration types.
15798 They may not work correctly for arrays whose element
15799 types have user-defined equality, for arrays of real values
15800 (in particular, IEEE-conformant floating point, because of negative
15801 zeroes and NaNs), and for arrays whose elements contain unused bits with
15802 indeterminate values.
15805 The other component-by-component array operations (@code{and}, @code{or},
15806 @code{xor}, @code{not}, and relational tests other than equality)
15807 are not implemented.
15810 @cindex array aggregates (Ada)
15811 @cindex record aggregates (Ada)
15812 @cindex aggregates (Ada)
15813 There is limited support for array and record aggregates. They are
15814 permitted only on the right sides of assignments, as in these examples:
15817 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
15818 (@value{GDBP}) set An_Array := (1, others => 0)
15819 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
15820 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
15821 (@value{GDBP}) set A_Record := (1, "Peter", True);
15822 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
15826 discriminant's value by assigning an aggregate has an
15827 undefined effect if that discriminant is used within the record.
15828 However, you can first modify discriminants by directly assigning to
15829 them (which normally would not be allowed in Ada), and then performing an
15830 aggregate assignment. For example, given a variable @code{A_Rec}
15831 declared to have a type such as:
15834 type Rec (Len : Small_Integer := 0) is record
15836 Vals : IntArray (1 .. Len);
15840 you can assign a value with a different size of @code{Vals} with two
15844 (@value{GDBP}) set A_Rec.Len := 4
15845 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
15848 As this example also illustrates, @value{GDBN} is very loose about the usual
15849 rules concerning aggregates. You may leave out some of the
15850 components of an array or record aggregate (such as the @code{Len}
15851 component in the assignment to @code{A_Rec} above); they will retain their
15852 original values upon assignment. You may freely use dynamic values as
15853 indices in component associations. You may even use overlapping or
15854 redundant component associations, although which component values are
15855 assigned in such cases is not defined.
15858 Calls to dispatching subprograms are not implemented.
15861 The overloading algorithm is much more limited (i.e., less selective)
15862 than that of real Ada. It makes only limited use of the context in
15863 which a subexpression appears to resolve its meaning, and it is much
15864 looser in its rules for allowing type matches. As a result, some
15865 function calls will be ambiguous, and the user will be asked to choose
15866 the proper resolution.
15869 The @code{new} operator is not implemented.
15872 Entry calls are not implemented.
15875 Aside from printing, arithmetic operations on the native VAX floating-point
15876 formats are not supported.
15879 It is not possible to slice a packed array.
15882 The names @code{True} and @code{False}, when not part of a qualified name,
15883 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
15885 Should your program
15886 redefine these names in a package or procedure (at best a dubious practice),
15887 you will have to use fully qualified names to access their new definitions.
15890 @node Additions to Ada
15891 @subsubsection Additions to Ada
15892 @cindex Ada, deviations from
15894 As it does for other languages, @value{GDBN} makes certain generic
15895 extensions to Ada (@pxref{Expressions}):
15899 If the expression @var{E} is a variable residing in memory (typically
15900 a local variable or array element) and @var{N} is a positive integer,
15901 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
15902 @var{N}-1 adjacent variables following it in memory as an array. In
15903 Ada, this operator is generally not necessary, since its prime use is
15904 in displaying parts of an array, and slicing will usually do this in
15905 Ada. However, there are occasional uses when debugging programs in
15906 which certain debugging information has been optimized away.
15909 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
15910 appears in function or file @var{B}.'' When @var{B} is a file name,
15911 you must typically surround it in single quotes.
15914 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
15915 @var{type} that appears at address @var{addr}.''
15918 A name starting with @samp{$} is a convenience variable
15919 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
15922 In addition, @value{GDBN} provides a few other shortcuts and outright
15923 additions specific to Ada:
15927 The assignment statement is allowed as an expression, returning
15928 its right-hand operand as its value. Thus, you may enter
15931 (@value{GDBP}) set x := y + 3
15932 (@value{GDBP}) print A(tmp := y + 1)
15936 The semicolon is allowed as an ``operator,'' returning as its value
15937 the value of its right-hand operand.
15938 This allows, for example,
15939 complex conditional breaks:
15942 (@value{GDBP}) break f
15943 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
15947 Rather than use catenation and symbolic character names to introduce special
15948 characters into strings, one may instead use a special bracket notation,
15949 which is also used to print strings. A sequence of characters of the form
15950 @samp{["@var{XX}"]} within a string or character literal denotes the
15951 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
15952 sequence of characters @samp{["""]} also denotes a single quotation mark
15953 in strings. For example,
15955 "One line.["0a"]Next line.["0a"]"
15958 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
15962 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
15963 @t{'Max} is optional (and is ignored in any case). For example, it is valid
15967 (@value{GDBP}) print 'max(x, y)
15971 When printing arrays, @value{GDBN} uses positional notation when the
15972 array has a lower bound of 1, and uses a modified named notation otherwise.
15973 For example, a one-dimensional array of three integers with a lower bound
15974 of 3 might print as
15981 That is, in contrast to valid Ada, only the first component has a @code{=>}
15985 You may abbreviate attributes in expressions with any unique,
15986 multi-character subsequence of
15987 their names (an exact match gets preference).
15988 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
15989 in place of @t{a'length}.
15992 @cindex quoting Ada internal identifiers
15993 Since Ada is case-insensitive, the debugger normally maps identifiers you type
15994 to lower case. The GNAT compiler uses upper-case characters for
15995 some of its internal identifiers, which are normally of no interest to users.
15996 For the rare occasions when you actually have to look at them,
15997 enclose them in angle brackets to avoid the lower-case mapping.
16000 (@value{GDBP}) print <JMPBUF_SAVE>[0]
16004 Printing an object of class-wide type or dereferencing an
16005 access-to-class-wide value will display all the components of the object's
16006 specific type (as indicated by its run-time tag). Likewise, component
16007 selection on such a value will operate on the specific type of the
16012 @node Overloading support for Ada
16013 @subsubsection Overloading support for Ada
16014 @cindex overloading, Ada
16016 The debugger supports limited overloading. Given a subprogram call in which
16017 the function symbol has multiple definitions, it will use the number of
16018 actual parameters and some information about their types to attempt to narrow
16019 the set of definitions. It also makes very limited use of context, preferring
16020 procedures to functions in the context of the @code{call} command, and
16021 functions to procedures elsewhere.
16023 If, after narrowing, the set of matching definitions still contains more than
16024 one definition, @value{GDBN} will display a menu to query which one it should
16028 (@value{GDBP}) print f(1)
16029 Multiple matches for f
16031 [1] foo.f (integer) return boolean at foo.adb:23
16032 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
16036 In this case, just select one menu entry either to cancel expression evaluation
16037 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
16038 instance (type the corresponding number and press @key{RET}).
16040 Here are a couple of commands to customize @value{GDBN}'s behavior in this
16045 @kindex set ada print-signatures
16046 @item set ada print-signatures
16047 Control whether parameter types and return types are displayed in overloads
16048 selection menus. It is @code{on} by default.
16049 @xref{Overloading support for Ada}.
16051 @kindex show ada print-signatures
16052 @item show ada print-signatures
16053 Show the current setting for displaying parameter types and return types in
16054 overloads selection menu.
16055 @xref{Overloading support for Ada}.
16059 @node Stopping Before Main Program
16060 @subsubsection Stopping at the Very Beginning
16062 @cindex breakpointing Ada elaboration code
16063 It is sometimes necessary to debug the program during elaboration, and
16064 before reaching the main procedure.
16065 As defined in the Ada Reference
16066 Manual, the elaboration code is invoked from a procedure called
16067 @code{adainit}. To run your program up to the beginning of
16068 elaboration, simply use the following two commands:
16069 @code{tbreak adainit} and @code{run}.
16071 @node Ada Exceptions
16072 @subsubsection Ada Exceptions
16074 A command is provided to list all Ada exceptions:
16077 @kindex info exceptions
16078 @item info exceptions
16079 @itemx info exceptions @var{regexp}
16080 The @code{info exceptions} command allows you to list all Ada exceptions
16081 defined within the program being debugged, as well as their addresses.
16082 With a regular expression, @var{regexp}, as argument, only those exceptions
16083 whose names match @var{regexp} are listed.
16086 Below is a small example, showing how the command can be used, first
16087 without argument, and next with a regular expression passed as an
16091 (@value{GDBP}) info exceptions
16092 All defined Ada exceptions:
16093 constraint_error: 0x613da0
16094 program_error: 0x613d20
16095 storage_error: 0x613ce0
16096 tasking_error: 0x613ca0
16097 const.aint_global_e: 0x613b00
16098 (@value{GDBP}) info exceptions const.aint
16099 All Ada exceptions matching regular expression "const.aint":
16100 constraint_error: 0x613da0
16101 const.aint_global_e: 0x613b00
16104 It is also possible to ask @value{GDBN} to stop your program's execution
16105 when an exception is raised. For more details, see @ref{Set Catchpoints}.
16108 @subsubsection Extensions for Ada Tasks
16109 @cindex Ada, tasking
16111 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
16112 @value{GDBN} provides the following task-related commands:
16117 This command shows a list of current Ada tasks, as in the following example:
16124 (@value{GDBP}) info tasks
16125 ID TID P-ID Pri State Name
16126 1 8088000 0 15 Child Activation Wait main_task
16127 2 80a4000 1 15 Accept Statement b
16128 3 809a800 1 15 Child Activation Wait a
16129 * 4 80ae800 3 15 Runnable c
16134 In this listing, the asterisk before the last task indicates it to be the
16135 task currently being inspected.
16139 Represents @value{GDBN}'s internal task number.
16145 The parent's task ID (@value{GDBN}'s internal task number).
16148 The base priority of the task.
16151 Current state of the task.
16155 The task has been created but has not been activated. It cannot be
16159 The task is not blocked for any reason known to Ada. (It may be waiting
16160 for a mutex, though.) It is conceptually "executing" in normal mode.
16163 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
16164 that were waiting on terminate alternatives have been awakened and have
16165 terminated themselves.
16167 @item Child Activation Wait
16168 The task is waiting for created tasks to complete activation.
16170 @item Accept Statement
16171 The task is waiting on an accept or selective wait statement.
16173 @item Waiting on entry call
16174 The task is waiting on an entry call.
16176 @item Async Select Wait
16177 The task is waiting to start the abortable part of an asynchronous
16181 The task is waiting on a select statement with only a delay
16184 @item Child Termination Wait
16185 The task is sleeping having completed a master within itself, and is
16186 waiting for the tasks dependent on that master to become terminated or
16187 waiting on a terminate Phase.
16189 @item Wait Child in Term Alt
16190 The task is sleeping waiting for tasks on terminate alternatives to
16191 finish terminating.
16193 @item Accepting RV with @var{taskno}
16194 The task is accepting a rendez-vous with the task @var{taskno}.
16198 Name of the task in the program.
16202 @kindex info task @var{taskno}
16203 @item info task @var{taskno}
16204 This command shows detailled informations on the specified task, as in
16205 the following example:
16210 (@value{GDBP}) info tasks
16211 ID TID P-ID Pri State Name
16212 1 8077880 0 15 Child Activation Wait main_task
16213 * 2 807c468 1 15 Runnable task_1
16214 (@value{GDBP}) info task 2
16215 Ada Task: 0x807c468
16218 Parent: 1 (main_task)
16224 @kindex task@r{ (Ada)}
16225 @cindex current Ada task ID
16226 This command prints the ID of the current task.
16232 (@value{GDBP}) info tasks
16233 ID TID P-ID Pri State Name
16234 1 8077870 0 15 Child Activation Wait main_task
16235 * 2 807c458 1 15 Runnable t
16236 (@value{GDBP}) task
16237 [Current task is 2]
16240 @item task @var{taskno}
16241 @cindex Ada task switching
16242 This command is like the @code{thread @var{thread-id}}
16243 command (@pxref{Threads}). It switches the context of debugging
16244 from the current task to the given task.
16250 (@value{GDBP}) info tasks
16251 ID TID P-ID Pri State Name
16252 1 8077870 0 15 Child Activation Wait main_task
16253 * 2 807c458 1 15 Runnable t
16254 (@value{GDBP}) task 1
16255 [Switching to task 1]
16256 #0 0x8067726 in pthread_cond_wait ()
16258 #0 0x8067726 in pthread_cond_wait ()
16259 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
16260 #2 0x805cb63 in system.task_primitives.operations.sleep ()
16261 #3 0x806153e in system.tasking.stages.activate_tasks ()
16262 #4 0x804aacc in un () at un.adb:5
16265 @item break @var{location} task @var{taskno}
16266 @itemx break @var{location} task @var{taskno} if @dots{}
16267 @cindex breakpoints and tasks, in Ada
16268 @cindex task breakpoints, in Ada
16269 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
16270 These commands are like the @code{break @dots{} thread @dots{}}
16271 command (@pxref{Thread Stops}). The
16272 @var{location} argument specifies source lines, as described
16273 in @ref{Specify Location}.
16275 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
16276 to specify that you only want @value{GDBN} to stop the program when a
16277 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
16278 numeric task identifiers assigned by @value{GDBN}, shown in the first
16279 column of the @samp{info tasks} display.
16281 If you do not specify @samp{task @var{taskno}} when you set a
16282 breakpoint, the breakpoint applies to @emph{all} tasks of your
16285 You can use the @code{task} qualifier on conditional breakpoints as
16286 well; in this case, place @samp{task @var{taskno}} before the
16287 breakpoint condition (before the @code{if}).
16295 (@value{GDBP}) info tasks
16296 ID TID P-ID Pri State Name
16297 1 140022020 0 15 Child Activation Wait main_task
16298 2 140045060 1 15 Accept/Select Wait t2
16299 3 140044840 1 15 Runnable t1
16300 * 4 140056040 1 15 Runnable t3
16301 (@value{GDBP}) b 15 task 2
16302 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
16303 (@value{GDBP}) cont
16308 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
16310 (@value{GDBP}) info tasks
16311 ID TID P-ID Pri State Name
16312 1 140022020 0 15 Child Activation Wait main_task
16313 * 2 140045060 1 15 Runnable t2
16314 3 140044840 1 15 Runnable t1
16315 4 140056040 1 15 Delay Sleep t3
16319 @node Ada Tasks and Core Files
16320 @subsubsection Tasking Support when Debugging Core Files
16321 @cindex Ada tasking and core file debugging
16323 When inspecting a core file, as opposed to debugging a live program,
16324 tasking support may be limited or even unavailable, depending on
16325 the platform being used.
16326 For instance, on x86-linux, the list of tasks is available, but task
16327 switching is not supported.
16329 On certain platforms, the debugger needs to perform some
16330 memory writes in order to provide Ada tasking support. When inspecting
16331 a core file, this means that the core file must be opened with read-write
16332 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
16333 Under these circumstances, you should make a backup copy of the core
16334 file before inspecting it with @value{GDBN}.
16336 @node Ravenscar Profile
16337 @subsubsection Tasking Support when using the Ravenscar Profile
16338 @cindex Ravenscar Profile
16340 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
16341 specifically designed for systems with safety-critical real-time
16345 @kindex set ravenscar task-switching on
16346 @cindex task switching with program using Ravenscar Profile
16347 @item set ravenscar task-switching on
16348 Allows task switching when debugging a program that uses the Ravenscar
16349 Profile. This is the default.
16351 @kindex set ravenscar task-switching off
16352 @item set ravenscar task-switching off
16353 Turn off task switching when debugging a program that uses the Ravenscar
16354 Profile. This is mostly intended to disable the code that adds support
16355 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
16356 the Ravenscar runtime is preventing @value{GDBN} from working properly.
16357 To be effective, this command should be run before the program is started.
16359 @kindex show ravenscar task-switching
16360 @item show ravenscar task-switching
16361 Show whether it is possible to switch from task to task in a program
16362 using the Ravenscar Profile.
16367 @subsubsection Known Peculiarities of Ada Mode
16368 @cindex Ada, problems
16370 Besides the omissions listed previously (@pxref{Omissions from Ada}),
16371 we know of several problems with and limitations of Ada mode in
16373 some of which will be fixed with planned future releases of the debugger
16374 and the GNU Ada compiler.
16378 Static constants that the compiler chooses not to materialize as objects in
16379 storage are invisible to the debugger.
16382 Named parameter associations in function argument lists are ignored (the
16383 argument lists are treated as positional).
16386 Many useful library packages are currently invisible to the debugger.
16389 Fixed-point arithmetic, conversions, input, and output is carried out using
16390 floating-point arithmetic, and may give results that only approximate those on
16394 The GNAT compiler never generates the prefix @code{Standard} for any of
16395 the standard symbols defined by the Ada language. @value{GDBN} knows about
16396 this: it will strip the prefix from names when you use it, and will never
16397 look for a name you have so qualified among local symbols, nor match against
16398 symbols in other packages or subprograms. If you have
16399 defined entities anywhere in your program other than parameters and
16400 local variables whose simple names match names in @code{Standard},
16401 GNAT's lack of qualification here can cause confusion. When this happens,
16402 you can usually resolve the confusion
16403 by qualifying the problematic names with package
16404 @code{Standard} explicitly.
16407 Older versions of the compiler sometimes generate erroneous debugging
16408 information, resulting in the debugger incorrectly printing the value
16409 of affected entities. In some cases, the debugger is able to work
16410 around an issue automatically. In other cases, the debugger is able
16411 to work around the issue, but the work-around has to be specifically
16414 @kindex set ada trust-PAD-over-XVS
16415 @kindex show ada trust-PAD-over-XVS
16418 @item set ada trust-PAD-over-XVS on
16419 Configure GDB to strictly follow the GNAT encoding when computing the
16420 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
16421 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
16422 a complete description of the encoding used by the GNAT compiler).
16423 This is the default.
16425 @item set ada trust-PAD-over-XVS off
16426 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
16427 sometimes prints the wrong value for certain entities, changing @code{ada
16428 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
16429 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
16430 @code{off}, but this incurs a slight performance penalty, so it is
16431 recommended to leave this setting to @code{on} unless necessary.
16435 @cindex GNAT descriptive types
16436 @cindex GNAT encoding
16437 Internally, the debugger also relies on the compiler following a number
16438 of conventions known as the @samp{GNAT Encoding}, all documented in
16439 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
16440 how the debugging information should be generated for certain types.
16441 In particular, this convention makes use of @dfn{descriptive types},
16442 which are artificial types generated purely to help the debugger.
16444 These encodings were defined at a time when the debugging information
16445 format used was not powerful enough to describe some of the more complex
16446 types available in Ada. Since DWARF allows us to express nearly all
16447 Ada features, the long-term goal is to slowly replace these descriptive
16448 types by their pure DWARF equivalent. To facilitate that transition,
16449 a new maintenance option is available to force the debugger to ignore
16450 those descriptive types. It allows the user to quickly evaluate how
16451 well @value{GDBN} works without them.
16455 @kindex maint ada set ignore-descriptive-types
16456 @item maintenance ada set ignore-descriptive-types [on|off]
16457 Control whether the debugger should ignore descriptive types.
16458 The default is not to ignore descriptives types (@code{off}).
16460 @kindex maint ada show ignore-descriptive-types
16461 @item maintenance ada show ignore-descriptive-types
16462 Show if descriptive types are ignored by @value{GDBN}.
16466 @node Unsupported Languages
16467 @section Unsupported Languages
16469 @cindex unsupported languages
16470 @cindex minimal language
16471 In addition to the other fully-supported programming languages,
16472 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
16473 It does not represent a real programming language, but provides a set
16474 of capabilities close to what the C or assembly languages provide.
16475 This should allow most simple operations to be performed while debugging
16476 an application that uses a language currently not supported by @value{GDBN}.
16478 If the language is set to @code{auto}, @value{GDBN} will automatically
16479 select this language if the current frame corresponds to an unsupported
16483 @chapter Examining the Symbol Table
16485 The commands described in this chapter allow you to inquire about the
16486 symbols (names of variables, functions and types) defined in your
16487 program. This information is inherent in the text of your program and
16488 does not change as your program executes. @value{GDBN} finds it in your
16489 program's symbol table, in the file indicated when you started @value{GDBN}
16490 (@pxref{File Options, ,Choosing Files}), or by one of the
16491 file-management commands (@pxref{Files, ,Commands to Specify Files}).
16493 @cindex symbol names
16494 @cindex names of symbols
16495 @cindex quoting names
16496 Occasionally, you may need to refer to symbols that contain unusual
16497 characters, which @value{GDBN} ordinarily treats as word delimiters. The
16498 most frequent case is in referring to static variables in other
16499 source files (@pxref{Variables,,Program Variables}). File names
16500 are recorded in object files as debugging symbols, but @value{GDBN} would
16501 ordinarily parse a typical file name, like @file{foo.c}, as the three words
16502 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
16503 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
16510 looks up the value of @code{x} in the scope of the file @file{foo.c}.
16513 @cindex case-insensitive symbol names
16514 @cindex case sensitivity in symbol names
16515 @kindex set case-sensitive
16516 @item set case-sensitive on
16517 @itemx set case-sensitive off
16518 @itemx set case-sensitive auto
16519 Normally, when @value{GDBN} looks up symbols, it matches their names
16520 with case sensitivity determined by the current source language.
16521 Occasionally, you may wish to control that. The command @code{set
16522 case-sensitive} lets you do that by specifying @code{on} for
16523 case-sensitive matches or @code{off} for case-insensitive ones. If
16524 you specify @code{auto}, case sensitivity is reset to the default
16525 suitable for the source language. The default is case-sensitive
16526 matches for all languages except for Fortran, for which the default is
16527 case-insensitive matches.
16529 @kindex show case-sensitive
16530 @item show case-sensitive
16531 This command shows the current setting of case sensitivity for symbols
16534 @kindex set print type methods
16535 @item set print type methods
16536 @itemx set print type methods on
16537 @itemx set print type methods off
16538 Normally, when @value{GDBN} prints a class, it displays any methods
16539 declared in that class. You can control this behavior either by
16540 passing the appropriate flag to @code{ptype}, or using @command{set
16541 print type methods}. Specifying @code{on} will cause @value{GDBN} to
16542 display the methods; this is the default. Specifying @code{off} will
16543 cause @value{GDBN} to omit the methods.
16545 @kindex show print type methods
16546 @item show print type methods
16547 This command shows the current setting of method display when printing
16550 @kindex set print type typedefs
16551 @item set print type typedefs
16552 @itemx set print type typedefs on
16553 @itemx set print type typedefs off
16555 Normally, when @value{GDBN} prints a class, it displays any typedefs
16556 defined in that class. You can control this behavior either by
16557 passing the appropriate flag to @code{ptype}, or using @command{set
16558 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
16559 display the typedef definitions; this is the default. Specifying
16560 @code{off} will cause @value{GDBN} to omit the typedef definitions.
16561 Note that this controls whether the typedef definition itself is
16562 printed, not whether typedef names are substituted when printing other
16565 @kindex show print type typedefs
16566 @item show print type typedefs
16567 This command shows the current setting of typedef display when
16570 @kindex info address
16571 @cindex address of a symbol
16572 @item info address @var{symbol}
16573 Describe where the data for @var{symbol} is stored. For a register
16574 variable, this says which register it is kept in. For a non-register
16575 local variable, this prints the stack-frame offset at which the variable
16578 Note the contrast with @samp{print &@var{symbol}}, which does not work
16579 at all for a register variable, and for a stack local variable prints
16580 the exact address of the current instantiation of the variable.
16582 @kindex info symbol
16583 @cindex symbol from address
16584 @cindex closest symbol and offset for an address
16585 @item info symbol @var{addr}
16586 Print the name of a symbol which is stored at the address @var{addr}.
16587 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
16588 nearest symbol and an offset from it:
16591 (@value{GDBP}) info symbol 0x54320
16592 _initialize_vx + 396 in section .text
16596 This is the opposite of the @code{info address} command. You can use
16597 it to find out the name of a variable or a function given its address.
16599 For dynamically linked executables, the name of executable or shared
16600 library containing the symbol is also printed:
16603 (@value{GDBP}) info symbol 0x400225
16604 _start + 5 in section .text of /tmp/a.out
16605 (@value{GDBP}) info symbol 0x2aaaac2811cf
16606 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
16611 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
16612 Demangle @var{name}.
16613 If @var{language} is provided it is the name of the language to demangle
16614 @var{name} in. Otherwise @var{name} is demangled in the current language.
16616 The @samp{--} option specifies the end of options,
16617 and is useful when @var{name} begins with a dash.
16619 The parameter @code{demangle-style} specifies how to interpret the kind
16620 of mangling used. @xref{Print Settings}.
16623 @item whatis[/@var{flags}] [@var{arg}]
16624 Print the data type of @var{arg}, which can be either an expression
16625 or a name of a data type. With no argument, print the data type of
16626 @code{$}, the last value in the value history.
16628 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
16629 is not actually evaluated, and any side-effecting operations (such as
16630 assignments or function calls) inside it do not take place.
16632 If @var{arg} is a variable or an expression, @code{whatis} prints its
16633 literal type as it is used in the source code. If the type was
16634 defined using a @code{typedef}, @code{whatis} will @emph{not} print
16635 the data type underlying the @code{typedef}. If the type of the
16636 variable or the expression is a compound data type, such as
16637 @code{struct} or @code{class}, @code{whatis} never prints their
16638 fields or methods. It just prints the @code{struct}/@code{class}
16639 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
16640 such a compound data type, use @code{ptype}.
16642 If @var{arg} is a type name that was defined using @code{typedef},
16643 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
16644 Unrolling means that @code{whatis} will show the underlying type used
16645 in the @code{typedef} declaration of @var{arg}. However, if that
16646 underlying type is also a @code{typedef}, @code{whatis} will not
16649 For C code, the type names may also have the form @samp{class
16650 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
16651 @var{union-tag}} or @samp{enum @var{enum-tag}}.
16653 @var{flags} can be used to modify how the type is displayed.
16654 Available flags are:
16658 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
16659 parameters and typedefs defined in a class when printing the class'
16660 members. The @code{/r} flag disables this.
16663 Do not print methods defined in the class.
16666 Print methods defined in the class. This is the default, but the flag
16667 exists in case you change the default with @command{set print type methods}.
16670 Do not print typedefs defined in the class. Note that this controls
16671 whether the typedef definition itself is printed, not whether typedef
16672 names are substituted when printing other types.
16675 Print typedefs defined in the class. This is the default, but the flag
16676 exists in case you change the default with @command{set print type typedefs}.
16680 @item ptype[/@var{flags}] [@var{arg}]
16681 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
16682 detailed description of the type, instead of just the name of the type.
16683 @xref{Expressions, ,Expressions}.
16685 Contrary to @code{whatis}, @code{ptype} always unrolls any
16686 @code{typedef}s in its argument declaration, whether the argument is
16687 a variable, expression, or a data type. This means that @code{ptype}
16688 of a variable or an expression will not print literally its type as
16689 present in the source code---use @code{whatis} for that. @code{typedef}s at
16690 the pointer or reference targets are also unrolled. Only @code{typedef}s of
16691 fields, methods and inner @code{class typedef}s of @code{struct}s,
16692 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
16694 For example, for this variable declaration:
16697 typedef double real_t;
16698 struct complex @{ real_t real; double imag; @};
16699 typedef struct complex complex_t;
16701 real_t *real_pointer_var;
16705 the two commands give this output:
16709 (@value{GDBP}) whatis var
16711 (@value{GDBP}) ptype var
16712 type = struct complex @{
16716 (@value{GDBP}) whatis complex_t
16717 type = struct complex
16718 (@value{GDBP}) whatis struct complex
16719 type = struct complex
16720 (@value{GDBP}) ptype struct complex
16721 type = struct complex @{
16725 (@value{GDBP}) whatis real_pointer_var
16727 (@value{GDBP}) ptype real_pointer_var
16733 As with @code{whatis}, using @code{ptype} without an argument refers to
16734 the type of @code{$}, the last value in the value history.
16736 @cindex incomplete type
16737 Sometimes, programs use opaque data types or incomplete specifications
16738 of complex data structure. If the debug information included in the
16739 program does not allow @value{GDBN} to display a full declaration of
16740 the data type, it will say @samp{<incomplete type>}. For example,
16741 given these declarations:
16745 struct foo *fooptr;
16749 but no definition for @code{struct foo} itself, @value{GDBN} will say:
16752 (@value{GDBP}) ptype foo
16753 $1 = <incomplete type>
16757 ``Incomplete type'' is C terminology for data types that are not
16758 completely specified.
16761 @item info types @var{regexp}
16763 Print a brief description of all types whose names match the regular
16764 expression @var{regexp} (or all types in your program, if you supply
16765 no argument). Each complete typename is matched as though it were a
16766 complete line; thus, @samp{i type value} gives information on all
16767 types in your program whose names include the string @code{value}, but
16768 @samp{i type ^value$} gives information only on types whose complete
16769 name is @code{value}.
16771 This command differs from @code{ptype} in two ways: first, like
16772 @code{whatis}, it does not print a detailed description; second, it
16773 lists all source files where a type is defined.
16775 @kindex info type-printers
16776 @item info type-printers
16777 Versions of @value{GDBN} that ship with Python scripting enabled may
16778 have ``type printers'' available. When using @command{ptype} or
16779 @command{whatis}, these printers are consulted when the name of a type
16780 is needed. @xref{Type Printing API}, for more information on writing
16783 @code{info type-printers} displays all the available type printers.
16785 @kindex enable type-printer
16786 @kindex disable type-printer
16787 @item enable type-printer @var{name}@dots{}
16788 @item disable type-printer @var{name}@dots{}
16789 These commands can be used to enable or disable type printers.
16792 @cindex local variables
16793 @item info scope @var{location}
16794 List all the variables local to a particular scope. This command
16795 accepts a @var{location} argument---a function name, a source line, or
16796 an address preceded by a @samp{*}, and prints all the variables local
16797 to the scope defined by that location. (@xref{Specify Location}, for
16798 details about supported forms of @var{location}.) For example:
16801 (@value{GDBP}) @b{info scope command_line_handler}
16802 Scope for command_line_handler:
16803 Symbol rl is an argument at stack/frame offset 8, length 4.
16804 Symbol linebuffer is in static storage at address 0x150a18, length 4.
16805 Symbol linelength is in static storage at address 0x150a1c, length 4.
16806 Symbol p is a local variable in register $esi, length 4.
16807 Symbol p1 is a local variable in register $ebx, length 4.
16808 Symbol nline is a local variable in register $edx, length 4.
16809 Symbol repeat is a local variable at frame offset -8, length 4.
16813 This command is especially useful for determining what data to collect
16814 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
16817 @kindex info source
16819 Show information about the current source file---that is, the source file for
16820 the function containing the current point of execution:
16823 the name of the source file, and the directory containing it,
16825 the directory it was compiled in,
16827 its length, in lines,
16829 which programming language it is written in,
16831 if the debug information provides it, the program that compiled the file
16832 (which may include, e.g., the compiler version and command line arguments),
16834 whether the executable includes debugging information for that file, and
16835 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
16837 whether the debugging information includes information about
16838 preprocessor macros.
16842 @kindex info sources
16844 Print the names of all source files in your program for which there is
16845 debugging information, organized into two lists: files whose symbols
16846 have already been read, and files whose symbols will be read when needed.
16848 @kindex info functions
16849 @item info functions
16850 Print the names and data types of all defined functions.
16852 @item info functions @var{regexp}
16853 Print the names and data types of all defined functions
16854 whose names contain a match for regular expression @var{regexp}.
16855 Thus, @samp{info fun step} finds all functions whose names
16856 include @code{step}; @samp{info fun ^step} finds those whose names
16857 start with @code{step}. If a function name contains characters
16858 that conflict with the regular expression language (e.g.@:
16859 @samp{operator*()}), they may be quoted with a backslash.
16861 @kindex info variables
16862 @item info variables
16863 Print the names and data types of all variables that are defined
16864 outside of functions (i.e.@: excluding local variables).
16866 @item info variables @var{regexp}
16867 Print the names and data types of all variables (except for local
16868 variables) whose names contain a match for regular expression
16871 @kindex info classes
16872 @cindex Objective-C, classes and selectors
16874 @itemx info classes @var{regexp}
16875 Display all Objective-C classes in your program, or
16876 (with the @var{regexp} argument) all those matching a particular regular
16879 @kindex info selectors
16880 @item info selectors
16881 @itemx info selectors @var{regexp}
16882 Display all Objective-C selectors in your program, or
16883 (with the @var{regexp} argument) all those matching a particular regular
16887 This was never implemented.
16888 @kindex info methods
16890 @itemx info methods @var{regexp}
16891 The @code{info methods} command permits the user to examine all defined
16892 methods within C@t{++} program, or (with the @var{regexp} argument) a
16893 specific set of methods found in the various C@t{++} classes. Many
16894 C@t{++} classes provide a large number of methods. Thus, the output
16895 from the @code{ptype} command can be overwhelming and hard to use. The
16896 @code{info-methods} command filters the methods, printing only those
16897 which match the regular-expression @var{regexp}.
16900 @cindex opaque data types
16901 @kindex set opaque-type-resolution
16902 @item set opaque-type-resolution on
16903 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
16904 declared as a pointer to a @code{struct}, @code{class}, or
16905 @code{union}---for example, @code{struct MyType *}---that is used in one
16906 source file although the full declaration of @code{struct MyType} is in
16907 another source file. The default is on.
16909 A change in the setting of this subcommand will not take effect until
16910 the next time symbols for a file are loaded.
16912 @item set opaque-type-resolution off
16913 Tell @value{GDBN} not to resolve opaque types. In this case, the type
16914 is printed as follows:
16916 @{<no data fields>@}
16919 @kindex show opaque-type-resolution
16920 @item show opaque-type-resolution
16921 Show whether opaque types are resolved or not.
16923 @kindex set print symbol-loading
16924 @cindex print messages when symbols are loaded
16925 @item set print symbol-loading
16926 @itemx set print symbol-loading full
16927 @itemx set print symbol-loading brief
16928 @itemx set print symbol-loading off
16929 The @code{set print symbol-loading} command allows you to control the
16930 printing of messages when @value{GDBN} loads symbol information.
16931 By default a message is printed for the executable and one for each
16932 shared library, and normally this is what you want. However, when
16933 debugging apps with large numbers of shared libraries these messages
16935 When set to @code{brief} a message is printed for each executable,
16936 and when @value{GDBN} loads a collection of shared libraries at once
16937 it will only print one message regardless of the number of shared
16938 libraries. When set to @code{off} no messages are printed.
16940 @kindex show print symbol-loading
16941 @item show print symbol-loading
16942 Show whether messages will be printed when a @value{GDBN} command
16943 entered from the keyboard causes symbol information to be loaded.
16945 @kindex maint print symbols
16946 @cindex symbol dump
16947 @kindex maint print psymbols
16948 @cindex partial symbol dump
16949 @kindex maint print msymbols
16950 @cindex minimal symbol dump
16951 @item maint print symbols @var{filename}
16952 @itemx maint print psymbols @var{filename}
16953 @itemx maint print msymbols @var{filename}
16954 Write a dump of debugging symbol data into the file @var{filename}.
16955 These commands are used to debug the @value{GDBN} symbol-reading code. Only
16956 symbols with debugging data are included. If you use @samp{maint print
16957 symbols}, @value{GDBN} includes all the symbols for which it has already
16958 collected full details: that is, @var{filename} reflects symbols for
16959 only those files whose symbols @value{GDBN} has read. You can use the
16960 command @code{info sources} to find out which files these are. If you
16961 use @samp{maint print psymbols} instead, the dump shows information about
16962 symbols that @value{GDBN} only knows partially---that is, symbols defined in
16963 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
16964 @samp{maint print msymbols} dumps just the minimal symbol information
16965 required for each object file from which @value{GDBN} has read some symbols.
16966 @xref{Files, ,Commands to Specify Files}, for a discussion of how
16967 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
16969 @kindex maint info symtabs
16970 @kindex maint info psymtabs
16971 @cindex listing @value{GDBN}'s internal symbol tables
16972 @cindex symbol tables, listing @value{GDBN}'s internal
16973 @cindex full symbol tables, listing @value{GDBN}'s internal
16974 @cindex partial symbol tables, listing @value{GDBN}'s internal
16975 @item maint info symtabs @r{[} @var{regexp} @r{]}
16976 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
16978 List the @code{struct symtab} or @code{struct partial_symtab}
16979 structures whose names match @var{regexp}. If @var{regexp} is not
16980 given, list them all. The output includes expressions which you can
16981 copy into a @value{GDBN} debugging this one to examine a particular
16982 structure in more detail. For example:
16985 (@value{GDBP}) maint info psymtabs dwarf2read
16986 @{ objfile /home/gnu/build/gdb/gdb
16987 ((struct objfile *) 0x82e69d0)
16988 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
16989 ((struct partial_symtab *) 0x8474b10)
16992 text addresses 0x814d3c8 -- 0x8158074
16993 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
16994 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
16995 dependencies (none)
16998 (@value{GDBP}) maint info symtabs
17002 We see that there is one partial symbol table whose filename contains
17003 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
17004 and we see that @value{GDBN} has not read in any symtabs yet at all.
17005 If we set a breakpoint on a function, that will cause @value{GDBN} to
17006 read the symtab for the compilation unit containing that function:
17009 (@value{GDBP}) break dwarf2_psymtab_to_symtab
17010 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
17012 (@value{GDBP}) maint info symtabs
17013 @{ objfile /home/gnu/build/gdb/gdb
17014 ((struct objfile *) 0x82e69d0)
17015 @{ symtab /home/gnu/src/gdb/dwarf2read.c
17016 ((struct symtab *) 0x86c1f38)
17019 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
17020 linetable ((struct linetable *) 0x8370fa0)
17021 debugformat DWARF 2
17027 @kindex maint set symbol-cache-size
17028 @cindex symbol cache size
17029 @item maint set symbol-cache-size @var{size}
17030 Set the size of the symbol cache to @var{size}.
17031 The default size is intended to be good enough for debugging
17032 most applications. This option exists to allow for experimenting
17033 with different sizes.
17035 @kindex maint show symbol-cache-size
17036 @item maint show symbol-cache-size
17037 Show the size of the symbol cache.
17039 @kindex maint print symbol-cache
17040 @cindex symbol cache, printing its contents
17041 @item maint print symbol-cache
17042 Print the contents of the symbol cache.
17043 This is useful when debugging symbol cache issues.
17045 @kindex maint print symbol-cache-statistics
17046 @cindex symbol cache, printing usage statistics
17047 @item maint print symbol-cache-statistics
17048 Print symbol cache usage statistics.
17049 This helps determine how well the cache is being utilized.
17051 @kindex maint flush-symbol-cache
17052 @cindex symbol cache, flushing
17053 @item maint flush-symbol-cache
17054 Flush the contents of the symbol cache, all entries are removed.
17055 This command is useful when debugging the symbol cache.
17056 It is also useful when collecting performance data.
17061 @chapter Altering Execution
17063 Once you think you have found an error in your program, you might want to
17064 find out for certain whether correcting the apparent error would lead to
17065 correct results in the rest of the run. You can find the answer by
17066 experiment, using the @value{GDBN} features for altering execution of the
17069 For example, you can store new values into variables or memory
17070 locations, give your program a signal, restart it at a different
17071 address, or even return prematurely from a function.
17074 * Assignment:: Assignment to variables
17075 * Jumping:: Continuing at a different address
17076 * Signaling:: Giving your program a signal
17077 * Returning:: Returning from a function
17078 * Calling:: Calling your program's functions
17079 * Patching:: Patching your program
17080 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
17084 @section Assignment to Variables
17087 @cindex setting variables
17088 To alter the value of a variable, evaluate an assignment expression.
17089 @xref{Expressions, ,Expressions}. For example,
17096 stores the value 4 into the variable @code{x}, and then prints the
17097 value of the assignment expression (which is 4).
17098 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
17099 information on operators in supported languages.
17101 @kindex set variable
17102 @cindex variables, setting
17103 If you are not interested in seeing the value of the assignment, use the
17104 @code{set} command instead of the @code{print} command. @code{set} is
17105 really the same as @code{print} except that the expression's value is
17106 not printed and is not put in the value history (@pxref{Value History,
17107 ,Value History}). The expression is evaluated only for its effects.
17109 If the beginning of the argument string of the @code{set} command
17110 appears identical to a @code{set} subcommand, use the @code{set
17111 variable} command instead of just @code{set}. This command is identical
17112 to @code{set} except for its lack of subcommands. For example, if your
17113 program has a variable @code{width}, you get an error if you try to set
17114 a new value with just @samp{set width=13}, because @value{GDBN} has the
17115 command @code{set width}:
17118 (@value{GDBP}) whatis width
17120 (@value{GDBP}) p width
17122 (@value{GDBP}) set width=47
17123 Invalid syntax in expression.
17127 The invalid expression, of course, is @samp{=47}. In
17128 order to actually set the program's variable @code{width}, use
17131 (@value{GDBP}) set var width=47
17134 Because the @code{set} command has many subcommands that can conflict
17135 with the names of program variables, it is a good idea to use the
17136 @code{set variable} command instead of just @code{set}. For example, if
17137 your program has a variable @code{g}, you run into problems if you try
17138 to set a new value with just @samp{set g=4}, because @value{GDBN} has
17139 the command @code{set gnutarget}, abbreviated @code{set g}:
17143 (@value{GDBP}) whatis g
17147 (@value{GDBP}) set g=4
17151 The program being debugged has been started already.
17152 Start it from the beginning? (y or n) y
17153 Starting program: /home/smith/cc_progs/a.out
17154 "/home/smith/cc_progs/a.out": can't open to read symbols:
17155 Invalid bfd target.
17156 (@value{GDBP}) show g
17157 The current BFD target is "=4".
17162 The program variable @code{g} did not change, and you silently set the
17163 @code{gnutarget} to an invalid value. In order to set the variable
17167 (@value{GDBP}) set var g=4
17170 @value{GDBN} allows more implicit conversions in assignments than C; you can
17171 freely store an integer value into a pointer variable or vice versa,
17172 and you can convert any structure to any other structure that is the
17173 same length or shorter.
17174 @comment FIXME: how do structs align/pad in these conversions?
17175 @comment /doc@cygnus.com 18dec1990
17177 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
17178 construct to generate a value of specified type at a specified address
17179 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
17180 to memory location @code{0x83040} as an integer (which implies a certain size
17181 and representation in memory), and
17184 set @{int@}0x83040 = 4
17188 stores the value 4 into that memory location.
17191 @section Continuing at a Different Address
17193 Ordinarily, when you continue your program, you do so at the place where
17194 it stopped, with the @code{continue} command. You can instead continue at
17195 an address of your own choosing, with the following commands:
17199 @kindex j @r{(@code{jump})}
17200 @item jump @var{location}
17201 @itemx j @var{location}
17202 Resume execution at @var{location}. Execution stops again immediately
17203 if there is a breakpoint there. @xref{Specify Location}, for a description
17204 of the different forms of @var{location}. It is common
17205 practice to use the @code{tbreak} command in conjunction with
17206 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
17208 The @code{jump} command does not change the current stack frame, or
17209 the stack pointer, or the contents of any memory location or any
17210 register other than the program counter. If @var{location} is in
17211 a different function from the one currently executing, the results may
17212 be bizarre if the two functions expect different patterns of arguments or
17213 of local variables. For this reason, the @code{jump} command requests
17214 confirmation if the specified line is not in the function currently
17215 executing. However, even bizarre results are predictable if you are
17216 well acquainted with the machine-language code of your program.
17219 On many systems, you can get much the same effect as the @code{jump}
17220 command by storing a new value into the register @code{$pc}. The
17221 difference is that this does not start your program running; it only
17222 changes the address of where it @emph{will} run when you continue. For
17230 makes the next @code{continue} command or stepping command execute at
17231 address @code{0x485}, rather than at the address where your program stopped.
17232 @xref{Continuing and Stepping, ,Continuing and Stepping}.
17234 The most common occasion to use the @code{jump} command is to back
17235 up---perhaps with more breakpoints set---over a portion of a program
17236 that has already executed, in order to examine its execution in more
17241 @section Giving your Program a Signal
17242 @cindex deliver a signal to a program
17246 @item signal @var{signal}
17247 Resume execution where your program is stopped, but immediately give it the
17248 signal @var{signal}. The @var{signal} can be the name or the number of a
17249 signal. For example, on many systems @code{signal 2} and @code{signal
17250 SIGINT} are both ways of sending an interrupt signal.
17252 Alternatively, if @var{signal} is zero, continue execution without
17253 giving a signal. This is useful when your program stopped on account of
17254 a signal and would ordinarily see the signal when resumed with the
17255 @code{continue} command; @samp{signal 0} causes it to resume without a
17258 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
17259 delivered to the currently selected thread, not the thread that last
17260 reported a stop. This includes the situation where a thread was
17261 stopped due to a signal. So if you want to continue execution
17262 suppressing the signal that stopped a thread, you should select that
17263 same thread before issuing the @samp{signal 0} command. If you issue
17264 the @samp{signal 0} command with another thread as the selected one,
17265 @value{GDBN} detects that and asks for confirmation.
17267 Invoking the @code{signal} command is not the same as invoking the
17268 @code{kill} utility from the shell. Sending a signal with @code{kill}
17269 causes @value{GDBN} to decide what to do with the signal depending on
17270 the signal handling tables (@pxref{Signals}). The @code{signal} command
17271 passes the signal directly to your program.
17273 @code{signal} does not repeat when you press @key{RET} a second time
17274 after executing the command.
17276 @kindex queue-signal
17277 @item queue-signal @var{signal}
17278 Queue @var{signal} to be delivered immediately to the current thread
17279 when execution of the thread resumes. The @var{signal} can be the name or
17280 the number of a signal. For example, on many systems @code{signal 2} and
17281 @code{signal SIGINT} are both ways of sending an interrupt signal.
17282 The handling of the signal must be set to pass the signal to the program,
17283 otherwise @value{GDBN} will report an error.
17284 You can control the handling of signals from @value{GDBN} with the
17285 @code{handle} command (@pxref{Signals}).
17287 Alternatively, if @var{signal} is zero, any currently queued signal
17288 for the current thread is discarded and when execution resumes no signal
17289 will be delivered. This is useful when your program stopped on account
17290 of a signal and would ordinarily see the signal when resumed with the
17291 @code{continue} command.
17293 This command differs from the @code{signal} command in that the signal
17294 is just queued, execution is not resumed. And @code{queue-signal} cannot
17295 be used to pass a signal whose handling state has been set to @code{nopass}
17300 @xref{stepping into signal handlers}, for information on how stepping
17301 commands behave when the thread has a signal queued.
17304 @section Returning from a Function
17307 @cindex returning from a function
17310 @itemx return @var{expression}
17311 You can cancel execution of a function call with the @code{return}
17312 command. If you give an
17313 @var{expression} argument, its value is used as the function's return
17317 When you use @code{return}, @value{GDBN} discards the selected stack frame
17318 (and all frames within it). You can think of this as making the
17319 discarded frame return prematurely. If you wish to specify a value to
17320 be returned, give that value as the argument to @code{return}.
17322 This pops the selected stack frame (@pxref{Selection, ,Selecting a
17323 Frame}), and any other frames inside of it, leaving its caller as the
17324 innermost remaining frame. That frame becomes selected. The
17325 specified value is stored in the registers used for returning values
17328 The @code{return} command does not resume execution; it leaves the
17329 program stopped in the state that would exist if the function had just
17330 returned. In contrast, the @code{finish} command (@pxref{Continuing
17331 and Stepping, ,Continuing and Stepping}) resumes execution until the
17332 selected stack frame returns naturally.
17334 @value{GDBN} needs to know how the @var{expression} argument should be set for
17335 the inferior. The concrete registers assignment depends on the OS ABI and the
17336 type being returned by the selected stack frame. For example it is common for
17337 OS ABI to return floating point values in FPU registers while integer values in
17338 CPU registers. Still some ABIs return even floating point values in CPU
17339 registers. Larger integer widths (such as @code{long long int}) also have
17340 specific placement rules. @value{GDBN} already knows the OS ABI from its
17341 current target so it needs to find out also the type being returned to make the
17342 assignment into the right register(s).
17344 Normally, the selected stack frame has debug info. @value{GDBN} will always
17345 use the debug info instead of the implicit type of @var{expression} when the
17346 debug info is available. For example, if you type @kbd{return -1}, and the
17347 function in the current stack frame is declared to return a @code{long long
17348 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
17349 into a @code{long long int}:
17352 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
17354 (@value{GDBP}) return -1
17355 Make func return now? (y or n) y
17356 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
17357 43 printf ("result=%lld\n", func ());
17361 However, if the selected stack frame does not have a debug info, e.g., if the
17362 function was compiled without debug info, @value{GDBN} has to find out the type
17363 to return from user. Specifying a different type by mistake may set the value
17364 in different inferior registers than the caller code expects. For example,
17365 typing @kbd{return -1} with its implicit type @code{int} would set only a part
17366 of a @code{long long int} result for a debug info less function (on 32-bit
17367 architectures). Therefore the user is required to specify the return type by
17368 an appropriate cast explicitly:
17371 Breakpoint 2, 0x0040050b in func ()
17372 (@value{GDBP}) return -1
17373 Return value type not available for selected stack frame.
17374 Please use an explicit cast of the value to return.
17375 (@value{GDBP}) return (long long int) -1
17376 Make selected stack frame return now? (y or n) y
17377 #0 0x00400526 in main ()
17382 @section Calling Program Functions
17385 @cindex calling functions
17386 @cindex inferior functions, calling
17387 @item print @var{expr}
17388 Evaluate the expression @var{expr} and display the resulting value.
17389 The expression may include calls to functions in the program being
17393 @item call @var{expr}
17394 Evaluate the expression @var{expr} without displaying @code{void}
17397 You can use this variant of the @code{print} command if you want to
17398 execute a function from your program that does not return anything
17399 (a.k.a.@: @dfn{a void function}), but without cluttering the output
17400 with @code{void} returned values that @value{GDBN} will otherwise
17401 print. If the result is not void, it is printed and saved in the
17405 It is possible for the function you call via the @code{print} or
17406 @code{call} command to generate a signal (e.g., if there's a bug in
17407 the function, or if you passed it incorrect arguments). What happens
17408 in that case is controlled by the @code{set unwindonsignal} command.
17410 Similarly, with a C@t{++} program it is possible for the function you
17411 call via the @code{print} or @code{call} command to generate an
17412 exception that is not handled due to the constraints of the dummy
17413 frame. In this case, any exception that is raised in the frame, but has
17414 an out-of-frame exception handler will not be found. GDB builds a
17415 dummy-frame for the inferior function call, and the unwinder cannot
17416 seek for exception handlers outside of this dummy-frame. What happens
17417 in that case is controlled by the
17418 @code{set unwind-on-terminating-exception} command.
17421 @item set unwindonsignal
17422 @kindex set unwindonsignal
17423 @cindex unwind stack in called functions
17424 @cindex call dummy stack unwinding
17425 Set unwinding of the stack if a signal is received while in a function
17426 that @value{GDBN} called in the program being debugged. If set to on,
17427 @value{GDBN} unwinds the stack it created for the call and restores
17428 the context to what it was before the call. If set to off (the
17429 default), @value{GDBN} stops in the frame where the signal was
17432 @item show unwindonsignal
17433 @kindex show unwindonsignal
17434 Show the current setting of stack unwinding in the functions called by
17437 @item set unwind-on-terminating-exception
17438 @kindex set unwind-on-terminating-exception
17439 @cindex unwind stack in called functions with unhandled exceptions
17440 @cindex call dummy stack unwinding on unhandled exception.
17441 Set unwinding of the stack if a C@t{++} exception is raised, but left
17442 unhandled while in a function that @value{GDBN} called in the program being
17443 debugged. If set to on (the default), @value{GDBN} unwinds the stack
17444 it created for the call and restores the context to what it was before
17445 the call. If set to off, @value{GDBN} the exception is delivered to
17446 the default C@t{++} exception handler and the inferior terminated.
17448 @item show unwind-on-terminating-exception
17449 @kindex show unwind-on-terminating-exception
17450 Show the current setting of stack unwinding in the functions called by
17455 @cindex weak alias functions
17456 Sometimes, a function you wish to call is actually a @dfn{weak alias}
17457 for another function. In such case, @value{GDBN} might not pick up
17458 the type information, including the types of the function arguments,
17459 which causes @value{GDBN} to call the inferior function incorrectly.
17460 As a result, the called function will function erroneously and may
17461 even crash. A solution to that is to use the name of the aliased
17465 @section Patching Programs
17467 @cindex patching binaries
17468 @cindex writing into executables
17469 @cindex writing into corefiles
17471 By default, @value{GDBN} opens the file containing your program's
17472 executable code (or the corefile) read-only. This prevents accidental
17473 alterations to machine code; but it also prevents you from intentionally
17474 patching your program's binary.
17476 If you'd like to be able to patch the binary, you can specify that
17477 explicitly with the @code{set write} command. For example, you might
17478 want to turn on internal debugging flags, or even to make emergency
17484 @itemx set write off
17485 If you specify @samp{set write on}, @value{GDBN} opens executable and
17486 core files for both reading and writing; if you specify @kbd{set write
17487 off} (the default), @value{GDBN} opens them read-only.
17489 If you have already loaded a file, you must load it again (using the
17490 @code{exec-file} or @code{core-file} command) after changing @code{set
17491 write}, for your new setting to take effect.
17495 Display whether executable files and core files are opened for writing
17496 as well as reading.
17499 @node Compiling and Injecting Code
17500 @section Compiling and injecting code in @value{GDBN}
17501 @cindex injecting code
17502 @cindex writing into executables
17503 @cindex compiling code
17505 @value{GDBN} supports on-demand compilation and code injection into
17506 programs running under @value{GDBN}. GCC 5.0 or higher built with
17507 @file{libcc1.so} must be installed for this functionality to be enabled.
17508 This functionality is implemented with the following commands.
17511 @kindex compile code
17512 @item compile code @var{source-code}
17513 @itemx compile code -raw @var{--} @var{source-code}
17514 Compile @var{source-code} with the compiler language found as the current
17515 language in @value{GDBN} (@pxref{Languages}). If compilation and
17516 injection is not supported with the current language specified in
17517 @value{GDBN}, or the compiler does not support this feature, an error
17518 message will be printed. If @var{source-code} compiles and links
17519 successfully, @value{GDBN} will load the object-code emitted,
17520 and execute it within the context of the currently selected inferior.
17521 It is important to note that the compiled code is executed immediately.
17522 After execution, the compiled code is removed from @value{GDBN} and any
17523 new types or variables you have defined will be deleted.
17525 The command allows you to specify @var{source-code} in two ways.
17526 The simplest method is to provide a single line of code to the command.
17530 compile code printf ("hello world\n");
17533 If you specify options on the command line as well as source code, they
17534 may conflict. The @samp{--} delimiter can be used to separate options
17535 from actual source code. E.g.:
17538 compile code -r -- printf ("hello world\n");
17541 Alternatively you can enter source code as multiple lines of text. To
17542 enter this mode, invoke the @samp{compile code} command without any text
17543 following the command. This will start the multiple-line editor and
17544 allow you to type as many lines of source code as required. When you
17545 have completed typing, enter @samp{end} on its own line to exit the
17550 >printf ("hello\n");
17551 >printf ("world\n");
17555 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
17556 provided @var{source-code} in a callable scope. In this case, you must
17557 specify the entry point of the code by defining a function named
17558 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
17559 inferior. Using @samp{-raw} option may be needed for example when
17560 @var{source-code} requires @samp{#include} lines which may conflict with
17561 inferior symbols otherwise.
17563 @kindex compile file
17564 @item compile file @var{filename}
17565 @itemx compile file -raw @var{filename}
17566 Like @code{compile code}, but take the source code from @var{filename}.
17569 compile file /home/user/example.c
17574 @item compile print @var{expr}
17575 @itemx compile print /@var{f} @var{expr}
17576 Compile and execute @var{expr} with the compiler language found as the
17577 current language in @value{GDBN} (@pxref{Languages}). By default the
17578 value of @var{expr} is printed in a format appropriate to its data type;
17579 you can choose a different format by specifying @samp{/@var{f}}, where
17580 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
17583 @item compile print
17584 @itemx compile print /@var{f}
17585 @cindex reprint the last value
17586 Alternatively you can enter the expression (source code producing it) as
17587 multiple lines of text. To enter this mode, invoke the @samp{compile print}
17588 command without any text following the command. This will start the
17589 multiple-line editor.
17593 The process of compiling and injecting the code can be inspected using:
17596 @anchor{set debug compile}
17597 @item set debug compile
17598 @cindex compile command debugging info
17599 Turns on or off display of @value{GDBN} process of compiling and
17600 injecting the code. The default is off.
17602 @item show debug compile
17603 Displays the current state of displaying @value{GDBN} process of
17604 compiling and injecting the code.
17607 @subsection Compilation options for the @code{compile} command
17609 @value{GDBN} needs to specify the right compilation options for the code
17610 to be injected, in part to make its ABI compatible with the inferior
17611 and in part to make the injected code compatible with @value{GDBN}'s
17615 The options used, in increasing precedence:
17618 @item target architecture and OS options (@code{gdbarch})
17619 These options depend on target processor type and target operating
17620 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
17621 (@code{-m64}) compilation option.
17623 @item compilation options recorded in the target
17624 @value{NGCC} (since version 4.7) stores the options used for compilation
17625 into @code{DW_AT_producer} part of DWARF debugging information according
17626 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
17627 explicitly specify @code{-g} during inferior compilation otherwise
17628 @value{NGCC} produces no DWARF. This feature is only relevant for
17629 platforms where @code{-g} produces DWARF by default, otherwise one may
17630 try to enforce DWARF by using @code{-gdwarf-4}.
17632 @item compilation options set by @code{set compile-args}
17636 You can override compilation options using the following command:
17639 @item set compile-args
17640 @cindex compile command options override
17641 Set compilation options used for compiling and injecting code with the
17642 @code{compile} commands. These options override any conflicting ones
17643 from the target architecture and/or options stored during inferior
17646 @item show compile-args
17647 Displays the current state of compilation options override.
17648 This does not show all the options actually used during compilation,
17649 use @ref{set debug compile} for that.
17652 @subsection Caveats when using the @code{compile} command
17654 There are a few caveats to keep in mind when using the @code{compile}
17655 command. As the caveats are different per language, the table below
17656 highlights specific issues on a per language basis.
17659 @item C code examples and caveats
17660 When the language in @value{GDBN} is set to @samp{C}, the compiler will
17661 attempt to compile the source code with a @samp{C} compiler. The source
17662 code provided to the @code{compile} command will have much the same
17663 access to variables and types as it normally would if it were part of
17664 the program currently being debugged in @value{GDBN}.
17666 Below is a sample program that forms the basis of the examples that
17667 follow. This program has been compiled and loaded into @value{GDBN},
17668 much like any other normal debugging session.
17671 void function1 (void)
17674 printf ("function 1\n");
17677 void function2 (void)
17692 For the purposes of the examples in this section, the program above has
17693 been compiled, loaded into @value{GDBN}, stopped at the function
17694 @code{main}, and @value{GDBN} is awaiting input from the user.
17696 To access variables and types for any program in @value{GDBN}, the
17697 program must be compiled and packaged with debug information. The
17698 @code{compile} command is not an exception to this rule. Without debug
17699 information, you can still use the @code{compile} command, but you will
17700 be very limited in what variables and types you can access.
17702 So with that in mind, the example above has been compiled with debug
17703 information enabled. The @code{compile} command will have access to
17704 all variables and types (except those that may have been optimized
17705 out). Currently, as @value{GDBN} has stopped the program in the
17706 @code{main} function, the @code{compile} command would have access to
17707 the variable @code{k}. You could invoke the @code{compile} command
17708 and type some source code to set the value of @code{k}. You can also
17709 read it, or do anything with that variable you would normally do in
17710 @code{C}. Be aware that changes to inferior variables in the
17711 @code{compile} command are persistent. In the following example:
17714 compile code k = 3;
17718 the variable @code{k} is now 3. It will retain that value until
17719 something else in the example program changes it, or another
17720 @code{compile} command changes it.
17722 Normal scope and access rules apply to source code compiled and
17723 injected by the @code{compile} command. In the example, the variables
17724 @code{j} and @code{k} are not accessible yet, because the program is
17725 currently stopped in the @code{main} function, where these variables
17726 are not in scope. Therefore, the following command
17729 compile code j = 3;
17733 will result in a compilation error message.
17735 Once the program is continued, execution will bring these variables in
17736 scope, and they will become accessible; then the code you specify via
17737 the @code{compile} command will be able to access them.
17739 You can create variables and types with the @code{compile} command as
17740 part of your source code. Variables and types that are created as part
17741 of the @code{compile} command are not visible to the rest of the program for
17742 the duration of its run. This example is valid:
17745 compile code int ff = 5; printf ("ff is %d\n", ff);
17748 However, if you were to type the following into @value{GDBN} after that
17749 command has completed:
17752 compile code printf ("ff is %d\n'', ff);
17756 a compiler error would be raised as the variable @code{ff} no longer
17757 exists. Object code generated and injected by the @code{compile}
17758 command is removed when its execution ends. Caution is advised
17759 when assigning to program variables values of variables created by the
17760 code submitted to the @code{compile} command. This example is valid:
17763 compile code int ff = 5; k = ff;
17766 The value of the variable @code{ff} is assigned to @code{k}. The variable
17767 @code{k} does not require the existence of @code{ff} to maintain the value
17768 it has been assigned. However, pointers require particular care in
17769 assignment. If the source code compiled with the @code{compile} command
17770 changed the address of a pointer in the example program, perhaps to a
17771 variable created in the @code{compile} command, that pointer would point
17772 to an invalid location when the command exits. The following example
17773 would likely cause issues with your debugged program:
17776 compile code int ff = 5; p = &ff;
17779 In this example, @code{p} would point to @code{ff} when the
17780 @code{compile} command is executing the source code provided to it.
17781 However, as variables in the (example) program persist with their
17782 assigned values, the variable @code{p} would point to an invalid
17783 location when the command exists. A general rule should be followed
17784 in that you should either assign @code{NULL} to any assigned pointers,
17785 or restore a valid location to the pointer before the command exits.
17787 Similar caution must be exercised with any structs, unions, and typedefs
17788 defined in @code{compile} command. Types defined in the @code{compile}
17789 command will no longer be available in the next @code{compile} command.
17790 Therefore, if you cast a variable to a type defined in the
17791 @code{compile} command, care must be taken to ensure that any future
17792 need to resolve the type can be achieved.
17795 (gdb) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
17796 (gdb) compile code printf ("%d\n", ((struct a *) argv)->a);
17797 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
17798 Compilation failed.
17799 (gdb) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
17803 Variables that have been optimized away by the compiler are not
17804 accessible to the code submitted to the @code{compile} command.
17805 Access to those variables will generate a compiler error which @value{GDBN}
17806 will print to the console.
17809 @subsection Compiler search for the @code{compile} command
17811 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged which
17812 may not be obvious for remote targets of different architecture than where
17813 @value{GDBN} is running. Environment variable @code{PATH} (@code{PATH} from
17814 shell that executed @value{GDBN}, not the one set by @value{GDBN}
17815 command @code{set environment}). @xref{Environment}. @code{PATH} on
17816 @value{GDBN} host is searched for @value{NGCC} binary matching the
17817 target architecture and operating system.
17819 Specifically @code{PATH} is searched for binaries matching regular expression
17820 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
17821 debugged. @var{arch} is processor name --- multiarch is supported, so for
17822 example both @code{i386} and @code{x86_64} targets look for pattern
17823 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
17824 for pattern @code{s390x?}. @var{os} is currently supported only for
17825 pattern @code{linux(-gnu)?}.
17828 @chapter @value{GDBN} Files
17830 @value{GDBN} needs to know the file name of the program to be debugged,
17831 both in order to read its symbol table and in order to start your
17832 program. To debug a core dump of a previous run, you must also tell
17833 @value{GDBN} the name of the core dump file.
17836 * Files:: Commands to specify files
17837 * File Caching:: Information about @value{GDBN}'s file caching
17838 * Separate Debug Files:: Debugging information in separate files
17839 * MiniDebugInfo:: Debugging information in a special section
17840 * Index Files:: Index files speed up GDB
17841 * Symbol Errors:: Errors reading symbol files
17842 * Data Files:: GDB data files
17846 @section Commands to Specify Files
17848 @cindex symbol table
17849 @cindex core dump file
17851 You may want to specify executable and core dump file names. The usual
17852 way to do this is at start-up time, using the arguments to
17853 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
17854 Out of @value{GDBN}}).
17856 Occasionally it is necessary to change to a different file during a
17857 @value{GDBN} session. Or you may run @value{GDBN} and forget to
17858 specify a file you want to use. Or you are debugging a remote target
17859 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
17860 Program}). In these situations the @value{GDBN} commands to specify
17861 new files are useful.
17864 @cindex executable file
17866 @item file @var{filename}
17867 Use @var{filename} as the program to be debugged. It is read for its
17868 symbols and for the contents of pure memory. It is also the program
17869 executed when you use the @code{run} command. If you do not specify a
17870 directory and the file is not found in the @value{GDBN} working directory,
17871 @value{GDBN} uses the environment variable @code{PATH} as a list of
17872 directories to search, just as the shell does when looking for a program
17873 to run. You can change the value of this variable, for both @value{GDBN}
17874 and your program, using the @code{path} command.
17876 @cindex unlinked object files
17877 @cindex patching object files
17878 You can load unlinked object @file{.o} files into @value{GDBN} using
17879 the @code{file} command. You will not be able to ``run'' an object
17880 file, but you can disassemble functions and inspect variables. Also,
17881 if the underlying BFD functionality supports it, you could use
17882 @kbd{gdb -write} to patch object files using this technique. Note
17883 that @value{GDBN} can neither interpret nor modify relocations in this
17884 case, so branches and some initialized variables will appear to go to
17885 the wrong place. But this feature is still handy from time to time.
17888 @code{file} with no argument makes @value{GDBN} discard any information it
17889 has on both executable file and the symbol table.
17892 @item exec-file @r{[} @var{filename} @r{]}
17893 Specify that the program to be run (but not the symbol table) is found
17894 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
17895 if necessary to locate your program. Omitting @var{filename} means to
17896 discard information on the executable file.
17898 @kindex symbol-file
17899 @item symbol-file @r{[} @var{filename} @r{]}
17900 Read symbol table information from file @var{filename}. @code{PATH} is
17901 searched when necessary. Use the @code{file} command to get both symbol
17902 table and program to run from the same file.
17904 @code{symbol-file} with no argument clears out @value{GDBN} information on your
17905 program's symbol table.
17907 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
17908 some breakpoints and auto-display expressions. This is because they may
17909 contain pointers to the internal data recording symbols and data types,
17910 which are part of the old symbol table data being discarded inside
17913 @code{symbol-file} does not repeat if you press @key{RET} again after
17916 When @value{GDBN} is configured for a particular environment, it
17917 understands debugging information in whatever format is the standard
17918 generated for that environment; you may use either a @sc{gnu} compiler, or
17919 other compilers that adhere to the local conventions.
17920 Best results are usually obtained from @sc{gnu} compilers; for example,
17921 using @code{@value{NGCC}} you can generate debugging information for
17924 For most kinds of object files, with the exception of old SVR3 systems
17925 using COFF, the @code{symbol-file} command does not normally read the
17926 symbol table in full right away. Instead, it scans the symbol table
17927 quickly to find which source files and which symbols are present. The
17928 details are read later, one source file at a time, as they are needed.
17930 The purpose of this two-stage reading strategy is to make @value{GDBN}
17931 start up faster. For the most part, it is invisible except for
17932 occasional pauses while the symbol table details for a particular source
17933 file are being read. (The @code{set verbose} command can turn these
17934 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
17935 Warnings and Messages}.)
17937 We have not implemented the two-stage strategy for COFF yet. When the
17938 symbol table is stored in COFF format, @code{symbol-file} reads the
17939 symbol table data in full right away. Note that ``stabs-in-COFF''
17940 still does the two-stage strategy, since the debug info is actually
17944 @cindex reading symbols immediately
17945 @cindex symbols, reading immediately
17946 @item symbol-file @r{[} -readnow @r{]} @var{filename}
17947 @itemx file @r{[} -readnow @r{]} @var{filename}
17948 You can override the @value{GDBN} two-stage strategy for reading symbol
17949 tables by using the @samp{-readnow} option with any of the commands that
17950 load symbol table information, if you want to be sure @value{GDBN} has the
17951 entire symbol table available.
17953 @c FIXME: for now no mention of directories, since this seems to be in
17954 @c flux. 13mar1992 status is that in theory GDB would look either in
17955 @c current dir or in same dir as myprog; but issues like competing
17956 @c GDB's, or clutter in system dirs, mean that in practice right now
17957 @c only current dir is used. FFish says maybe a special GDB hierarchy
17958 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
17962 @item core-file @r{[}@var{filename}@r{]}
17964 Specify the whereabouts of a core dump file to be used as the ``contents
17965 of memory''. Traditionally, core files contain only some parts of the
17966 address space of the process that generated them; @value{GDBN} can access the
17967 executable file itself for other parts.
17969 @code{core-file} with no argument specifies that no core file is
17972 Note that the core file is ignored when your program is actually running
17973 under @value{GDBN}. So, if you have been running your program and you
17974 wish to debug a core file instead, you must kill the subprocess in which
17975 the program is running. To do this, use the @code{kill} command
17976 (@pxref{Kill Process, ,Killing the Child Process}).
17978 @kindex add-symbol-file
17979 @cindex dynamic linking
17980 @item add-symbol-file @var{filename} @var{address}
17981 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
17982 @itemx add-symbol-file @var{filename} @var{address} -s @var{section} @var{address} @dots{}
17983 The @code{add-symbol-file} command reads additional symbol table
17984 information from the file @var{filename}. You would use this command
17985 when @var{filename} has been dynamically loaded (by some other means)
17986 into the program that is running. The @var{address} should give the memory
17987 address at which the file has been loaded; @value{GDBN} cannot figure
17988 this out for itself. You can additionally specify an arbitrary number
17989 of @samp{-s @var{section} @var{address}} pairs, to give an explicit
17990 section name and base address for that section. You can specify any
17991 @var{address} as an expression.
17993 The symbol table of the file @var{filename} is added to the symbol table
17994 originally read with the @code{symbol-file} command. You can use the
17995 @code{add-symbol-file} command any number of times; the new symbol data
17996 thus read is kept in addition to the old.
17998 Changes can be reverted using the command @code{remove-symbol-file}.
18000 @cindex relocatable object files, reading symbols from
18001 @cindex object files, relocatable, reading symbols from
18002 @cindex reading symbols from relocatable object files
18003 @cindex symbols, reading from relocatable object files
18004 @cindex @file{.o} files, reading symbols from
18005 Although @var{filename} is typically a shared library file, an
18006 executable file, or some other object file which has been fully
18007 relocated for loading into a process, you can also load symbolic
18008 information from relocatable @file{.o} files, as long as:
18012 the file's symbolic information refers only to linker symbols defined in
18013 that file, not to symbols defined by other object files,
18015 every section the file's symbolic information refers to has actually
18016 been loaded into the inferior, as it appears in the file, and
18018 you can determine the address at which every section was loaded, and
18019 provide these to the @code{add-symbol-file} command.
18023 Some embedded operating systems, like Sun Chorus and VxWorks, can load
18024 relocatable files into an already running program; such systems
18025 typically make the requirements above easy to meet. However, it's
18026 important to recognize that many native systems use complex link
18027 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
18028 assembly, for example) that make the requirements difficult to meet. In
18029 general, one cannot assume that using @code{add-symbol-file} to read a
18030 relocatable object file's symbolic information will have the same effect
18031 as linking the relocatable object file into the program in the normal
18034 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
18036 @kindex remove-symbol-file
18037 @item remove-symbol-file @var{filename}
18038 @item remove-symbol-file -a @var{address}
18039 Remove a symbol file added via the @code{add-symbol-file} command. The
18040 file to remove can be identified by its @var{filename} or by an @var{address}
18041 that lies within the boundaries of this symbol file in memory. Example:
18044 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
18045 add symbol table from file "/home/user/gdb/mylib.so" at
18046 .text_addr = 0x7ffff7ff9480
18048 Reading symbols from /home/user/gdb/mylib.so...done.
18049 (gdb) remove-symbol-file -a 0x7ffff7ff9480
18050 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
18055 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
18057 @kindex add-symbol-file-from-memory
18058 @cindex @code{syscall DSO}
18059 @cindex load symbols from memory
18060 @item add-symbol-file-from-memory @var{address}
18061 Load symbols from the given @var{address} in a dynamically loaded
18062 object file whose image is mapped directly into the inferior's memory.
18063 For example, the Linux kernel maps a @code{syscall DSO} into each
18064 process's address space; this DSO provides kernel-specific code for
18065 some system calls. The argument can be any expression whose
18066 evaluation yields the address of the file's shared object file header.
18067 For this command to work, you must have used @code{symbol-file} or
18068 @code{exec-file} commands in advance.
18071 @item section @var{section} @var{addr}
18072 The @code{section} command changes the base address of the named
18073 @var{section} of the exec file to @var{addr}. This can be used if the
18074 exec file does not contain section addresses, (such as in the
18075 @code{a.out} format), or when the addresses specified in the file
18076 itself are wrong. Each section must be changed separately. The
18077 @code{info files} command, described below, lists all the sections and
18081 @kindex info target
18084 @code{info files} and @code{info target} are synonymous; both print the
18085 current target (@pxref{Targets, ,Specifying a Debugging Target}),
18086 including the names of the executable and core dump files currently in
18087 use by @value{GDBN}, and the files from which symbols were loaded. The
18088 command @code{help target} lists all possible targets rather than
18091 @kindex maint info sections
18092 @item maint info sections
18093 Another command that can give you extra information about program sections
18094 is @code{maint info sections}. In addition to the section information
18095 displayed by @code{info files}, this command displays the flags and file
18096 offset of each section in the executable and core dump files. In addition,
18097 @code{maint info sections} provides the following command options (which
18098 may be arbitrarily combined):
18102 Display sections for all loaded object files, including shared libraries.
18103 @item @var{sections}
18104 Display info only for named @var{sections}.
18105 @item @var{section-flags}
18106 Display info only for sections for which @var{section-flags} are true.
18107 The section flags that @value{GDBN} currently knows about are:
18110 Section will have space allocated in the process when loaded.
18111 Set for all sections except those containing debug information.
18113 Section will be loaded from the file into the child process memory.
18114 Set for pre-initialized code and data, clear for @code{.bss} sections.
18116 Section needs to be relocated before loading.
18118 Section cannot be modified by the child process.
18120 Section contains executable code only.
18122 Section contains data only (no executable code).
18124 Section will reside in ROM.
18126 Section contains data for constructor/destructor lists.
18128 Section is not empty.
18130 An instruction to the linker to not output the section.
18131 @item COFF_SHARED_LIBRARY
18132 A notification to the linker that the section contains
18133 COFF shared library information.
18135 Section contains common symbols.
18138 @kindex set trust-readonly-sections
18139 @cindex read-only sections
18140 @item set trust-readonly-sections on
18141 Tell @value{GDBN} that readonly sections in your object file
18142 really are read-only (i.e.@: that their contents will not change).
18143 In that case, @value{GDBN} can fetch values from these sections
18144 out of the object file, rather than from the target program.
18145 For some targets (notably embedded ones), this can be a significant
18146 enhancement to debugging performance.
18148 The default is off.
18150 @item set trust-readonly-sections off
18151 Tell @value{GDBN} not to trust readonly sections. This means that
18152 the contents of the section might change while the program is running,
18153 and must therefore be fetched from the target when needed.
18155 @item show trust-readonly-sections
18156 Show the current setting of trusting readonly sections.
18159 All file-specifying commands allow both absolute and relative file names
18160 as arguments. @value{GDBN} always converts the file name to an absolute file
18161 name and remembers it that way.
18163 @cindex shared libraries
18164 @anchor{Shared Libraries}
18165 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
18166 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
18167 DSBT (TIC6X) shared libraries.
18169 On MS-Windows @value{GDBN} must be linked with the Expat library to support
18170 shared libraries. @xref{Expat}.
18172 @value{GDBN} automatically loads symbol definitions from shared libraries
18173 when you use the @code{run} command, or when you examine a core file.
18174 (Before you issue the @code{run} command, @value{GDBN} does not understand
18175 references to a function in a shared library, however---unless you are
18176 debugging a core file).
18178 @c FIXME: some @value{GDBN} release may permit some refs to undef
18179 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
18180 @c FIXME...lib; check this from time to time when updating manual
18182 There are times, however, when you may wish to not automatically load
18183 symbol definitions from shared libraries, such as when they are
18184 particularly large or there are many of them.
18186 To control the automatic loading of shared library symbols, use the
18190 @kindex set auto-solib-add
18191 @item set auto-solib-add @var{mode}
18192 If @var{mode} is @code{on}, symbols from all shared object libraries
18193 will be loaded automatically when the inferior begins execution, you
18194 attach to an independently started inferior, or when the dynamic linker
18195 informs @value{GDBN} that a new library has been loaded. If @var{mode}
18196 is @code{off}, symbols must be loaded manually, using the
18197 @code{sharedlibrary} command. The default value is @code{on}.
18199 @cindex memory used for symbol tables
18200 If your program uses lots of shared libraries with debug info that
18201 takes large amounts of memory, you can decrease the @value{GDBN}
18202 memory footprint by preventing it from automatically loading the
18203 symbols from shared libraries. To that end, type @kbd{set
18204 auto-solib-add off} before running the inferior, then load each
18205 library whose debug symbols you do need with @kbd{sharedlibrary
18206 @var{regexp}}, where @var{regexp} is a regular expression that matches
18207 the libraries whose symbols you want to be loaded.
18209 @kindex show auto-solib-add
18210 @item show auto-solib-add
18211 Display the current autoloading mode.
18214 @cindex load shared library
18215 To explicitly load shared library symbols, use the @code{sharedlibrary}
18219 @kindex info sharedlibrary
18221 @item info share @var{regex}
18222 @itemx info sharedlibrary @var{regex}
18223 Print the names of the shared libraries which are currently loaded
18224 that match @var{regex}. If @var{regex} is omitted then print
18225 all shared libraries that are loaded.
18228 @item info dll @var{regex}
18229 This is an alias of @code{info sharedlibrary}.
18231 @kindex sharedlibrary
18233 @item sharedlibrary @var{regex}
18234 @itemx share @var{regex}
18235 Load shared object library symbols for files matching a
18236 Unix regular expression.
18237 As with files loaded automatically, it only loads shared libraries
18238 required by your program for a core file or after typing @code{run}. If
18239 @var{regex} is omitted all shared libraries required by your program are
18242 @item nosharedlibrary
18243 @kindex nosharedlibrary
18244 @cindex unload symbols from shared libraries
18245 Unload all shared object library symbols. This discards all symbols
18246 that have been loaded from all shared libraries. Symbols from shared
18247 libraries that were loaded by explicit user requests are not
18251 Sometimes you may wish that @value{GDBN} stops and gives you control
18252 when any of shared library events happen. The best way to do this is
18253 to use @code{catch load} and @code{catch unload} (@pxref{Set
18256 @value{GDBN} also supports the the @code{set stop-on-solib-events}
18257 command for this. This command exists for historical reasons. It is
18258 less useful than setting a catchpoint, because it does not allow for
18259 conditions or commands as a catchpoint does.
18262 @item set stop-on-solib-events
18263 @kindex set stop-on-solib-events
18264 This command controls whether @value{GDBN} should give you control
18265 when the dynamic linker notifies it about some shared library event.
18266 The most common event of interest is loading or unloading of a new
18269 @item show stop-on-solib-events
18270 @kindex show stop-on-solib-events
18271 Show whether @value{GDBN} stops and gives you control when shared
18272 library events happen.
18275 Shared libraries are also supported in many cross or remote debugging
18276 configurations. @value{GDBN} needs to have access to the target's libraries;
18277 this can be accomplished either by providing copies of the libraries
18278 on the host system, or by asking @value{GDBN} to automatically retrieve the
18279 libraries from the target. If copies of the target libraries are
18280 provided, they need to be the same as the target libraries, although the
18281 copies on the target can be stripped as long as the copies on the host are
18284 @cindex where to look for shared libraries
18285 For remote debugging, you need to tell @value{GDBN} where the target
18286 libraries are, so that it can load the correct copies---otherwise, it
18287 may try to load the host's libraries. @value{GDBN} has two variables
18288 to specify the search directories for target libraries.
18291 @cindex prefix for executable and shared library file names
18292 @cindex system root, alternate
18293 @kindex set solib-absolute-prefix
18294 @kindex set sysroot
18295 @item set sysroot @var{path}
18296 Use @var{path} as the system root for the program being debugged. Any
18297 absolute shared library paths will be prefixed with @var{path}; many
18298 runtime loaders store the absolute paths to the shared library in the
18299 target program's memory. When starting processes remotely, and when
18300 attaching to already-running processes (local or remote), their
18301 executable filenames will be prefixed with @var{path} if reported to
18302 @value{GDBN} as absolute by the operating system. If you use
18303 @code{set sysroot} to find executables and shared libraries, they need
18304 to be laid out in the same way that they are on the target, with
18305 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
18308 If @var{path} starts with the sequence @file{target:} and the target
18309 system is remote then @value{GDBN} will retrieve the target binaries
18310 from the remote system. This is only supported when using a remote
18311 target that supports the @code{remote get} command (@pxref{File
18312 Transfer,,Sending files to a remote system}). The part of @var{path}
18313 following the initial @file{target:} (if present) is used as system
18314 root prefix on the remote file system. If @var{path} starts with the
18315 sequence @file{remote:} this is converted to the sequence
18316 @file{target:} by @code{set sysroot}@footnote{Historically the
18317 functionality to retrieve binaries from the remote system was
18318 provided by prefixing @var{path} with @file{remote:}}. If you want
18319 to specify a local system root using a directory that happens to be
18320 named @file{target:} or @file{remote:}, you need to use some
18321 equivalent variant of the name like @file{./target:}.
18323 For targets with an MS-DOS based filesystem, such as MS-Windows and
18324 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
18325 absolute file name with @var{path}. But first, on Unix hosts,
18326 @value{GDBN} converts all backslash directory separators into forward
18327 slashes, because the backslash is not a directory separator on Unix:
18330 c:\foo\bar.dll @result{} c:/foo/bar.dll
18333 Then, @value{GDBN} attempts prefixing the target file name with
18334 @var{path}, and looks for the resulting file name in the host file
18338 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
18341 If that does not find the binary, @value{GDBN} tries removing
18342 the @samp{:} character from the drive spec, both for convenience, and,
18343 for the case of the host file system not supporting file names with
18347 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
18350 This makes it possible to have a system root that mirrors a target
18351 with more than one drive. E.g., you may want to setup your local
18352 copies of the target system shared libraries like so (note @samp{c} vs
18356 @file{/path/to/sysroot/c/sys/bin/foo.dll}
18357 @file{/path/to/sysroot/c/sys/bin/bar.dll}
18358 @file{/path/to/sysroot/z/sys/bin/bar.dll}
18362 and point the system root at @file{/path/to/sysroot}, so that
18363 @value{GDBN} can find the correct copies of both
18364 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
18366 If that still does not find the binary, @value{GDBN} tries
18367 removing the whole drive spec from the target file name:
18370 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
18373 This last lookup makes it possible to not care about the drive name,
18374 if you don't want or need to.
18376 The @code{set solib-absolute-prefix} command is an alias for @code{set
18379 @cindex default system root
18380 @cindex @samp{--with-sysroot}
18381 You can set the default system root by using the configure-time
18382 @samp{--with-sysroot} option. If the system root is inside
18383 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
18384 @samp{--exec-prefix}), then the default system root will be updated
18385 automatically if the installed @value{GDBN} is moved to a new
18388 @kindex show sysroot
18390 Display the current executable and shared library prefix.
18392 @kindex set solib-search-path
18393 @item set solib-search-path @var{path}
18394 If this variable is set, @var{path} is a colon-separated list of
18395 directories to search for shared libraries. @samp{solib-search-path}
18396 is used after @samp{sysroot} fails to locate the library, or if the
18397 path to the library is relative instead of absolute. If you want to
18398 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
18399 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
18400 finding your host's libraries. @samp{sysroot} is preferred; setting
18401 it to a nonexistent directory may interfere with automatic loading
18402 of shared library symbols.
18404 @kindex show solib-search-path
18405 @item show solib-search-path
18406 Display the current shared library search path.
18408 @cindex DOS file-name semantics of file names.
18409 @kindex set target-file-system-kind (unix|dos-based|auto)
18410 @kindex show target-file-system-kind
18411 @item set target-file-system-kind @var{kind}
18412 Set assumed file system kind for target reported file names.
18414 Shared library file names as reported by the target system may not
18415 make sense as is on the system @value{GDBN} is running on. For
18416 example, when remote debugging a target that has MS-DOS based file
18417 system semantics, from a Unix host, the target may be reporting to
18418 @value{GDBN} a list of loaded shared libraries with file names such as
18419 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
18420 drive letters, so the @samp{c:\} prefix is not normally understood as
18421 indicating an absolute file name, and neither is the backslash
18422 normally considered a directory separator character. In that case,
18423 the native file system would interpret this whole absolute file name
18424 as a relative file name with no directory components. This would make
18425 it impossible to point @value{GDBN} at a copy of the remote target's
18426 shared libraries on the host using @code{set sysroot}, and impractical
18427 with @code{set solib-search-path}. Setting
18428 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
18429 to interpret such file names similarly to how the target would, and to
18430 map them to file names valid on @value{GDBN}'s native file system
18431 semantics. The value of @var{kind} can be @code{"auto"}, in addition
18432 to one of the supported file system kinds. In that case, @value{GDBN}
18433 tries to determine the appropriate file system variant based on the
18434 current target's operating system (@pxref{ABI, ,Configuring the
18435 Current ABI}). The supported file system settings are:
18439 Instruct @value{GDBN} to assume the target file system is of Unix
18440 kind. Only file names starting the forward slash (@samp{/}) character
18441 are considered absolute, and the directory separator character is also
18445 Instruct @value{GDBN} to assume the target file system is DOS based.
18446 File names starting with either a forward slash, or a drive letter
18447 followed by a colon (e.g., @samp{c:}), are considered absolute, and
18448 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
18449 considered directory separators.
18452 Instruct @value{GDBN} to use the file system kind associated with the
18453 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
18454 This is the default.
18458 @cindex file name canonicalization
18459 @cindex base name differences
18460 When processing file names provided by the user, @value{GDBN}
18461 frequently needs to compare them to the file names recorded in the
18462 program's debug info. Normally, @value{GDBN} compares just the
18463 @dfn{base names} of the files as strings, which is reasonably fast
18464 even for very large programs. (The base name of a file is the last
18465 portion of its name, after stripping all the leading directories.)
18466 This shortcut in comparison is based upon the assumption that files
18467 cannot have more than one base name. This is usually true, but
18468 references to files that use symlinks or similar filesystem
18469 facilities violate that assumption. If your program records files
18470 using such facilities, or if you provide file names to @value{GDBN}
18471 using symlinks etc., you can set @code{basenames-may-differ} to
18472 @code{true} to instruct @value{GDBN} to completely canonicalize each
18473 pair of file names it needs to compare. This will make file-name
18474 comparisons accurate, but at a price of a significant slowdown.
18477 @item set basenames-may-differ
18478 @kindex set basenames-may-differ
18479 Set whether a source file may have multiple base names.
18481 @item show basenames-may-differ
18482 @kindex show basenames-may-differ
18483 Show whether a source file may have multiple base names.
18487 @section File Caching
18488 @cindex caching of opened files
18489 @cindex caching of bfd objects
18491 To speed up file loading, and reduce memory usage, @value{GDBN} will
18492 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
18493 BFD, bfd, The Binary File Descriptor Library}. The following commands
18494 allow visibility and control of the caching behavior.
18497 @kindex maint info bfds
18498 @item maint info bfds
18499 This prints information about each @code{bfd} object that is known to
18502 @kindex maint set bfd-sharing
18503 @kindex maint show bfd-sharing
18504 @kindex bfd caching
18505 @item maint set bfd-sharing
18506 @item maint show bfd-sharing
18507 Control whether @code{bfd} objects can be shared. When sharing is
18508 enabled @value{GDBN} reuses already open @code{bfd} objects rather
18509 than reopening the same file. Turning sharing off does not cause
18510 already shared @code{bfd} objects to be unshared, but all future files
18511 that are opened will create a new @code{bfd} object. Similarly,
18512 re-enabling sharing does not cause multiple existing @code{bfd}
18513 objects to be collapsed into a single shared @code{bfd} object.
18515 @kindex set debug bfd-cache @var{level}
18516 @kindex bfd caching
18517 @item set debug bfd-cache @var{level}
18518 Turns on debugging of the bfd cache, setting the level to @var{level}.
18520 @kindex show debug bfd-cache
18521 @kindex bfd caching
18522 @item show debug bfd-cache
18523 Show the current debugging level of the bfd cache.
18526 @node Separate Debug Files
18527 @section Debugging Information in Separate Files
18528 @cindex separate debugging information files
18529 @cindex debugging information in separate files
18530 @cindex @file{.debug} subdirectories
18531 @cindex debugging information directory, global
18532 @cindex global debugging information directories
18533 @cindex build ID, and separate debugging files
18534 @cindex @file{.build-id} directory
18536 @value{GDBN} allows you to put a program's debugging information in a
18537 file separate from the executable itself, in a way that allows
18538 @value{GDBN} to find and load the debugging information automatically.
18539 Since debugging information can be very large---sometimes larger
18540 than the executable code itself---some systems distribute debugging
18541 information for their executables in separate files, which users can
18542 install only when they need to debug a problem.
18544 @value{GDBN} supports two ways of specifying the separate debug info
18549 The executable contains a @dfn{debug link} that specifies the name of
18550 the separate debug info file. The separate debug file's name is
18551 usually @file{@var{executable}.debug}, where @var{executable} is the
18552 name of the corresponding executable file without leading directories
18553 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
18554 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
18555 checksum for the debug file, which @value{GDBN} uses to validate that
18556 the executable and the debug file came from the same build.
18559 The executable contains a @dfn{build ID}, a unique bit string that is
18560 also present in the corresponding debug info file. (This is supported
18561 only on some operating systems, when using the ELF or PE file formats
18562 for binary files and the @sc{gnu} Binutils.) For more details about
18563 this feature, see the description of the @option{--build-id}
18564 command-line option in @ref{Options, , Command Line Options, ld.info,
18565 The GNU Linker}. The debug info file's name is not specified
18566 explicitly by the build ID, but can be computed from the build ID, see
18570 Depending on the way the debug info file is specified, @value{GDBN}
18571 uses two different methods of looking for the debug file:
18575 For the ``debug link'' method, @value{GDBN} looks up the named file in
18576 the directory of the executable file, then in a subdirectory of that
18577 directory named @file{.debug}, and finally under each one of the global debug
18578 directories, in a subdirectory whose name is identical to the leading
18579 directories of the executable's absolute file name.
18582 For the ``build ID'' method, @value{GDBN} looks in the
18583 @file{.build-id} subdirectory of each one of the global debug directories for
18584 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
18585 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
18586 are the rest of the bit string. (Real build ID strings are 32 or more
18587 hex characters, not 10.)
18590 So, for example, suppose you ask @value{GDBN} to debug
18591 @file{/usr/bin/ls}, which has a debug link that specifies the
18592 file @file{ls.debug}, and a build ID whose value in hex is
18593 @code{abcdef1234}. If the list of the global debug directories includes
18594 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
18595 debug information files, in the indicated order:
18599 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
18601 @file{/usr/bin/ls.debug}
18603 @file{/usr/bin/.debug/ls.debug}
18605 @file{/usr/lib/debug/usr/bin/ls.debug}.
18608 @anchor{debug-file-directory}
18609 Global debugging info directories default to what is set by @value{GDBN}
18610 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
18611 you can also set the global debugging info directories, and view the list
18612 @value{GDBN} is currently using.
18616 @kindex set debug-file-directory
18617 @item set debug-file-directory @var{directories}
18618 Set the directories which @value{GDBN} searches for separate debugging
18619 information files to @var{directory}. Multiple path components can be set
18620 concatenating them by a path separator.
18622 @kindex show debug-file-directory
18623 @item show debug-file-directory
18624 Show the directories @value{GDBN} searches for separate debugging
18629 @cindex @code{.gnu_debuglink} sections
18630 @cindex debug link sections
18631 A debug link is a special section of the executable file named
18632 @code{.gnu_debuglink}. The section must contain:
18636 A filename, with any leading directory components removed, followed by
18639 zero to three bytes of padding, as needed to reach the next four-byte
18640 boundary within the section, and
18642 a four-byte CRC checksum, stored in the same endianness used for the
18643 executable file itself. The checksum is computed on the debugging
18644 information file's full contents by the function given below, passing
18645 zero as the @var{crc} argument.
18648 Any executable file format can carry a debug link, as long as it can
18649 contain a section named @code{.gnu_debuglink} with the contents
18652 @cindex @code{.note.gnu.build-id} sections
18653 @cindex build ID sections
18654 The build ID is a special section in the executable file (and in other
18655 ELF binary files that @value{GDBN} may consider). This section is
18656 often named @code{.note.gnu.build-id}, but that name is not mandatory.
18657 It contains unique identification for the built files---the ID remains
18658 the same across multiple builds of the same build tree. The default
18659 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
18660 content for the build ID string. The same section with an identical
18661 value is present in the original built binary with symbols, in its
18662 stripped variant, and in the separate debugging information file.
18664 The debugging information file itself should be an ordinary
18665 executable, containing a full set of linker symbols, sections, and
18666 debugging information. The sections of the debugging information file
18667 should have the same names, addresses, and sizes as the original file,
18668 but they need not contain any data---much like a @code{.bss} section
18669 in an ordinary executable.
18671 The @sc{gnu} binary utilities (Binutils) package includes the
18672 @samp{objcopy} utility that can produce
18673 the separated executable / debugging information file pairs using the
18674 following commands:
18677 @kbd{objcopy --only-keep-debug foo foo.debug}
18682 These commands remove the debugging
18683 information from the executable file @file{foo} and place it in the file
18684 @file{foo.debug}. You can use the first, second or both methods to link the
18689 The debug link method needs the following additional command to also leave
18690 behind a debug link in @file{foo}:
18693 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
18696 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
18697 a version of the @code{strip} command such that the command @kbd{strip foo -f
18698 foo.debug} has the same functionality as the two @code{objcopy} commands and
18699 the @code{ln -s} command above, together.
18702 Build ID gets embedded into the main executable using @code{ld --build-id} or
18703 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
18704 compatibility fixes for debug files separation are present in @sc{gnu} binary
18705 utilities (Binutils) package since version 2.18.
18710 @cindex CRC algorithm definition
18711 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
18712 IEEE 802.3 using the polynomial:
18714 @c TexInfo requires naked braces for multi-digit exponents for Tex
18715 @c output, but this causes HTML output to barf. HTML has to be set using
18716 @c raw commands. So we end up having to specify this equation in 2
18721 <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>
18722 + <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
18728 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
18729 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
18733 The function is computed byte at a time, taking the least
18734 significant bit of each byte first. The initial pattern
18735 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
18736 the final result is inverted to ensure trailing zeros also affect the
18739 @emph{Note:} This is the same CRC polynomial as used in handling the
18740 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
18741 However in the case of the Remote Serial Protocol, the CRC is computed
18742 @emph{most} significant bit first, and the result is not inverted, so
18743 trailing zeros have no effect on the CRC value.
18745 To complete the description, we show below the code of the function
18746 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
18747 initially supplied @code{crc} argument means that an initial call to
18748 this function passing in zero will start computing the CRC using
18751 @kindex gnu_debuglink_crc32
18754 gnu_debuglink_crc32 (unsigned long crc,
18755 unsigned char *buf, size_t len)
18757 static const unsigned long crc32_table[256] =
18759 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
18760 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
18761 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
18762 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
18763 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
18764 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
18765 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
18766 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
18767 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
18768 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
18769 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
18770 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
18771 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
18772 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
18773 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
18774 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
18775 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
18776 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
18777 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
18778 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
18779 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
18780 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
18781 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
18782 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
18783 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
18784 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
18785 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
18786 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
18787 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
18788 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
18789 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
18790 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
18791 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
18792 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
18793 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
18794 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
18795 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
18796 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
18797 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
18798 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
18799 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
18800 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
18801 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
18802 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
18803 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
18804 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
18805 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
18806 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
18807 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
18808 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
18809 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
18812 unsigned char *end;
18814 crc = ~crc & 0xffffffff;
18815 for (end = buf + len; buf < end; ++buf)
18816 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
18817 return ~crc & 0xffffffff;
18822 This computation does not apply to the ``build ID'' method.
18824 @node MiniDebugInfo
18825 @section Debugging information in a special section
18826 @cindex separate debug sections
18827 @cindex @samp{.gnu_debugdata} section
18829 Some systems ship pre-built executables and libraries that have a
18830 special @samp{.gnu_debugdata} section. This feature is called
18831 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
18832 is used to supply extra symbols for backtraces.
18834 The intent of this section is to provide extra minimal debugging
18835 information for use in simple backtraces. It is not intended to be a
18836 replacement for full separate debugging information (@pxref{Separate
18837 Debug Files}). The example below shows the intended use; however,
18838 @value{GDBN} does not currently put restrictions on what sort of
18839 debugging information might be included in the section.
18841 @value{GDBN} has support for this extension. If the section exists,
18842 then it is used provided that no other source of debugging information
18843 can be found, and that @value{GDBN} was configured with LZMA support.
18845 This section can be easily created using @command{objcopy} and other
18846 standard utilities:
18849 # Extract the dynamic symbols from the main binary, there is no need
18850 # to also have these in the normal symbol table.
18851 nm -D @var{binary} --format=posix --defined-only \
18852 | awk '@{ print $1 @}' | sort > dynsyms
18854 # Extract all the text (i.e. function) symbols from the debuginfo.
18855 # (Note that we actually also accept "D" symbols, for the benefit
18856 # of platforms like PowerPC64 that use function descriptors.)
18857 nm @var{binary} --format=posix --defined-only \
18858 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
18861 # Keep all the function symbols not already in the dynamic symbol
18863 comm -13 dynsyms funcsyms > keep_symbols
18865 # Separate full debug info into debug binary.
18866 objcopy --only-keep-debug @var{binary} debug
18868 # Copy the full debuginfo, keeping only a minimal set of symbols and
18869 # removing some unnecessary sections.
18870 objcopy -S --remove-section .gdb_index --remove-section .comment \
18871 --keep-symbols=keep_symbols debug mini_debuginfo
18873 # Drop the full debug info from the original binary.
18874 strip --strip-all -R .comment @var{binary}
18876 # Inject the compressed data into the .gnu_debugdata section of the
18879 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
18883 @section Index Files Speed Up @value{GDBN}
18884 @cindex index files
18885 @cindex @samp{.gdb_index} section
18887 When @value{GDBN} finds a symbol file, it scans the symbols in the
18888 file in order to construct an internal symbol table. This lets most
18889 @value{GDBN} operations work quickly---at the cost of a delay early
18890 on. For large programs, this delay can be quite lengthy, so
18891 @value{GDBN} provides a way to build an index, which speeds up
18894 The index is stored as a section in the symbol file. @value{GDBN} can
18895 write the index to a file, then you can put it into the symbol file
18896 using @command{objcopy}.
18898 To create an index file, use the @code{save gdb-index} command:
18901 @item save gdb-index @var{directory}
18902 @kindex save gdb-index
18903 Create an index file for each symbol file currently known by
18904 @value{GDBN}. Each file is named after its corresponding symbol file,
18905 with @samp{.gdb-index} appended, and is written into the given
18909 Once you have created an index file you can merge it into your symbol
18910 file, here named @file{symfile}, using @command{objcopy}:
18913 $ objcopy --add-section .gdb_index=symfile.gdb-index \
18914 --set-section-flags .gdb_index=readonly symfile symfile
18917 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
18918 sections that have been deprecated. Usually they are deprecated because
18919 they are missing a new feature or have performance issues.
18920 To tell @value{GDBN} to use a deprecated index section anyway
18921 specify @code{set use-deprecated-index-sections on}.
18922 The default is @code{off}.
18923 This can speed up startup, but may result in some functionality being lost.
18924 @xref{Index Section Format}.
18926 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
18927 must be done before gdb reads the file. The following will not work:
18930 $ gdb -ex "set use-deprecated-index-sections on" <program>
18933 Instead you must do, for example,
18936 $ gdb -iex "set use-deprecated-index-sections on" <program>
18939 There are currently some limitation on indices. They only work when
18940 for DWARF debugging information, not stabs. And, they do not
18941 currently work for programs using Ada.
18943 @node Symbol Errors
18944 @section Errors Reading Symbol Files
18946 While reading a symbol file, @value{GDBN} occasionally encounters problems,
18947 such as symbol types it does not recognize, or known bugs in compiler
18948 output. By default, @value{GDBN} does not notify you of such problems, since
18949 they are relatively common and primarily of interest to people
18950 debugging compilers. If you are interested in seeing information
18951 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
18952 only one message about each such type of problem, no matter how many
18953 times the problem occurs; or you can ask @value{GDBN} to print more messages,
18954 to see how many times the problems occur, with the @code{set
18955 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
18958 The messages currently printed, and their meanings, include:
18961 @item inner block not inside outer block in @var{symbol}
18963 The symbol information shows where symbol scopes begin and end
18964 (such as at the start of a function or a block of statements). This
18965 error indicates that an inner scope block is not fully contained
18966 in its outer scope blocks.
18968 @value{GDBN} circumvents the problem by treating the inner block as if it had
18969 the same scope as the outer block. In the error message, @var{symbol}
18970 may be shown as ``@code{(don't know)}'' if the outer block is not a
18973 @item block at @var{address} out of order
18975 The symbol information for symbol scope blocks should occur in
18976 order of increasing addresses. This error indicates that it does not
18979 @value{GDBN} does not circumvent this problem, and has trouble
18980 locating symbols in the source file whose symbols it is reading. (You
18981 can often determine what source file is affected by specifying
18982 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
18985 @item bad block start address patched
18987 The symbol information for a symbol scope block has a start address
18988 smaller than the address of the preceding source line. This is known
18989 to occur in the SunOS 4.1.1 (and earlier) C compiler.
18991 @value{GDBN} circumvents the problem by treating the symbol scope block as
18992 starting on the previous source line.
18994 @item bad string table offset in symbol @var{n}
18997 Symbol number @var{n} contains a pointer into the string table which is
18998 larger than the size of the string table.
19000 @value{GDBN} circumvents the problem by considering the symbol to have the
19001 name @code{foo}, which may cause other problems if many symbols end up
19004 @item unknown symbol type @code{0x@var{nn}}
19006 The symbol information contains new data types that @value{GDBN} does
19007 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
19008 uncomprehended information, in hexadecimal.
19010 @value{GDBN} circumvents the error by ignoring this symbol information.
19011 This usually allows you to debug your program, though certain symbols
19012 are not accessible. If you encounter such a problem and feel like
19013 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
19014 on @code{complain}, then go up to the function @code{read_dbx_symtab}
19015 and examine @code{*bufp} to see the symbol.
19017 @item stub type has NULL name
19019 @value{GDBN} could not find the full definition for a struct or class.
19021 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
19022 The symbol information for a C@t{++} member function is missing some
19023 information that recent versions of the compiler should have output for
19026 @item info mismatch between compiler and debugger
19028 @value{GDBN} could not parse a type specification output by the compiler.
19033 @section GDB Data Files
19035 @cindex prefix for data files
19036 @value{GDBN} will sometimes read an auxiliary data file. These files
19037 are kept in a directory known as the @dfn{data directory}.
19039 You can set the data directory's name, and view the name @value{GDBN}
19040 is currently using.
19043 @kindex set data-directory
19044 @item set data-directory @var{directory}
19045 Set the directory which @value{GDBN} searches for auxiliary data files
19046 to @var{directory}.
19048 @kindex show data-directory
19049 @item show data-directory
19050 Show the directory @value{GDBN} searches for auxiliary data files.
19053 @cindex default data directory
19054 @cindex @samp{--with-gdb-datadir}
19055 You can set the default data directory by using the configure-time
19056 @samp{--with-gdb-datadir} option. If the data directory is inside
19057 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
19058 @samp{--exec-prefix}), then the default data directory will be updated
19059 automatically if the installed @value{GDBN} is moved to a new
19062 The data directory may also be specified with the
19063 @code{--data-directory} command line option.
19064 @xref{Mode Options}.
19067 @chapter Specifying a Debugging Target
19069 @cindex debugging target
19070 A @dfn{target} is the execution environment occupied by your program.
19072 Often, @value{GDBN} runs in the same host environment as your program;
19073 in that case, the debugging target is specified as a side effect when
19074 you use the @code{file} or @code{core} commands. When you need more
19075 flexibility---for example, running @value{GDBN} on a physically separate
19076 host, or controlling a standalone system over a serial port or a
19077 realtime system over a TCP/IP connection---you can use the @code{target}
19078 command to specify one of the target types configured for @value{GDBN}
19079 (@pxref{Target Commands, ,Commands for Managing Targets}).
19081 @cindex target architecture
19082 It is possible to build @value{GDBN} for several different @dfn{target
19083 architectures}. When @value{GDBN} is built like that, you can choose
19084 one of the available architectures with the @kbd{set architecture}
19088 @kindex set architecture
19089 @kindex show architecture
19090 @item set architecture @var{arch}
19091 This command sets the current target architecture to @var{arch}. The
19092 value of @var{arch} can be @code{"auto"}, in addition to one of the
19093 supported architectures.
19095 @item show architecture
19096 Show the current target architecture.
19098 @item set processor
19100 @kindex set processor
19101 @kindex show processor
19102 These are alias commands for, respectively, @code{set architecture}
19103 and @code{show architecture}.
19107 * Active Targets:: Active targets
19108 * Target Commands:: Commands for managing targets
19109 * Byte Order:: Choosing target byte order
19112 @node Active Targets
19113 @section Active Targets
19115 @cindex stacking targets
19116 @cindex active targets
19117 @cindex multiple targets
19119 There are multiple classes of targets such as: processes, executable files or
19120 recording sessions. Core files belong to the process class, making core file
19121 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
19122 on multiple active targets, one in each class. This allows you to (for
19123 example) start a process and inspect its activity, while still having access to
19124 the executable file after the process finishes. Or if you start process
19125 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
19126 presented a virtual layer of the recording target, while the process target
19127 remains stopped at the chronologically last point of the process execution.
19129 Use the @code{core-file} and @code{exec-file} commands to select a new core
19130 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
19131 specify as a target a process that is already running, use the @code{attach}
19132 command (@pxref{Attach, ,Debugging an Already-running Process}).
19134 @node Target Commands
19135 @section Commands for Managing Targets
19138 @item target @var{type} @var{parameters}
19139 Connects the @value{GDBN} host environment to a target machine or
19140 process. A target is typically a protocol for talking to debugging
19141 facilities. You use the argument @var{type} to specify the type or
19142 protocol of the target machine.
19144 Further @var{parameters} are interpreted by the target protocol, but
19145 typically include things like device names or host names to connect
19146 with, process numbers, and baud rates.
19148 The @code{target} command does not repeat if you press @key{RET} again
19149 after executing the command.
19151 @kindex help target
19153 Displays the names of all targets available. To display targets
19154 currently selected, use either @code{info target} or @code{info files}
19155 (@pxref{Files, ,Commands to Specify Files}).
19157 @item help target @var{name}
19158 Describe a particular target, including any parameters necessary to
19161 @kindex set gnutarget
19162 @item set gnutarget @var{args}
19163 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
19164 knows whether it is reading an @dfn{executable},
19165 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
19166 with the @code{set gnutarget} command. Unlike most @code{target} commands,
19167 with @code{gnutarget} the @code{target} refers to a program, not a machine.
19170 @emph{Warning:} To specify a file format with @code{set gnutarget},
19171 you must know the actual BFD name.
19175 @xref{Files, , Commands to Specify Files}.
19177 @kindex show gnutarget
19178 @item show gnutarget
19179 Use the @code{show gnutarget} command to display what file format
19180 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
19181 @value{GDBN} will determine the file format for each file automatically,
19182 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
19185 @cindex common targets
19186 Here are some common targets (available, or not, depending on the GDB
19191 @item target exec @var{program}
19192 @cindex executable file target
19193 An executable file. @samp{target exec @var{program}} is the same as
19194 @samp{exec-file @var{program}}.
19196 @item target core @var{filename}
19197 @cindex core dump file target
19198 A core dump file. @samp{target core @var{filename}} is the same as
19199 @samp{core-file @var{filename}}.
19201 @item target remote @var{medium}
19202 @cindex remote target
19203 A remote system connected to @value{GDBN} via a serial line or network
19204 connection. This command tells @value{GDBN} to use its own remote
19205 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
19207 For example, if you have a board connected to @file{/dev/ttya} on the
19208 machine running @value{GDBN}, you could say:
19211 target remote /dev/ttya
19214 @code{target remote} supports the @code{load} command. This is only
19215 useful if you have some other way of getting the stub to the target
19216 system, and you can put it somewhere in memory where it won't get
19217 clobbered by the download.
19219 @item target sim @r{[}@var{simargs}@r{]} @dots{}
19220 @cindex built-in simulator target
19221 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
19229 works; however, you cannot assume that a specific memory map, device
19230 drivers, or even basic I/O is available, although some simulators do
19231 provide these. For info about any processor-specific simulator details,
19232 see the appropriate section in @ref{Embedded Processors, ,Embedded
19235 @item target native
19236 @cindex native target
19237 Setup for local/native process debugging. Useful to make the
19238 @code{run} command spawn native processes (likewise @code{attach},
19239 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
19240 (@pxref{set auto-connect-native-target}).
19244 Different targets are available on different configurations of @value{GDBN};
19245 your configuration may have more or fewer targets.
19247 Many remote targets require you to download the executable's code once
19248 you've successfully established a connection. You may wish to control
19249 various aspects of this process.
19254 @kindex set hash@r{, for remote monitors}
19255 @cindex hash mark while downloading
19256 This command controls whether a hash mark @samp{#} is displayed while
19257 downloading a file to the remote monitor. If on, a hash mark is
19258 displayed after each S-record is successfully downloaded to the
19262 @kindex show hash@r{, for remote monitors}
19263 Show the current status of displaying the hash mark.
19265 @item set debug monitor
19266 @kindex set debug monitor
19267 @cindex display remote monitor communications
19268 Enable or disable display of communications messages between
19269 @value{GDBN} and the remote monitor.
19271 @item show debug monitor
19272 @kindex show debug monitor
19273 Show the current status of displaying communications between
19274 @value{GDBN} and the remote monitor.
19279 @kindex load @var{filename}
19280 @item load @var{filename}
19282 Depending on what remote debugging facilities are configured into
19283 @value{GDBN}, the @code{load} command may be available. Where it exists, it
19284 is meant to make @var{filename} (an executable) available for debugging
19285 on the remote system---by downloading, or dynamic linking, for example.
19286 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
19287 the @code{add-symbol-file} command.
19289 If your @value{GDBN} does not have a @code{load} command, attempting to
19290 execute it gets the error message ``@code{You can't do that when your
19291 target is @dots{}}''
19293 The file is loaded at whatever address is specified in the executable.
19294 For some object file formats, you can specify the load address when you
19295 link the program; for other formats, like a.out, the object file format
19296 specifies a fixed address.
19297 @c FIXME! This would be a good place for an xref to the GNU linker doc.
19299 Depending on the remote side capabilities, @value{GDBN} may be able to
19300 load programs into flash memory.
19302 @code{load} does not repeat if you press @key{RET} again after using it.
19306 @section Choosing Target Byte Order
19308 @cindex choosing target byte order
19309 @cindex target byte order
19311 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
19312 offer the ability to run either big-endian or little-endian byte
19313 orders. Usually the executable or symbol will include a bit to
19314 designate the endian-ness, and you will not need to worry about
19315 which to use. However, you may still find it useful to adjust
19316 @value{GDBN}'s idea of processor endian-ness manually.
19320 @item set endian big
19321 Instruct @value{GDBN} to assume the target is big-endian.
19323 @item set endian little
19324 Instruct @value{GDBN} to assume the target is little-endian.
19326 @item set endian auto
19327 Instruct @value{GDBN} to use the byte order associated with the
19331 Display @value{GDBN}'s current idea of the target byte order.
19335 Note that these commands merely adjust interpretation of symbolic
19336 data on the host, and that they have absolutely no effect on the
19340 @node Remote Debugging
19341 @chapter Debugging Remote Programs
19342 @cindex remote debugging
19344 If you are trying to debug a program running on a machine that cannot run
19345 @value{GDBN} in the usual way, it is often useful to use remote debugging.
19346 For example, you might use remote debugging on an operating system kernel,
19347 or on a small system which does not have a general purpose operating system
19348 powerful enough to run a full-featured debugger.
19350 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
19351 to make this work with particular debugging targets. In addition,
19352 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
19353 but not specific to any particular target system) which you can use if you
19354 write the remote stubs---the code that runs on the remote system to
19355 communicate with @value{GDBN}.
19357 Other remote targets may be available in your
19358 configuration of @value{GDBN}; use @code{help target} to list them.
19361 * Connecting:: Connecting to a remote target
19362 * File Transfer:: Sending files to a remote system
19363 * Server:: Using the gdbserver program
19364 * Remote Configuration:: Remote configuration
19365 * Remote Stub:: Implementing a remote stub
19369 @section Connecting to a Remote Target
19370 @cindex remote debugging, connecting
19371 @cindex @code{gdbserver}, connecting
19372 @cindex remote debugging, types of connections
19373 @cindex @code{gdbserver}, types of connections
19374 @cindex @code{gdbserver}, @code{target remote} mode
19375 @cindex @code{gdbserver}, @code{target extended-remote} mode
19377 This section describes how to connect to a remote target, including the
19378 types of connections and their differences, how to set up executable and
19379 symbol files on the host and target, and the commands used for
19380 connecting to and disconnecting from the remote target.
19382 @subsection Types of Remote Connections
19384 @value{GDBN} supports two types of remote connections, @code{target remote}
19385 mode and @code{target extended-remote} mode. Note that many remote targets
19386 support only @code{target remote} mode. There are several major
19387 differences between the two types of connections, enumerated here:
19391 @cindex remote debugging, detach and program exit
19392 @item Result of detach or program exit
19393 @strong{With target remote mode:} When the debugged program exits or you
19394 detach from it, @value{GDBN} disconnects from the target. When using
19395 @code{gdbserver}, @code{gdbserver} will exit.
19397 @strong{With target extended-remote mode:} When the debugged program exits or
19398 you detach from it, @value{GDBN} remains connected to the target, even
19399 though no program is running. You can rerun the program, attach to a
19400 running program, or use @code{monitor} commands specific to the target.
19402 When using @code{gdbserver} in this case, it does not exit unless it was
19403 invoked using the @option{--once} option. If the @option{--once} option
19404 was not used, you can ask @code{gdbserver} to exit using the
19405 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
19407 @item Specifying the program to debug
19408 For both connection types you use the @code{file} command to specify the
19409 program on the host system. If you are using @code{gdbserver} there are
19410 some differences in how to specify the location of the program on the
19413 @strong{With target remote mode:} You must either specify the program to debug
19414 on the @code{gdbserver} command line or use the @option{--attach} option
19415 (@pxref{Attaching to a program,,Attaching to a Running Program}).
19417 @cindex @option{--multi}, @code{gdbserver} option
19418 @strong{With target extended-remote mode:} You may specify the program to debug
19419 on the @code{gdbserver} command line, or you can load the program or attach
19420 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
19422 @anchor{--multi Option in Types of Remote Connnections}
19423 You can start @code{gdbserver} without supplying an initial command to run
19424 or process ID to attach. To do this, use the @option{--multi} command line
19425 option. Then you can connect using @code{target extended-remote} and start
19426 the program you want to debug (see below for details on using the
19427 @code{run} command in this scenario). Note that the conditions under which
19428 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
19429 (@code{target remote} or @code{target extended-remote}). The
19430 @option{--multi} option to @code{gdbserver} has no influence on that.
19432 @item The @code{run} command
19433 @strong{With target remote mode:} The @code{run} command is not
19434 supported. Once a connection has been established, you can use all
19435 the usual @value{GDBN} commands to examine and change data. The
19436 remote program is already running, so you can use commands like
19437 @kbd{step} and @kbd{continue}.
19439 @strong{With target extended-remote mode:} The @code{run} command is
19440 supported. The @code{run} command uses the value set by
19441 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
19442 the program to run. Command line arguments are supported, except for
19443 wildcard expansion and I/O redirection (@pxref{Arguments}).
19445 If you specify the program to debug on the command line, then the
19446 @code{run} command is not required to start execution, and you can
19447 resume using commands like @kbd{step} and @kbd{continue} as with
19448 @code{target remote} mode.
19450 @anchor{Attaching in Types of Remote Connections}
19452 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
19453 not supported. To attach to a running program using @code{gdbserver}, you
19454 must use the @option{--attach} option (@pxref{Running gdbserver}).
19456 @strong{With target extended-remote mode:} To attach to a running program,
19457 you may use the @code{attach} command after the connection has been
19458 established. If you are using @code{gdbserver}, you may also invoke
19459 @code{gdbserver} using the @option{--attach} option
19460 (@pxref{Running gdbserver}).
19464 @anchor{Host and target files}
19465 @subsection Host and Target Files
19466 @cindex remote debugging, symbol files
19467 @cindex symbol files, remote debugging
19469 @value{GDBN}, running on the host, needs access to symbol and debugging
19470 information for your program running on the target. This requires
19471 access to an unstripped copy of your program, and possibly any associated
19472 symbol files. Note that this section applies equally to both @code{target
19473 remote} mode and @code{target extended-remote} mode.
19475 Some remote targets (@pxref{qXfer executable filename read}, and
19476 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
19477 the same connection used to communicate with @value{GDBN}. With such a
19478 target, if the remote program is unstripped, the only command you need is
19479 @code{target remote} (or @code{target extended-remote}).
19481 If the remote program is stripped, or the target does not support remote
19482 program file access, start up @value{GDBN} using the name of the local
19483 unstripped copy of your program as the first argument, or use the
19484 @code{file} command. Use @code{set sysroot} to specify the location (on
19485 the host) of target libraries (unless your @value{GDBN} was compiled with
19486 the correct sysroot using @code{--with-sysroot}). Alternatively, you
19487 may use @code{set solib-search-path} to specify how @value{GDBN} locates
19490 The symbol file and target libraries must exactly match the executable
19491 and libraries on the target, with one exception: the files on the host
19492 system should not be stripped, even if the files on the target system
19493 are. Mismatched or missing files will lead to confusing results
19494 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
19495 files may also prevent @code{gdbserver} from debugging multi-threaded
19498 @subsection Remote Connection Commands
19499 @cindex remote connection commands
19500 @value{GDBN} can communicate with the target over a serial line, or
19501 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
19502 each case, @value{GDBN} uses the same protocol for debugging your
19503 program; only the medium carrying the debugging packets varies. The
19504 @code{target remote} and @code{target extended-remote} commands
19505 establish a connection to the target. Both commands accept the same
19506 arguments, which indicate the medium to use:
19510 @item target remote @var{serial-device}
19511 @itemx target extended-remote @var{serial-device}
19512 @cindex serial line, @code{target remote}
19513 Use @var{serial-device} to communicate with the target. For example,
19514 to use a serial line connected to the device named @file{/dev/ttyb}:
19517 target remote /dev/ttyb
19520 If you're using a serial line, you may want to give @value{GDBN} the
19521 @samp{--baud} option, or use the @code{set serial baud} command
19522 (@pxref{Remote Configuration, set serial baud}) before the
19523 @code{target} command.
19525 @item target remote @code{@var{host}:@var{port}}
19526 @itemx target remote @code{tcp:@var{host}:@var{port}}
19527 @itemx target extended-remote @code{@var{host}:@var{port}}
19528 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
19529 @cindex @acronym{TCP} port, @code{target remote}
19530 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
19531 The @var{host} may be either a host name or a numeric @acronym{IP}
19532 address; @var{port} must be a decimal number. The @var{host} could be
19533 the target machine itself, if it is directly connected to the net, or
19534 it might be a terminal server which in turn has a serial line to the
19537 For example, to connect to port 2828 on a terminal server named
19541 target remote manyfarms:2828
19544 If your remote target is actually running on the same machine as your
19545 debugger session (e.g.@: a simulator for your target running on the
19546 same host), you can omit the hostname. For example, to connect to
19547 port 1234 on your local machine:
19550 target remote :1234
19554 Note that the colon is still required here.
19556 @item target remote @code{udp:@var{host}:@var{port}}
19557 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
19558 @cindex @acronym{UDP} port, @code{target remote}
19559 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
19560 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
19563 target remote udp:manyfarms:2828
19566 When using a @acronym{UDP} connection for remote debugging, you should
19567 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
19568 can silently drop packets on busy or unreliable networks, which will
19569 cause havoc with your debugging session.
19571 @item target remote | @var{command}
19572 @itemx target extended-remote | @var{command}
19573 @cindex pipe, @code{target remote} to
19574 Run @var{command} in the background and communicate with it using a
19575 pipe. The @var{command} is a shell command, to be parsed and expanded
19576 by the system's command shell, @code{/bin/sh}; it should expect remote
19577 protocol packets on its standard input, and send replies on its
19578 standard output. You could use this to run a stand-alone simulator
19579 that speaks the remote debugging protocol, to make net connections
19580 using programs like @code{ssh}, or for other similar tricks.
19582 If @var{command} closes its standard output (perhaps by exiting),
19583 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
19584 program has already exited, this will have no effect.)
19588 @cindex interrupting remote programs
19589 @cindex remote programs, interrupting
19590 Whenever @value{GDBN} is waiting for the remote program, if you type the
19591 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
19592 program. This may or may not succeed, depending in part on the hardware
19593 and the serial drivers the remote system uses. If you type the
19594 interrupt character once again, @value{GDBN} displays this prompt:
19597 Interrupted while waiting for the program.
19598 Give up (and stop debugging it)? (y or n)
19601 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
19602 the remote debugging session. (If you decide you want to try again later,
19603 you can use @kbd{target remote} again to connect once more.) If you type
19604 @kbd{n}, @value{GDBN} goes back to waiting.
19606 In @code{target extended-remote} mode, typing @kbd{n} will leave
19607 @value{GDBN} connected to the target.
19610 @kindex detach (remote)
19612 When you have finished debugging the remote program, you can use the
19613 @code{detach} command to release it from @value{GDBN} control.
19614 Detaching from the target normally resumes its execution, but the results
19615 will depend on your particular remote stub. After the @code{detach}
19616 command in @code{target remote} mode, @value{GDBN} is free to connect to
19617 another target. In @code{target extended-remote} mode, @value{GDBN} is
19618 still connected to the target.
19622 The @code{disconnect} command closes the connection to the target, and
19623 the target is generally not resumed. It will wait for @value{GDBN}
19624 (this instance or another one) to connect and continue debugging. After
19625 the @code{disconnect} command, @value{GDBN} is again free to connect to
19628 @cindex send command to remote monitor
19629 @cindex extend @value{GDBN} for remote targets
19630 @cindex add new commands for external monitor
19632 @item monitor @var{cmd}
19633 This command allows you to send arbitrary commands directly to the
19634 remote monitor. Since @value{GDBN} doesn't care about the commands it
19635 sends like this, this command is the way to extend @value{GDBN}---you
19636 can add new commands that only the external monitor will understand
19640 @node File Transfer
19641 @section Sending files to a remote system
19642 @cindex remote target, file transfer
19643 @cindex file transfer
19644 @cindex sending files to remote systems
19646 Some remote targets offer the ability to transfer files over the same
19647 connection used to communicate with @value{GDBN}. This is convenient
19648 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
19649 running @code{gdbserver} over a network interface. For other targets,
19650 e.g.@: embedded devices with only a single serial port, this may be
19651 the only way to upload or download files.
19653 Not all remote targets support these commands.
19657 @item remote put @var{hostfile} @var{targetfile}
19658 Copy file @var{hostfile} from the host system (the machine running
19659 @value{GDBN}) to @var{targetfile} on the target system.
19662 @item remote get @var{targetfile} @var{hostfile}
19663 Copy file @var{targetfile} from the target system to @var{hostfile}
19664 on the host system.
19666 @kindex remote delete
19667 @item remote delete @var{targetfile}
19668 Delete @var{targetfile} from the target system.
19673 @section Using the @code{gdbserver} Program
19676 @cindex remote connection without stubs
19677 @code{gdbserver} is a control program for Unix-like systems, which
19678 allows you to connect your program with a remote @value{GDBN} via
19679 @code{target remote} or @code{target extended-remote}---but without
19680 linking in the usual debugging stub.
19682 @code{gdbserver} is not a complete replacement for the debugging stubs,
19683 because it requires essentially the same operating-system facilities
19684 that @value{GDBN} itself does. In fact, a system that can run
19685 @code{gdbserver} to connect to a remote @value{GDBN} could also run
19686 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
19687 because it is a much smaller program than @value{GDBN} itself. It is
19688 also easier to port than all of @value{GDBN}, so you may be able to get
19689 started more quickly on a new system by using @code{gdbserver}.
19690 Finally, if you develop code for real-time systems, you may find that
19691 the tradeoffs involved in real-time operation make it more convenient to
19692 do as much development work as possible on another system, for example
19693 by cross-compiling. You can use @code{gdbserver} to make a similar
19694 choice for debugging.
19696 @value{GDBN} and @code{gdbserver} communicate via either a serial line
19697 or a TCP connection, using the standard @value{GDBN} remote serial
19701 @emph{Warning:} @code{gdbserver} does not have any built-in security.
19702 Do not run @code{gdbserver} connected to any public network; a
19703 @value{GDBN} connection to @code{gdbserver} provides access to the
19704 target system with the same privileges as the user running
19708 @anchor{Running gdbserver}
19709 @subsection Running @code{gdbserver}
19710 @cindex arguments, to @code{gdbserver}
19711 @cindex @code{gdbserver}, command-line arguments
19713 Run @code{gdbserver} on the target system. You need a copy of the
19714 program you want to debug, including any libraries it requires.
19715 @code{gdbserver} does not need your program's symbol table, so you can
19716 strip the program if necessary to save space. @value{GDBN} on the host
19717 system does all the symbol handling.
19719 To use the server, you must tell it how to communicate with @value{GDBN};
19720 the name of your program; and the arguments for your program. The usual
19724 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
19727 @var{comm} is either a device name (to use a serial line), or a TCP
19728 hostname and portnumber, or @code{-} or @code{stdio} to use
19729 stdin/stdout of @code{gdbserver}.
19730 For example, to debug Emacs with the argument
19731 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
19735 target> gdbserver /dev/com1 emacs foo.txt
19738 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
19741 To use a TCP connection instead of a serial line:
19744 target> gdbserver host:2345 emacs foo.txt
19747 The only difference from the previous example is the first argument,
19748 specifying that you are communicating with the host @value{GDBN} via
19749 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
19750 expect a TCP connection from machine @samp{host} to local TCP port 2345.
19751 (Currently, the @samp{host} part is ignored.) You can choose any number
19752 you want for the port number as long as it does not conflict with any
19753 TCP ports already in use on the target system (for example, @code{23} is
19754 reserved for @code{telnet}).@footnote{If you choose a port number that
19755 conflicts with another service, @code{gdbserver} prints an error message
19756 and exits.} You must use the same port number with the host @value{GDBN}
19757 @code{target remote} command.
19759 The @code{stdio} connection is useful when starting @code{gdbserver}
19763 (gdb) target remote | ssh -T hostname gdbserver - hello
19766 The @samp{-T} option to ssh is provided because we don't need a remote pty,
19767 and we don't want escape-character handling. Ssh does this by default when
19768 a command is provided, the flag is provided to make it explicit.
19769 You could elide it if you want to.
19771 Programs started with stdio-connected gdbserver have @file{/dev/null} for
19772 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
19773 display through a pipe connected to gdbserver.
19774 Both @code{stdout} and @code{stderr} use the same pipe.
19776 @anchor{Attaching to a program}
19777 @subsubsection Attaching to a Running Program
19778 @cindex attach to a program, @code{gdbserver}
19779 @cindex @option{--attach}, @code{gdbserver} option
19781 On some targets, @code{gdbserver} can also attach to running programs.
19782 This is accomplished via the @code{--attach} argument. The syntax is:
19785 target> gdbserver --attach @var{comm} @var{pid}
19788 @var{pid} is the process ID of a currently running process. It isn't
19789 necessary to point @code{gdbserver} at a binary for the running process.
19791 In @code{target extended-remote} mode, you can also attach using the
19792 @value{GDBN} attach command
19793 (@pxref{Attaching in Types of Remote Connections}).
19796 You can debug processes by name instead of process ID if your target has the
19797 @code{pidof} utility:
19800 target> gdbserver --attach @var{comm} `pidof @var{program}`
19803 In case more than one copy of @var{program} is running, or @var{program}
19804 has multiple threads, most versions of @code{pidof} support the
19805 @code{-s} option to only return the first process ID.
19807 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
19809 This section applies only when @code{gdbserver} is run to listen on a TCP
19812 @code{gdbserver} normally terminates after all of its debugged processes have
19813 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
19814 extended-remote}, @code{gdbserver} stays running even with no processes left.
19815 @value{GDBN} normally terminates the spawned debugged process on its exit,
19816 which normally also terminates @code{gdbserver} in the @kbd{target remote}
19817 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
19818 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
19819 stays running even in the @kbd{target remote} mode.
19821 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
19822 Such reconnecting is useful for features like @ref{disconnected tracing}. For
19823 completeness, at most one @value{GDBN} can be connected at a time.
19825 @cindex @option{--once}, @code{gdbserver} option
19826 By default, @code{gdbserver} keeps the listening TCP port open, so that
19827 subsequent connections are possible. However, if you start @code{gdbserver}
19828 with the @option{--once} option, it will stop listening for any further
19829 connection attempts after connecting to the first @value{GDBN} session. This
19830 means no further connections to @code{gdbserver} will be possible after the
19831 first one. It also means @code{gdbserver} will terminate after the first
19832 connection with remote @value{GDBN} has closed, even for unexpectedly closed
19833 connections and even in the @kbd{target extended-remote} mode. The
19834 @option{--once} option allows reusing the same port number for connecting to
19835 multiple instances of @code{gdbserver} running on the same host, since each
19836 instance closes its port after the first connection.
19838 @anchor{Other Command-Line Arguments for gdbserver}
19839 @subsubsection Other Command-Line Arguments for @code{gdbserver}
19841 You can use the @option{--multi} option to start @code{gdbserver} without
19842 specifying a program to debug or a process to attach to. Then you can
19843 attach in @code{target extended-remote} mode and run or attach to a
19844 program. For more information,
19845 @pxref{--multi Option in Types of Remote Connnections}.
19847 @cindex @option{--debug}, @code{gdbserver} option
19848 The @option{--debug} option tells @code{gdbserver} to display extra
19849 status information about the debugging process.
19850 @cindex @option{--remote-debug}, @code{gdbserver} option
19851 The @option{--remote-debug} option tells @code{gdbserver} to display
19852 remote protocol debug output. These options are intended for
19853 @code{gdbserver} development and for bug reports to the developers.
19855 @cindex @option{--debug-format}, @code{gdbserver} option
19856 The @option{--debug-format=option1[,option2,...]} option tells
19857 @code{gdbserver} to include additional information in each output.
19858 Possible options are:
19862 Turn off all extra information in debugging output.
19864 Turn on all extra information in debugging output.
19866 Include a timestamp in each line of debugging output.
19869 Options are processed in order. Thus, for example, if @option{none}
19870 appears last then no additional information is added to debugging output.
19872 @cindex @option{--wrapper}, @code{gdbserver} option
19873 The @option{--wrapper} option specifies a wrapper to launch programs
19874 for debugging. The option should be followed by the name of the
19875 wrapper, then any command-line arguments to pass to the wrapper, then
19876 @kbd{--} indicating the end of the wrapper arguments.
19878 @code{gdbserver} runs the specified wrapper program with a combined
19879 command line including the wrapper arguments, then the name of the
19880 program to debug, then any arguments to the program. The wrapper
19881 runs until it executes your program, and then @value{GDBN} gains control.
19883 You can use any program that eventually calls @code{execve} with
19884 its arguments as a wrapper. Several standard Unix utilities do
19885 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
19886 with @code{exec "$@@"} will also work.
19888 For example, you can use @code{env} to pass an environment variable to
19889 the debugged program, without setting the variable in @code{gdbserver}'s
19893 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
19896 @subsection Connecting to @code{gdbserver}
19898 The basic procedure for connecting to the remote target is:
19902 Run @value{GDBN} on the host system.
19905 Make sure you have the necessary symbol files
19906 (@pxref{Host and target files}).
19907 Load symbols for your application using the @code{file} command before you
19908 connect. Use @code{set sysroot} to locate target libraries (unless your
19909 @value{GDBN} was compiled with the correct sysroot using
19910 @code{--with-sysroot}).
19913 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
19914 For TCP connections, you must start up @code{gdbserver} prior to using
19915 the @code{target} command. Otherwise you may get an error whose
19916 text depends on the host system, but which usually looks something like
19917 @samp{Connection refused}. Don't use the @code{load}
19918 command in @value{GDBN} when using @code{target remote} mode, since the
19919 program is already on the target.
19923 @anchor{Monitor Commands for gdbserver}
19924 @subsection Monitor Commands for @code{gdbserver}
19925 @cindex monitor commands, for @code{gdbserver}
19927 During a @value{GDBN} session using @code{gdbserver}, you can use the
19928 @code{monitor} command to send special requests to @code{gdbserver}.
19929 Here are the available commands.
19933 List the available monitor commands.
19935 @item monitor set debug 0
19936 @itemx monitor set debug 1
19937 Disable or enable general debugging messages.
19939 @item monitor set remote-debug 0
19940 @itemx monitor set remote-debug 1
19941 Disable or enable specific debugging messages associated with the remote
19942 protocol (@pxref{Remote Protocol}).
19944 @item monitor set debug-format option1@r{[},option2,...@r{]}
19945 Specify additional text to add to debugging messages.
19946 Possible options are:
19950 Turn off all extra information in debugging output.
19952 Turn on all extra information in debugging output.
19954 Include a timestamp in each line of debugging output.
19957 Options are processed in order. Thus, for example, if @option{none}
19958 appears last then no additional information is added to debugging output.
19960 @item monitor set libthread-db-search-path [PATH]
19961 @cindex gdbserver, search path for @code{libthread_db}
19962 When this command is issued, @var{path} is a colon-separated list of
19963 directories to search for @code{libthread_db} (@pxref{Threads,,set
19964 libthread-db-search-path}). If you omit @var{path},
19965 @samp{libthread-db-search-path} will be reset to its default value.
19967 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
19968 not supported in @code{gdbserver}.
19971 Tell gdbserver to exit immediately. This command should be followed by
19972 @code{disconnect} to close the debugging session. @code{gdbserver} will
19973 detach from any attached processes and kill any processes it created.
19974 Use @code{monitor exit} to terminate @code{gdbserver} at the end
19975 of a multi-process mode debug session.
19979 @subsection Tracepoints support in @code{gdbserver}
19980 @cindex tracepoints support in @code{gdbserver}
19982 On some targets, @code{gdbserver} supports tracepoints, fast
19983 tracepoints and static tracepoints.
19985 For fast or static tracepoints to work, a special library called the
19986 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
19987 This library is built and distributed as an integral part of
19988 @code{gdbserver}. In addition, support for static tracepoints
19989 requires building the in-process agent library with static tracepoints
19990 support. At present, the UST (LTTng Userspace Tracer,
19991 @url{http://lttng.org/ust}) tracing engine is supported. This support
19992 is automatically available if UST development headers are found in the
19993 standard include path when @code{gdbserver} is built, or if
19994 @code{gdbserver} was explicitly configured using @option{--with-ust}
19995 to point at such headers. You can explicitly disable the support
19996 using @option{--with-ust=no}.
19998 There are several ways to load the in-process agent in your program:
20001 @item Specifying it as dependency at link time
20003 You can link your program dynamically with the in-process agent
20004 library. On most systems, this is accomplished by adding
20005 @code{-linproctrace} to the link command.
20007 @item Using the system's preloading mechanisms
20009 You can force loading the in-process agent at startup time by using
20010 your system's support for preloading shared libraries. Many Unixes
20011 support the concept of preloading user defined libraries. In most
20012 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
20013 in the environment. See also the description of @code{gdbserver}'s
20014 @option{--wrapper} command line option.
20016 @item Using @value{GDBN} to force loading the agent at run time
20018 On some systems, you can force the inferior to load a shared library,
20019 by calling a dynamic loader function in the inferior that takes care
20020 of dynamically looking up and loading a shared library. On most Unix
20021 systems, the function is @code{dlopen}. You'll use the @code{call}
20022 command for that. For example:
20025 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
20028 Note that on most Unix systems, for the @code{dlopen} function to be
20029 available, the program needs to be linked with @code{-ldl}.
20032 On systems that have a userspace dynamic loader, like most Unix
20033 systems, when you connect to @code{gdbserver} using @code{target
20034 remote}, you'll find that the program is stopped at the dynamic
20035 loader's entry point, and no shared library has been loaded in the
20036 program's address space yet, including the in-process agent. In that
20037 case, before being able to use any of the fast or static tracepoints
20038 features, you need to let the loader run and load the shared
20039 libraries. The simplest way to do that is to run the program to the
20040 main procedure. E.g., if debugging a C or C@t{++} program, start
20041 @code{gdbserver} like so:
20044 $ gdbserver :9999 myprogram
20047 Start GDB and connect to @code{gdbserver} like so, and run to main:
20051 (@value{GDBP}) target remote myhost:9999
20052 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
20053 (@value{GDBP}) b main
20054 (@value{GDBP}) continue
20057 The in-process tracing agent library should now be loaded into the
20058 process; you can confirm it with the @code{info sharedlibrary}
20059 command, which will list @file{libinproctrace.so} as loaded in the
20060 process. You are now ready to install fast tracepoints, list static
20061 tracepoint markers, probe static tracepoints markers, and start
20064 @node Remote Configuration
20065 @section Remote Configuration
20068 @kindex show remote
20069 This section documents the configuration options available when
20070 debugging remote programs. For the options related to the File I/O
20071 extensions of the remote protocol, see @ref{system,
20072 system-call-allowed}.
20075 @item set remoteaddresssize @var{bits}
20076 @cindex address size for remote targets
20077 @cindex bits in remote address
20078 Set the maximum size of address in a memory packet to the specified
20079 number of bits. @value{GDBN} will mask off the address bits above
20080 that number, when it passes addresses to the remote target. The
20081 default value is the number of bits in the target's address.
20083 @item show remoteaddresssize
20084 Show the current value of remote address size in bits.
20086 @item set serial baud @var{n}
20087 @cindex baud rate for remote targets
20088 Set the baud rate for the remote serial I/O to @var{n} baud. The
20089 value is used to set the speed of the serial port used for debugging
20092 @item show serial baud
20093 Show the current speed of the remote connection.
20095 @item set serial parity @var{parity}
20096 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
20097 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
20099 @item show serial parity
20100 Show the current parity of the serial port.
20102 @item set remotebreak
20103 @cindex interrupt remote programs
20104 @cindex BREAK signal instead of Ctrl-C
20105 @anchor{set remotebreak}
20106 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
20107 when you type @kbd{Ctrl-c} to interrupt the program running
20108 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
20109 character instead. The default is off, since most remote systems
20110 expect to see @samp{Ctrl-C} as the interrupt signal.
20112 @item show remotebreak
20113 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
20114 interrupt the remote program.
20116 @item set remoteflow on
20117 @itemx set remoteflow off
20118 @kindex set remoteflow
20119 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
20120 on the serial port used to communicate to the remote target.
20122 @item show remoteflow
20123 @kindex show remoteflow
20124 Show the current setting of hardware flow control.
20126 @item set remotelogbase @var{base}
20127 Set the base (a.k.a.@: radix) of logging serial protocol
20128 communications to @var{base}. Supported values of @var{base} are:
20129 @code{ascii}, @code{octal}, and @code{hex}. The default is
20132 @item show remotelogbase
20133 Show the current setting of the radix for logging remote serial
20136 @item set remotelogfile @var{file}
20137 @cindex record serial communications on file
20138 Record remote serial communications on the named @var{file}. The
20139 default is not to record at all.
20141 @item show remotelogfile.
20142 Show the current setting of the file name on which to record the
20143 serial communications.
20145 @item set remotetimeout @var{num}
20146 @cindex timeout for serial communications
20147 @cindex remote timeout
20148 Set the timeout limit to wait for the remote target to respond to
20149 @var{num} seconds. The default is 2 seconds.
20151 @item show remotetimeout
20152 Show the current number of seconds to wait for the remote target
20155 @cindex limit hardware breakpoints and watchpoints
20156 @cindex remote target, limit break- and watchpoints
20157 @anchor{set remote hardware-watchpoint-limit}
20158 @anchor{set remote hardware-breakpoint-limit}
20159 @item set remote hardware-watchpoint-limit @var{limit}
20160 @itemx set remote hardware-breakpoint-limit @var{limit}
20161 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
20162 watchpoints. A limit of -1, the default, is treated as unlimited.
20164 @cindex limit hardware watchpoints length
20165 @cindex remote target, limit watchpoints length
20166 @anchor{set remote hardware-watchpoint-length-limit}
20167 @item set remote hardware-watchpoint-length-limit @var{limit}
20168 Restrict @value{GDBN} to using @var{limit} bytes for the maximum length of
20169 a remote hardware watchpoint. A limit of -1, the default, is treated
20172 @item show remote hardware-watchpoint-length-limit
20173 Show the current limit (in bytes) of the maximum length of
20174 a remote hardware watchpoint.
20176 @item set remote exec-file @var{filename}
20177 @itemx show remote exec-file
20178 @anchor{set remote exec-file}
20179 @cindex executable file, for remote target
20180 Select the file used for @code{run} with @code{target
20181 extended-remote}. This should be set to a filename valid on the
20182 target system. If it is not set, the target will use a default
20183 filename (e.g.@: the last program run).
20185 @item set remote interrupt-sequence
20186 @cindex interrupt remote programs
20187 @cindex select Ctrl-C, BREAK or BREAK-g
20188 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
20189 @samp{BREAK-g} as the
20190 sequence to the remote target in order to interrupt the execution.
20191 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
20192 is high level of serial line for some certain time.
20193 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
20194 It is @code{BREAK} signal followed by character @code{g}.
20196 @item show interrupt-sequence
20197 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
20198 is sent by @value{GDBN} to interrupt the remote program.
20199 @code{BREAK-g} is BREAK signal followed by @code{g} and
20200 also known as Magic SysRq g.
20202 @item set remote interrupt-on-connect
20203 @cindex send interrupt-sequence on start
20204 Specify whether interrupt-sequence is sent to remote target when
20205 @value{GDBN} connects to it. This is mostly needed when you debug
20206 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
20207 which is known as Magic SysRq g in order to connect @value{GDBN}.
20209 @item show interrupt-on-connect
20210 Show whether interrupt-sequence is sent
20211 to remote target when @value{GDBN} connects to it.
20215 @item set tcp auto-retry on
20216 @cindex auto-retry, for remote TCP target
20217 Enable auto-retry for remote TCP connections. This is useful if the remote
20218 debugging agent is launched in parallel with @value{GDBN}; there is a race
20219 condition because the agent may not become ready to accept the connection
20220 before @value{GDBN} attempts to connect. When auto-retry is
20221 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
20222 to establish the connection using the timeout specified by
20223 @code{set tcp connect-timeout}.
20225 @item set tcp auto-retry off
20226 Do not auto-retry failed TCP connections.
20228 @item show tcp auto-retry
20229 Show the current auto-retry setting.
20231 @item set tcp connect-timeout @var{seconds}
20232 @itemx set tcp connect-timeout unlimited
20233 @cindex connection timeout, for remote TCP target
20234 @cindex timeout, for remote target connection
20235 Set the timeout for establishing a TCP connection to the remote target to
20236 @var{seconds}. The timeout affects both polling to retry failed connections
20237 (enabled by @code{set tcp auto-retry on}) and waiting for connections
20238 that are merely slow to complete, and represents an approximate cumulative
20239 value. If @var{seconds} is @code{unlimited}, there is no timeout and
20240 @value{GDBN} will keep attempting to establish a connection forever,
20241 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
20243 @item show tcp connect-timeout
20244 Show the current connection timeout setting.
20247 @cindex remote packets, enabling and disabling
20248 The @value{GDBN} remote protocol autodetects the packets supported by
20249 your debugging stub. If you need to override the autodetection, you
20250 can use these commands to enable or disable individual packets. Each
20251 packet can be set to @samp{on} (the remote target supports this
20252 packet), @samp{off} (the remote target does not support this packet),
20253 or @samp{auto} (detect remote target support for this packet). They
20254 all default to @samp{auto}. For more information about each packet,
20255 see @ref{Remote Protocol}.
20257 During normal use, you should not have to use any of these commands.
20258 If you do, that may be a bug in your remote debugging stub, or a bug
20259 in @value{GDBN}. You may want to report the problem to the
20260 @value{GDBN} developers.
20262 For each packet @var{name}, the command to enable or disable the
20263 packet is @code{set remote @var{name}-packet}. The available settings
20266 @multitable @columnfractions 0.28 0.32 0.25
20269 @tab Related Features
20271 @item @code{fetch-register}
20273 @tab @code{info registers}
20275 @item @code{set-register}
20279 @item @code{binary-download}
20281 @tab @code{load}, @code{set}
20283 @item @code{read-aux-vector}
20284 @tab @code{qXfer:auxv:read}
20285 @tab @code{info auxv}
20287 @item @code{symbol-lookup}
20288 @tab @code{qSymbol}
20289 @tab Detecting multiple threads
20291 @item @code{attach}
20292 @tab @code{vAttach}
20295 @item @code{verbose-resume}
20297 @tab Stepping or resuming multiple threads
20303 @item @code{software-breakpoint}
20307 @item @code{hardware-breakpoint}
20311 @item @code{write-watchpoint}
20315 @item @code{read-watchpoint}
20319 @item @code{access-watchpoint}
20323 @item @code{pid-to-exec-file}
20324 @tab @code{qXfer:exec-file:read}
20325 @tab @code{attach}, @code{run}
20327 @item @code{target-features}
20328 @tab @code{qXfer:features:read}
20329 @tab @code{set architecture}
20331 @item @code{library-info}
20332 @tab @code{qXfer:libraries:read}
20333 @tab @code{info sharedlibrary}
20335 @item @code{memory-map}
20336 @tab @code{qXfer:memory-map:read}
20337 @tab @code{info mem}
20339 @item @code{read-sdata-object}
20340 @tab @code{qXfer:sdata:read}
20341 @tab @code{print $_sdata}
20343 @item @code{read-spu-object}
20344 @tab @code{qXfer:spu:read}
20345 @tab @code{info spu}
20347 @item @code{write-spu-object}
20348 @tab @code{qXfer:spu:write}
20349 @tab @code{info spu}
20351 @item @code{read-siginfo-object}
20352 @tab @code{qXfer:siginfo:read}
20353 @tab @code{print $_siginfo}
20355 @item @code{write-siginfo-object}
20356 @tab @code{qXfer:siginfo:write}
20357 @tab @code{set $_siginfo}
20359 @item @code{threads}
20360 @tab @code{qXfer:threads:read}
20361 @tab @code{info threads}
20363 @item @code{get-thread-local-@*storage-address}
20364 @tab @code{qGetTLSAddr}
20365 @tab Displaying @code{__thread} variables
20367 @item @code{get-thread-information-block-address}
20368 @tab @code{qGetTIBAddr}
20369 @tab Display MS-Windows Thread Information Block.
20371 @item @code{search-memory}
20372 @tab @code{qSearch:memory}
20375 @item @code{supported-packets}
20376 @tab @code{qSupported}
20377 @tab Remote communications parameters
20379 @item @code{catch-syscalls}
20380 @tab @code{QCatchSyscalls}
20381 @tab @code{catch syscall}
20383 @item @code{pass-signals}
20384 @tab @code{QPassSignals}
20385 @tab @code{handle @var{signal}}
20387 @item @code{program-signals}
20388 @tab @code{QProgramSignals}
20389 @tab @code{handle @var{signal}}
20391 @item @code{hostio-close-packet}
20392 @tab @code{vFile:close}
20393 @tab @code{remote get}, @code{remote put}
20395 @item @code{hostio-open-packet}
20396 @tab @code{vFile:open}
20397 @tab @code{remote get}, @code{remote put}
20399 @item @code{hostio-pread-packet}
20400 @tab @code{vFile:pread}
20401 @tab @code{remote get}, @code{remote put}
20403 @item @code{hostio-pwrite-packet}
20404 @tab @code{vFile:pwrite}
20405 @tab @code{remote get}, @code{remote put}
20407 @item @code{hostio-unlink-packet}
20408 @tab @code{vFile:unlink}
20409 @tab @code{remote delete}
20411 @item @code{hostio-readlink-packet}
20412 @tab @code{vFile:readlink}
20415 @item @code{hostio-fstat-packet}
20416 @tab @code{vFile:fstat}
20419 @item @code{hostio-setfs-packet}
20420 @tab @code{vFile:setfs}
20423 @item @code{noack-packet}
20424 @tab @code{QStartNoAckMode}
20425 @tab Packet acknowledgment
20427 @item @code{osdata}
20428 @tab @code{qXfer:osdata:read}
20429 @tab @code{info os}
20431 @item @code{query-attached}
20432 @tab @code{qAttached}
20433 @tab Querying remote process attach state.
20435 @item @code{trace-buffer-size}
20436 @tab @code{QTBuffer:size}
20437 @tab @code{set trace-buffer-size}
20439 @item @code{trace-status}
20440 @tab @code{qTStatus}
20441 @tab @code{tstatus}
20443 @item @code{traceframe-info}
20444 @tab @code{qXfer:traceframe-info:read}
20445 @tab Traceframe info
20447 @item @code{install-in-trace}
20448 @tab @code{InstallInTrace}
20449 @tab Install tracepoint in tracing
20451 @item @code{disable-randomization}
20452 @tab @code{QDisableRandomization}
20453 @tab @code{set disable-randomization}
20455 @item @code{conditional-breakpoints-packet}
20456 @tab @code{Z0 and Z1}
20457 @tab @code{Support for target-side breakpoint condition evaluation}
20459 @item @code{multiprocess-extensions}
20460 @tab @code{multiprocess extensions}
20461 @tab Debug multiple processes and remote process PID awareness
20463 @item @code{swbreak-feature}
20464 @tab @code{swbreak stop reason}
20467 @item @code{hwbreak-feature}
20468 @tab @code{hwbreak stop reason}
20471 @item @code{fork-event-feature}
20472 @tab @code{fork stop reason}
20475 @item @code{vfork-event-feature}
20476 @tab @code{vfork stop reason}
20479 @item @code{exec-event-feature}
20480 @tab @code{exec stop reason}
20483 @item @code{thread-events}
20484 @tab @code{QThreadEvents}
20485 @tab Tracking thread lifetime.
20487 @item @code{no-resumed-stop-reply}
20488 @tab @code{no resumed thread left stop reply}
20489 @tab Tracking thread lifetime.
20494 @section Implementing a Remote Stub
20496 @cindex debugging stub, example
20497 @cindex remote stub, example
20498 @cindex stub example, remote debugging
20499 The stub files provided with @value{GDBN} implement the target side of the
20500 communication protocol, and the @value{GDBN} side is implemented in the
20501 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
20502 these subroutines to communicate, and ignore the details. (If you're
20503 implementing your own stub file, you can still ignore the details: start
20504 with one of the existing stub files. @file{sparc-stub.c} is the best
20505 organized, and therefore the easiest to read.)
20507 @cindex remote serial debugging, overview
20508 To debug a program running on another machine (the debugging
20509 @dfn{target} machine), you must first arrange for all the usual
20510 prerequisites for the program to run by itself. For example, for a C
20515 A startup routine to set up the C runtime environment; these usually
20516 have a name like @file{crt0}. The startup routine may be supplied by
20517 your hardware supplier, or you may have to write your own.
20520 A C subroutine library to support your program's
20521 subroutine calls, notably managing input and output.
20524 A way of getting your program to the other machine---for example, a
20525 download program. These are often supplied by the hardware
20526 manufacturer, but you may have to write your own from hardware
20530 The next step is to arrange for your program to use a serial port to
20531 communicate with the machine where @value{GDBN} is running (the @dfn{host}
20532 machine). In general terms, the scheme looks like this:
20536 @value{GDBN} already understands how to use this protocol; when everything
20537 else is set up, you can simply use the @samp{target remote} command
20538 (@pxref{Targets,,Specifying a Debugging Target}).
20540 @item On the target,
20541 you must link with your program a few special-purpose subroutines that
20542 implement the @value{GDBN} remote serial protocol. The file containing these
20543 subroutines is called a @dfn{debugging stub}.
20545 On certain remote targets, you can use an auxiliary program
20546 @code{gdbserver} instead of linking a stub into your program.
20547 @xref{Server,,Using the @code{gdbserver} Program}, for details.
20550 The debugging stub is specific to the architecture of the remote
20551 machine; for example, use @file{sparc-stub.c} to debug programs on
20554 @cindex remote serial stub list
20555 These working remote stubs are distributed with @value{GDBN}:
20560 @cindex @file{i386-stub.c}
20563 For Intel 386 and compatible architectures.
20566 @cindex @file{m68k-stub.c}
20567 @cindex Motorola 680x0
20569 For Motorola 680x0 architectures.
20572 @cindex @file{sh-stub.c}
20575 For Renesas SH architectures.
20578 @cindex @file{sparc-stub.c}
20580 For @sc{sparc} architectures.
20582 @item sparcl-stub.c
20583 @cindex @file{sparcl-stub.c}
20586 For Fujitsu @sc{sparclite} architectures.
20590 The @file{README} file in the @value{GDBN} distribution may list other
20591 recently added stubs.
20594 * Stub Contents:: What the stub can do for you
20595 * Bootstrapping:: What you must do for the stub
20596 * Debug Session:: Putting it all together
20599 @node Stub Contents
20600 @subsection What the Stub Can Do for You
20602 @cindex remote serial stub
20603 The debugging stub for your architecture supplies these three
20607 @item set_debug_traps
20608 @findex set_debug_traps
20609 @cindex remote serial stub, initialization
20610 This routine arranges for @code{handle_exception} to run when your
20611 program stops. You must call this subroutine explicitly in your
20612 program's startup code.
20614 @item handle_exception
20615 @findex handle_exception
20616 @cindex remote serial stub, main routine
20617 This is the central workhorse, but your program never calls it
20618 explicitly---the setup code arranges for @code{handle_exception} to
20619 run when a trap is triggered.
20621 @code{handle_exception} takes control when your program stops during
20622 execution (for example, on a breakpoint), and mediates communications
20623 with @value{GDBN} on the host machine. This is where the communications
20624 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
20625 representative on the target machine. It begins by sending summary
20626 information on the state of your program, then continues to execute,
20627 retrieving and transmitting any information @value{GDBN} needs, until you
20628 execute a @value{GDBN} command that makes your program resume; at that point,
20629 @code{handle_exception} returns control to your own code on the target
20633 @cindex @code{breakpoint} subroutine, remote
20634 Use this auxiliary subroutine to make your program contain a
20635 breakpoint. Depending on the particular situation, this may be the only
20636 way for @value{GDBN} to get control. For instance, if your target
20637 machine has some sort of interrupt button, you won't need to call this;
20638 pressing the interrupt button transfers control to
20639 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
20640 simply receiving characters on the serial port may also trigger a trap;
20641 again, in that situation, you don't need to call @code{breakpoint} from
20642 your own program---simply running @samp{target remote} from the host
20643 @value{GDBN} session gets control.
20645 Call @code{breakpoint} if none of these is true, or if you simply want
20646 to make certain your program stops at a predetermined point for the
20647 start of your debugging session.
20650 @node Bootstrapping
20651 @subsection What You Must Do for the Stub
20653 @cindex remote stub, support routines
20654 The debugging stubs that come with @value{GDBN} are set up for a particular
20655 chip architecture, but they have no information about the rest of your
20656 debugging target machine.
20658 First of all you need to tell the stub how to communicate with the
20662 @item int getDebugChar()
20663 @findex getDebugChar
20664 Write this subroutine to read a single character from the serial port.
20665 It may be identical to @code{getchar} for your target system; a
20666 different name is used to allow you to distinguish the two if you wish.
20668 @item void putDebugChar(int)
20669 @findex putDebugChar
20670 Write this subroutine to write a single character to the serial port.
20671 It may be identical to @code{putchar} for your target system; a
20672 different name is used to allow you to distinguish the two if you wish.
20675 @cindex control C, and remote debugging
20676 @cindex interrupting remote targets
20677 If you want @value{GDBN} to be able to stop your program while it is
20678 running, you need to use an interrupt-driven serial driver, and arrange
20679 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
20680 character). That is the character which @value{GDBN} uses to tell the
20681 remote system to stop.
20683 Getting the debugging target to return the proper status to @value{GDBN}
20684 probably requires changes to the standard stub; one quick and dirty way
20685 is to just execute a breakpoint instruction (the ``dirty'' part is that
20686 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
20688 Other routines you need to supply are:
20691 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
20692 @findex exceptionHandler
20693 Write this function to install @var{exception_address} in the exception
20694 handling tables. You need to do this because the stub does not have any
20695 way of knowing what the exception handling tables on your target system
20696 are like (for example, the processor's table might be in @sc{rom},
20697 containing entries which point to a table in @sc{ram}).
20698 The @var{exception_number} specifies the exception which should be changed;
20699 its meaning is architecture-dependent (for example, different numbers
20700 might represent divide by zero, misaligned access, etc). When this
20701 exception occurs, control should be transferred directly to
20702 @var{exception_address}, and the processor state (stack, registers,
20703 and so on) should be just as it is when a processor exception occurs. So if
20704 you want to use a jump instruction to reach @var{exception_address}, it
20705 should be a simple jump, not a jump to subroutine.
20707 For the 386, @var{exception_address} should be installed as an interrupt
20708 gate so that interrupts are masked while the handler runs. The gate
20709 should be at privilege level 0 (the most privileged level). The
20710 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
20711 help from @code{exceptionHandler}.
20713 @item void flush_i_cache()
20714 @findex flush_i_cache
20715 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
20716 instruction cache, if any, on your target machine. If there is no
20717 instruction cache, this subroutine may be a no-op.
20719 On target machines that have instruction caches, @value{GDBN} requires this
20720 function to make certain that the state of your program is stable.
20724 You must also make sure this library routine is available:
20727 @item void *memset(void *, int, int)
20729 This is the standard library function @code{memset} that sets an area of
20730 memory to a known value. If you have one of the free versions of
20731 @code{libc.a}, @code{memset} can be found there; otherwise, you must
20732 either obtain it from your hardware manufacturer, or write your own.
20735 If you do not use the GNU C compiler, you may need other standard
20736 library subroutines as well; this varies from one stub to another,
20737 but in general the stubs are likely to use any of the common library
20738 subroutines which @code{@value{NGCC}} generates as inline code.
20741 @node Debug Session
20742 @subsection Putting it All Together
20744 @cindex remote serial debugging summary
20745 In summary, when your program is ready to debug, you must follow these
20750 Make sure you have defined the supporting low-level routines
20751 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
20753 @code{getDebugChar}, @code{putDebugChar},
20754 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
20758 Insert these lines in your program's startup code, before the main
20759 procedure is called:
20766 On some machines, when a breakpoint trap is raised, the hardware
20767 automatically makes the PC point to the instruction after the
20768 breakpoint. If your machine doesn't do that, you may need to adjust
20769 @code{handle_exception} to arrange for it to return to the instruction
20770 after the breakpoint on this first invocation, so that your program
20771 doesn't keep hitting the initial breakpoint instead of making
20775 For the 680x0 stub only, you need to provide a variable called
20776 @code{exceptionHook}. Normally you just use:
20779 void (*exceptionHook)() = 0;
20783 but if before calling @code{set_debug_traps}, you set it to point to a
20784 function in your program, that function is called when
20785 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
20786 error). The function indicated by @code{exceptionHook} is called with
20787 one parameter: an @code{int} which is the exception number.
20790 Compile and link together: your program, the @value{GDBN} debugging stub for
20791 your target architecture, and the supporting subroutines.
20794 Make sure you have a serial connection between your target machine and
20795 the @value{GDBN} host, and identify the serial port on the host.
20798 @c The "remote" target now provides a `load' command, so we should
20799 @c document that. FIXME.
20800 Download your program to your target machine (or get it there by
20801 whatever means the manufacturer provides), and start it.
20804 Start @value{GDBN} on the host, and connect to the target
20805 (@pxref{Connecting,,Connecting to a Remote Target}).
20809 @node Configurations
20810 @chapter Configuration-Specific Information
20812 While nearly all @value{GDBN} commands are available for all native and
20813 cross versions of the debugger, there are some exceptions. This chapter
20814 describes things that are only available in certain configurations.
20816 There are three major categories of configurations: native
20817 configurations, where the host and target are the same, embedded
20818 operating system configurations, which are usually the same for several
20819 different processor architectures, and bare embedded processors, which
20820 are quite different from each other.
20825 * Embedded Processors::
20832 This section describes details specific to particular native
20836 * BSD libkvm Interface:: Debugging BSD kernel memory images
20837 * SVR4 Process Information:: SVR4 process information
20838 * DJGPP Native:: Features specific to the DJGPP port
20839 * Cygwin Native:: Features specific to the Cygwin port
20840 * Hurd Native:: Features specific to @sc{gnu} Hurd
20841 * Darwin:: Features specific to Darwin
20844 @node BSD libkvm Interface
20845 @subsection BSD libkvm Interface
20848 @cindex kernel memory image
20849 @cindex kernel crash dump
20851 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
20852 interface that provides a uniform interface for accessing kernel virtual
20853 memory images, including live systems and crash dumps. @value{GDBN}
20854 uses this interface to allow you to debug live kernels and kernel crash
20855 dumps on many native BSD configurations. This is implemented as a
20856 special @code{kvm} debugging target. For debugging a live system, load
20857 the currently running kernel into @value{GDBN} and connect to the
20861 (@value{GDBP}) @b{target kvm}
20864 For debugging crash dumps, provide the file name of the crash dump as an
20868 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
20871 Once connected to the @code{kvm} target, the following commands are
20877 Set current context from the @dfn{Process Control Block} (PCB) address.
20880 Set current context from proc address. This command isn't available on
20881 modern FreeBSD systems.
20884 @node SVR4 Process Information
20885 @subsection SVR4 Process Information
20887 @cindex examine process image
20888 @cindex process info via @file{/proc}
20890 Many versions of SVR4 and compatible systems provide a facility called
20891 @samp{/proc} that can be used to examine the image of a running
20892 process using file-system subroutines.
20894 If @value{GDBN} is configured for an operating system with this
20895 facility, the command @code{info proc} is available to report
20896 information about the process running your program, or about any
20897 process running on your system. This includes, as of this writing,
20898 @sc{gnu}/Linux and Solaris, for example.
20900 This command may also work on core files that were created on a system
20901 that has the @samp{/proc} facility.
20907 @itemx info proc @var{process-id}
20908 Summarize available information about any running process. If a
20909 process ID is specified by @var{process-id}, display information about
20910 that process; otherwise display information about the program being
20911 debugged. The summary includes the debugged process ID, the command
20912 line used to invoke it, its current working directory, and its
20913 executable file's absolute file name.
20915 On some systems, @var{process-id} can be of the form
20916 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
20917 within a process. If the optional @var{pid} part is missing, it means
20918 a thread from the process being debugged (the leading @samp{/} still
20919 needs to be present, or else @value{GDBN} will interpret the number as
20920 a process ID rather than a thread ID).
20922 @item info proc cmdline
20923 @cindex info proc cmdline
20924 Show the original command line of the process. This command is
20925 specific to @sc{gnu}/Linux.
20927 @item info proc cwd
20928 @cindex info proc cwd
20929 Show the current working directory of the process. This command is
20930 specific to @sc{gnu}/Linux.
20932 @item info proc exe
20933 @cindex info proc exe
20934 Show the name of executable of the process. This command is specific
20937 @item info proc mappings
20938 @cindex memory address space mappings
20939 Report the memory address space ranges accessible in the program, with
20940 information on whether the process has read, write, or execute access
20941 rights to each range. On @sc{gnu}/Linux systems, each memory range
20942 includes the object file which is mapped to that range, instead of the
20943 memory access rights to that range.
20945 @item info proc stat
20946 @itemx info proc status
20947 @cindex process detailed status information
20948 These subcommands are specific to @sc{gnu}/Linux systems. They show
20949 the process-related information, including the user ID and group ID;
20950 how many threads are there in the process; its virtual memory usage;
20951 the signals that are pending, blocked, and ignored; its TTY; its
20952 consumption of system and user time; its stack size; its @samp{nice}
20953 value; etc. For more information, see the @samp{proc} man page
20954 (type @kbd{man 5 proc} from your shell prompt).
20956 @item info proc all
20957 Show all the information about the process described under all of the
20958 above @code{info proc} subcommands.
20961 @comment These sub-options of 'info proc' were not included when
20962 @comment procfs.c was re-written. Keep their descriptions around
20963 @comment against the day when someone finds the time to put them back in.
20964 @kindex info proc times
20965 @item info proc times
20966 Starting time, user CPU time, and system CPU time for your program and
20969 @kindex info proc id
20971 Report on the process IDs related to your program: its own process ID,
20972 the ID of its parent, the process group ID, and the session ID.
20975 @item set procfs-trace
20976 @kindex set procfs-trace
20977 @cindex @code{procfs} API calls
20978 This command enables and disables tracing of @code{procfs} API calls.
20980 @item show procfs-trace
20981 @kindex show procfs-trace
20982 Show the current state of @code{procfs} API call tracing.
20984 @item set procfs-file @var{file}
20985 @kindex set procfs-file
20986 Tell @value{GDBN} to write @code{procfs} API trace to the named
20987 @var{file}. @value{GDBN} appends the trace info to the previous
20988 contents of the file. The default is to display the trace on the
20991 @item show procfs-file
20992 @kindex show procfs-file
20993 Show the file to which @code{procfs} API trace is written.
20995 @item proc-trace-entry
20996 @itemx proc-trace-exit
20997 @itemx proc-untrace-entry
20998 @itemx proc-untrace-exit
20999 @kindex proc-trace-entry
21000 @kindex proc-trace-exit
21001 @kindex proc-untrace-entry
21002 @kindex proc-untrace-exit
21003 These commands enable and disable tracing of entries into and exits
21004 from the @code{syscall} interface.
21007 @kindex info pidlist
21008 @cindex process list, QNX Neutrino
21009 For QNX Neutrino only, this command displays the list of all the
21010 processes and all the threads within each process.
21013 @kindex info meminfo
21014 @cindex mapinfo list, QNX Neutrino
21015 For QNX Neutrino only, this command displays the list of all mapinfos.
21019 @subsection Features for Debugging @sc{djgpp} Programs
21020 @cindex @sc{djgpp} debugging
21021 @cindex native @sc{djgpp} debugging
21022 @cindex MS-DOS-specific commands
21025 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
21026 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
21027 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
21028 top of real-mode DOS systems and their emulations.
21030 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
21031 defines a few commands specific to the @sc{djgpp} port. This
21032 subsection describes those commands.
21037 This is a prefix of @sc{djgpp}-specific commands which print
21038 information about the target system and important OS structures.
21041 @cindex MS-DOS system info
21042 @cindex free memory information (MS-DOS)
21043 @item info dos sysinfo
21044 This command displays assorted information about the underlying
21045 platform: the CPU type and features, the OS version and flavor, the
21046 DPMI version, and the available conventional and DPMI memory.
21051 @cindex segment descriptor tables
21052 @cindex descriptor tables display
21054 @itemx info dos ldt
21055 @itemx info dos idt
21056 These 3 commands display entries from, respectively, Global, Local,
21057 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
21058 tables are data structures which store a descriptor for each segment
21059 that is currently in use. The segment's selector is an index into a
21060 descriptor table; the table entry for that index holds the
21061 descriptor's base address and limit, and its attributes and access
21064 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
21065 segment (used for both data and the stack), and a DOS segment (which
21066 allows access to DOS/BIOS data structures and absolute addresses in
21067 conventional memory). However, the DPMI host will usually define
21068 additional segments in order to support the DPMI environment.
21070 @cindex garbled pointers
21071 These commands allow to display entries from the descriptor tables.
21072 Without an argument, all entries from the specified table are
21073 displayed. An argument, which should be an integer expression, means
21074 display a single entry whose index is given by the argument. For
21075 example, here's a convenient way to display information about the
21076 debugged program's data segment:
21079 @exdent @code{(@value{GDBP}) info dos ldt $ds}
21080 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
21084 This comes in handy when you want to see whether a pointer is outside
21085 the data segment's limit (i.e.@: @dfn{garbled}).
21087 @cindex page tables display (MS-DOS)
21089 @itemx info dos pte
21090 These two commands display entries from, respectively, the Page
21091 Directory and the Page Tables. Page Directories and Page Tables are
21092 data structures which control how virtual memory addresses are mapped
21093 into physical addresses. A Page Table includes an entry for every
21094 page of memory that is mapped into the program's address space; there
21095 may be several Page Tables, each one holding up to 4096 entries. A
21096 Page Directory has up to 4096 entries, one each for every Page Table
21097 that is currently in use.
21099 Without an argument, @kbd{info dos pde} displays the entire Page
21100 Directory, and @kbd{info dos pte} displays all the entries in all of
21101 the Page Tables. An argument, an integer expression, given to the
21102 @kbd{info dos pde} command means display only that entry from the Page
21103 Directory table. An argument given to the @kbd{info dos pte} command
21104 means display entries from a single Page Table, the one pointed to by
21105 the specified entry in the Page Directory.
21107 @cindex direct memory access (DMA) on MS-DOS
21108 These commands are useful when your program uses @dfn{DMA} (Direct
21109 Memory Access), which needs physical addresses to program the DMA
21112 These commands are supported only with some DPMI servers.
21114 @cindex physical address from linear address
21115 @item info dos address-pte @var{addr}
21116 This command displays the Page Table entry for a specified linear
21117 address. The argument @var{addr} is a linear address which should
21118 already have the appropriate segment's base address added to it,
21119 because this command accepts addresses which may belong to @emph{any}
21120 segment. For example, here's how to display the Page Table entry for
21121 the page where a variable @code{i} is stored:
21124 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
21125 @exdent @code{Page Table entry for address 0x11a00d30:}
21126 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
21130 This says that @code{i} is stored at offset @code{0xd30} from the page
21131 whose physical base address is @code{0x02698000}, and shows all the
21132 attributes of that page.
21134 Note that you must cast the addresses of variables to a @code{char *},
21135 since otherwise the value of @code{__djgpp_base_address}, the base
21136 address of all variables and functions in a @sc{djgpp} program, will
21137 be added using the rules of C pointer arithmetics: if @code{i} is
21138 declared an @code{int}, @value{GDBN} will add 4 times the value of
21139 @code{__djgpp_base_address} to the address of @code{i}.
21141 Here's another example, it displays the Page Table entry for the
21145 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
21146 @exdent @code{Page Table entry for address 0x29110:}
21147 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
21151 (The @code{+ 3} offset is because the transfer buffer's address is the
21152 3rd member of the @code{_go32_info_block} structure.) The output
21153 clearly shows that this DPMI server maps the addresses in conventional
21154 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
21155 linear (@code{0x29110}) addresses are identical.
21157 This command is supported only with some DPMI servers.
21160 @cindex DOS serial data link, remote debugging
21161 In addition to native debugging, the DJGPP port supports remote
21162 debugging via a serial data link. The following commands are specific
21163 to remote serial debugging in the DJGPP port of @value{GDBN}.
21166 @kindex set com1base
21167 @kindex set com1irq
21168 @kindex set com2base
21169 @kindex set com2irq
21170 @kindex set com3base
21171 @kindex set com3irq
21172 @kindex set com4base
21173 @kindex set com4irq
21174 @item set com1base @var{addr}
21175 This command sets the base I/O port address of the @file{COM1} serial
21178 @item set com1irq @var{irq}
21179 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
21180 for the @file{COM1} serial port.
21182 There are similar commands @samp{set com2base}, @samp{set com3irq},
21183 etc.@: for setting the port address and the @code{IRQ} lines for the
21186 @kindex show com1base
21187 @kindex show com1irq
21188 @kindex show com2base
21189 @kindex show com2irq
21190 @kindex show com3base
21191 @kindex show com3irq
21192 @kindex show com4base
21193 @kindex show com4irq
21194 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
21195 display the current settings of the base address and the @code{IRQ}
21196 lines used by the COM ports.
21199 @kindex info serial
21200 @cindex DOS serial port status
21201 This command prints the status of the 4 DOS serial ports. For each
21202 port, it prints whether it's active or not, its I/O base address and
21203 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
21204 counts of various errors encountered so far.
21208 @node Cygwin Native
21209 @subsection Features for Debugging MS Windows PE Executables
21210 @cindex MS Windows debugging
21211 @cindex native Cygwin debugging
21212 @cindex Cygwin-specific commands
21214 @value{GDBN} supports native debugging of MS Windows programs, including
21215 DLLs with and without symbolic debugging information.
21217 @cindex Ctrl-BREAK, MS-Windows
21218 @cindex interrupt debuggee on MS-Windows
21219 MS-Windows programs that call @code{SetConsoleMode} to switch off the
21220 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
21221 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
21222 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
21223 sequence, which can be used to interrupt the debuggee even if it
21226 There are various additional Cygwin-specific commands, described in
21227 this section. Working with DLLs that have no debugging symbols is
21228 described in @ref{Non-debug DLL Symbols}.
21233 This is a prefix of MS Windows-specific commands which print
21234 information about the target system and important OS structures.
21236 @item info w32 selector
21237 This command displays information returned by
21238 the Win32 API @code{GetThreadSelectorEntry} function.
21239 It takes an optional argument that is evaluated to
21240 a long value to give the information about this given selector.
21241 Without argument, this command displays information
21242 about the six segment registers.
21244 @item info w32 thread-information-block
21245 This command displays thread specific information stored in the
21246 Thread Information Block (readable on the X86 CPU family using @code{$fs}
21247 selector for 32-bit programs and @code{$gs} for 64-bit programs).
21249 @kindex set cygwin-exceptions
21250 @cindex debugging the Cygwin DLL
21251 @cindex Cygwin DLL, debugging
21252 @item set cygwin-exceptions @var{mode}
21253 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
21254 happen inside the Cygwin DLL. If @var{mode} is @code{off},
21255 @value{GDBN} will delay recognition of exceptions, and may ignore some
21256 exceptions which seem to be caused by internal Cygwin DLL
21257 ``bookkeeping''. This option is meant primarily for debugging the
21258 Cygwin DLL itself; the default value is @code{off} to avoid annoying
21259 @value{GDBN} users with false @code{SIGSEGV} signals.
21261 @kindex show cygwin-exceptions
21262 @item show cygwin-exceptions
21263 Displays whether @value{GDBN} will break on exceptions that happen
21264 inside the Cygwin DLL itself.
21266 @kindex set new-console
21267 @item set new-console @var{mode}
21268 If @var{mode} is @code{on} the debuggee will
21269 be started in a new console on next start.
21270 If @var{mode} is @code{off}, the debuggee will
21271 be started in the same console as the debugger.
21273 @kindex show new-console
21274 @item show new-console
21275 Displays whether a new console is used
21276 when the debuggee is started.
21278 @kindex set new-group
21279 @item set new-group @var{mode}
21280 This boolean value controls whether the debuggee should
21281 start a new group or stay in the same group as the debugger.
21282 This affects the way the Windows OS handles
21285 @kindex show new-group
21286 @item show new-group
21287 Displays current value of new-group boolean.
21289 @kindex set debugevents
21290 @item set debugevents
21291 This boolean value adds debug output concerning kernel events related
21292 to the debuggee seen by the debugger. This includes events that
21293 signal thread and process creation and exit, DLL loading and
21294 unloading, console interrupts, and debugging messages produced by the
21295 Windows @code{OutputDebugString} API call.
21297 @kindex set debugexec
21298 @item set debugexec
21299 This boolean value adds debug output concerning execute events
21300 (such as resume thread) seen by the debugger.
21302 @kindex set debugexceptions
21303 @item set debugexceptions
21304 This boolean value adds debug output concerning exceptions in the
21305 debuggee seen by the debugger.
21307 @kindex set debugmemory
21308 @item set debugmemory
21309 This boolean value adds debug output concerning debuggee memory reads
21310 and writes by the debugger.
21314 This boolean values specifies whether the debuggee is called
21315 via a shell or directly (default value is on).
21319 Displays if the debuggee will be started with a shell.
21324 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
21327 @node Non-debug DLL Symbols
21328 @subsubsection Support for DLLs without Debugging Symbols
21329 @cindex DLLs with no debugging symbols
21330 @cindex Minimal symbols and DLLs
21332 Very often on windows, some of the DLLs that your program relies on do
21333 not include symbolic debugging information (for example,
21334 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
21335 symbols in a DLL, it relies on the minimal amount of symbolic
21336 information contained in the DLL's export table. This section
21337 describes working with such symbols, known internally to @value{GDBN} as
21338 ``minimal symbols''.
21340 Note that before the debugged program has started execution, no DLLs
21341 will have been loaded. The easiest way around this problem is simply to
21342 start the program --- either by setting a breakpoint or letting the
21343 program run once to completion.
21345 @subsubsection DLL Name Prefixes
21347 In keeping with the naming conventions used by the Microsoft debugging
21348 tools, DLL export symbols are made available with a prefix based on the
21349 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
21350 also entered into the symbol table, so @code{CreateFileA} is often
21351 sufficient. In some cases there will be name clashes within a program
21352 (particularly if the executable itself includes full debugging symbols)
21353 necessitating the use of the fully qualified name when referring to the
21354 contents of the DLL. Use single-quotes around the name to avoid the
21355 exclamation mark (``!'') being interpreted as a language operator.
21357 Note that the internal name of the DLL may be all upper-case, even
21358 though the file name of the DLL is lower-case, or vice-versa. Since
21359 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
21360 some confusion. If in doubt, try the @code{info functions} and
21361 @code{info variables} commands or even @code{maint print msymbols}
21362 (@pxref{Symbols}). Here's an example:
21365 (@value{GDBP}) info function CreateFileA
21366 All functions matching regular expression "CreateFileA":
21368 Non-debugging symbols:
21369 0x77e885f4 CreateFileA
21370 0x77e885f4 KERNEL32!CreateFileA
21374 (@value{GDBP}) info function !
21375 All functions matching regular expression "!":
21377 Non-debugging symbols:
21378 0x6100114c cygwin1!__assert
21379 0x61004034 cygwin1!_dll_crt0@@0
21380 0x61004240 cygwin1!dll_crt0(per_process *)
21384 @subsubsection Working with Minimal Symbols
21386 Symbols extracted from a DLL's export table do not contain very much
21387 type information. All that @value{GDBN} can do is guess whether a symbol
21388 refers to a function or variable depending on the linker section that
21389 contains the symbol. Also note that the actual contents of the memory
21390 contained in a DLL are not available unless the program is running. This
21391 means that you cannot examine the contents of a variable or disassemble
21392 a function within a DLL without a running program.
21394 Variables are generally treated as pointers and dereferenced
21395 automatically. For this reason, it is often necessary to prefix a
21396 variable name with the address-of operator (``&'') and provide explicit
21397 type information in the command. Here's an example of the type of
21401 (@value{GDBP}) print 'cygwin1!__argv'
21406 (@value{GDBP}) x 'cygwin1!__argv'
21407 0x10021610: "\230y\""
21410 And two possible solutions:
21413 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
21414 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
21418 (@value{GDBP}) x/2x &'cygwin1!__argv'
21419 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
21420 (@value{GDBP}) x/x 0x10021608
21421 0x10021608: 0x0022fd98
21422 (@value{GDBP}) x/s 0x0022fd98
21423 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
21426 Setting a break point within a DLL is possible even before the program
21427 starts execution. However, under these circumstances, @value{GDBN} can't
21428 examine the initial instructions of the function in order to skip the
21429 function's frame set-up code. You can work around this by using ``*&''
21430 to set the breakpoint at a raw memory address:
21433 (@value{GDBP}) break *&'python22!PyOS_Readline'
21434 Breakpoint 1 at 0x1e04eff0
21437 The author of these extensions is not entirely convinced that setting a
21438 break point within a shared DLL like @file{kernel32.dll} is completely
21442 @subsection Commands Specific to @sc{gnu} Hurd Systems
21443 @cindex @sc{gnu} Hurd debugging
21445 This subsection describes @value{GDBN} commands specific to the
21446 @sc{gnu} Hurd native debugging.
21451 @kindex set signals@r{, Hurd command}
21452 @kindex set sigs@r{, Hurd command}
21453 This command toggles the state of inferior signal interception by
21454 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
21455 affected by this command. @code{sigs} is a shorthand alias for
21460 @kindex show signals@r{, Hurd command}
21461 @kindex show sigs@r{, Hurd command}
21462 Show the current state of intercepting inferior's signals.
21464 @item set signal-thread
21465 @itemx set sigthread
21466 @kindex set signal-thread
21467 @kindex set sigthread
21468 This command tells @value{GDBN} which thread is the @code{libc} signal
21469 thread. That thread is run when a signal is delivered to a running
21470 process. @code{set sigthread} is the shorthand alias of @code{set
21473 @item show signal-thread
21474 @itemx show sigthread
21475 @kindex show signal-thread
21476 @kindex show sigthread
21477 These two commands show which thread will run when the inferior is
21478 delivered a signal.
21481 @kindex set stopped@r{, Hurd command}
21482 This commands tells @value{GDBN} that the inferior process is stopped,
21483 as with the @code{SIGSTOP} signal. The stopped process can be
21484 continued by delivering a signal to it.
21487 @kindex show stopped@r{, Hurd command}
21488 This command shows whether @value{GDBN} thinks the debuggee is
21491 @item set exceptions
21492 @kindex set exceptions@r{, Hurd command}
21493 Use this command to turn off trapping of exceptions in the inferior.
21494 When exception trapping is off, neither breakpoints nor
21495 single-stepping will work. To restore the default, set exception
21498 @item show exceptions
21499 @kindex show exceptions@r{, Hurd command}
21500 Show the current state of trapping exceptions in the inferior.
21502 @item set task pause
21503 @kindex set task@r{, Hurd commands}
21504 @cindex task attributes (@sc{gnu} Hurd)
21505 @cindex pause current task (@sc{gnu} Hurd)
21506 This command toggles task suspension when @value{GDBN} has control.
21507 Setting it to on takes effect immediately, and the task is suspended
21508 whenever @value{GDBN} gets control. Setting it to off will take
21509 effect the next time the inferior is continued. If this option is set
21510 to off, you can use @code{set thread default pause on} or @code{set
21511 thread pause on} (see below) to pause individual threads.
21513 @item show task pause
21514 @kindex show task@r{, Hurd commands}
21515 Show the current state of task suspension.
21517 @item set task detach-suspend-count
21518 @cindex task suspend count
21519 @cindex detach from task, @sc{gnu} Hurd
21520 This command sets the suspend count the task will be left with when
21521 @value{GDBN} detaches from it.
21523 @item show task detach-suspend-count
21524 Show the suspend count the task will be left with when detaching.
21526 @item set task exception-port
21527 @itemx set task excp
21528 @cindex task exception port, @sc{gnu} Hurd
21529 This command sets the task exception port to which @value{GDBN} will
21530 forward exceptions. The argument should be the value of the @dfn{send
21531 rights} of the task. @code{set task excp} is a shorthand alias.
21533 @item set noninvasive
21534 @cindex noninvasive task options
21535 This command switches @value{GDBN} to a mode that is the least
21536 invasive as far as interfering with the inferior is concerned. This
21537 is the same as using @code{set task pause}, @code{set exceptions}, and
21538 @code{set signals} to values opposite to the defaults.
21540 @item info send-rights
21541 @itemx info receive-rights
21542 @itemx info port-rights
21543 @itemx info port-sets
21544 @itemx info dead-names
21547 @cindex send rights, @sc{gnu} Hurd
21548 @cindex receive rights, @sc{gnu} Hurd
21549 @cindex port rights, @sc{gnu} Hurd
21550 @cindex port sets, @sc{gnu} Hurd
21551 @cindex dead names, @sc{gnu} Hurd
21552 These commands display information about, respectively, send rights,
21553 receive rights, port rights, port sets, and dead names of a task.
21554 There are also shorthand aliases: @code{info ports} for @code{info
21555 port-rights} and @code{info psets} for @code{info port-sets}.
21557 @item set thread pause
21558 @kindex set thread@r{, Hurd command}
21559 @cindex thread properties, @sc{gnu} Hurd
21560 @cindex pause current thread (@sc{gnu} Hurd)
21561 This command toggles current thread suspension when @value{GDBN} has
21562 control. Setting it to on takes effect immediately, and the current
21563 thread is suspended whenever @value{GDBN} gets control. Setting it to
21564 off will take effect the next time the inferior is continued.
21565 Normally, this command has no effect, since when @value{GDBN} has
21566 control, the whole task is suspended. However, if you used @code{set
21567 task pause off} (see above), this command comes in handy to suspend
21568 only the current thread.
21570 @item show thread pause
21571 @kindex show thread@r{, Hurd command}
21572 This command shows the state of current thread suspension.
21574 @item set thread run
21575 This command sets whether the current thread is allowed to run.
21577 @item show thread run
21578 Show whether the current thread is allowed to run.
21580 @item set thread detach-suspend-count
21581 @cindex thread suspend count, @sc{gnu} Hurd
21582 @cindex detach from thread, @sc{gnu} Hurd
21583 This command sets the suspend count @value{GDBN} will leave on a
21584 thread when detaching. This number is relative to the suspend count
21585 found by @value{GDBN} when it notices the thread; use @code{set thread
21586 takeover-suspend-count} to force it to an absolute value.
21588 @item show thread detach-suspend-count
21589 Show the suspend count @value{GDBN} will leave on the thread when
21592 @item set thread exception-port
21593 @itemx set thread excp
21594 Set the thread exception port to which to forward exceptions. This
21595 overrides the port set by @code{set task exception-port} (see above).
21596 @code{set thread excp} is the shorthand alias.
21598 @item set thread takeover-suspend-count
21599 Normally, @value{GDBN}'s thread suspend counts are relative to the
21600 value @value{GDBN} finds when it notices each thread. This command
21601 changes the suspend counts to be absolute instead.
21603 @item set thread default
21604 @itemx show thread default
21605 @cindex thread default settings, @sc{gnu} Hurd
21606 Each of the above @code{set thread} commands has a @code{set thread
21607 default} counterpart (e.g., @code{set thread default pause}, @code{set
21608 thread default exception-port}, etc.). The @code{thread default}
21609 variety of commands sets the default thread properties for all
21610 threads; you can then change the properties of individual threads with
21611 the non-default commands.
21618 @value{GDBN} provides the following commands specific to the Darwin target:
21621 @item set debug darwin @var{num}
21622 @kindex set debug darwin
21623 When set to a non zero value, enables debugging messages specific to
21624 the Darwin support. Higher values produce more verbose output.
21626 @item show debug darwin
21627 @kindex show debug darwin
21628 Show the current state of Darwin messages.
21630 @item set debug mach-o @var{num}
21631 @kindex set debug mach-o
21632 When set to a non zero value, enables debugging messages while
21633 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
21634 file format used on Darwin for object and executable files.) Higher
21635 values produce more verbose output. This is a command to diagnose
21636 problems internal to @value{GDBN} and should not be needed in normal
21639 @item show debug mach-o
21640 @kindex show debug mach-o
21641 Show the current state of Mach-O file messages.
21643 @item set mach-exceptions on
21644 @itemx set mach-exceptions off
21645 @kindex set mach-exceptions
21646 On Darwin, faults are first reported as a Mach exception and are then
21647 mapped to a Posix signal. Use this command to turn on trapping of
21648 Mach exceptions in the inferior. This might be sometimes useful to
21649 better understand the cause of a fault. The default is off.
21651 @item show mach-exceptions
21652 @kindex show mach-exceptions
21653 Show the current state of exceptions trapping.
21658 @section Embedded Operating Systems
21660 This section describes configurations involving the debugging of
21661 embedded operating systems that are available for several different
21664 @value{GDBN} includes the ability to debug programs running on
21665 various real-time operating systems.
21667 @node Embedded Processors
21668 @section Embedded Processors
21670 This section goes into details specific to particular embedded
21673 @cindex send command to simulator
21674 Whenever a specific embedded processor has a simulator, @value{GDBN}
21675 allows to send an arbitrary command to the simulator.
21678 @item sim @var{command}
21679 @kindex sim@r{, a command}
21680 Send an arbitrary @var{command} string to the simulator. Consult the
21681 documentation for the specific simulator in use for information about
21682 acceptable commands.
21688 * M32R/SDI:: Renesas M32R/SDI
21689 * M68K:: Motorola M68K
21690 * MicroBlaze:: Xilinx MicroBlaze
21691 * MIPS Embedded:: MIPS Embedded
21692 * PowerPC Embedded:: PowerPC Embedded
21695 * Super-H:: Renesas Super-H
21701 @value{GDBN} provides the following ARM-specific commands:
21704 @item set arm disassembler
21706 This commands selects from a list of disassembly styles. The
21707 @code{"std"} style is the standard style.
21709 @item show arm disassembler
21711 Show the current disassembly style.
21713 @item set arm apcs32
21714 @cindex ARM 32-bit mode
21715 This command toggles ARM operation mode between 32-bit and 26-bit.
21717 @item show arm apcs32
21718 Display the current usage of the ARM 32-bit mode.
21720 @item set arm fpu @var{fputype}
21721 This command sets the ARM floating-point unit (FPU) type. The
21722 argument @var{fputype} can be one of these:
21726 Determine the FPU type by querying the OS ABI.
21728 Software FPU, with mixed-endian doubles on little-endian ARM
21731 GCC-compiled FPA co-processor.
21733 Software FPU with pure-endian doubles.
21739 Show the current type of the FPU.
21742 This command forces @value{GDBN} to use the specified ABI.
21745 Show the currently used ABI.
21747 @item set arm fallback-mode (arm|thumb|auto)
21748 @value{GDBN} uses the symbol table, when available, to determine
21749 whether instructions are ARM or Thumb. This command controls
21750 @value{GDBN}'s default behavior when the symbol table is not
21751 available. The default is @samp{auto}, which causes @value{GDBN} to
21752 use the current execution mode (from the @code{T} bit in the @code{CPSR}
21755 @item show arm fallback-mode
21756 Show the current fallback instruction mode.
21758 @item set arm force-mode (arm|thumb|auto)
21759 This command overrides use of the symbol table to determine whether
21760 instructions are ARM or Thumb. The default is @samp{auto}, which
21761 causes @value{GDBN} to use the symbol table and then the setting
21762 of @samp{set arm fallback-mode}.
21764 @item show arm force-mode
21765 Show the current forced instruction mode.
21767 @item set debug arm
21768 Toggle whether to display ARM-specific debugging messages from the ARM
21769 target support subsystem.
21771 @item show debug arm
21772 Show whether ARM-specific debugging messages are enabled.
21776 @item target sim @r{[}@var{simargs}@r{]} @dots{}
21777 The @value{GDBN} ARM simulator accepts the following optional arguments.
21780 @item --swi-support=@var{type}
21781 Tell the simulator which SWI interfaces to support. The argument
21782 @var{type} may be a comma separated list of the following values.
21783 The default value is @code{all}.
21796 @subsection Renesas M32R/SDI
21798 The following commands are available for M32R/SDI:
21803 @cindex reset SDI connection, M32R
21804 This command resets the SDI connection.
21808 This command shows the SDI connection status.
21811 @kindex debug_chaos
21812 @cindex M32R/Chaos debugging
21813 Instructs the remote that M32R/Chaos debugging is to be used.
21815 @item use_debug_dma
21816 @kindex use_debug_dma
21817 Instructs the remote to use the DEBUG_DMA method of accessing memory.
21820 @kindex use_mon_code
21821 Instructs the remote to use the MON_CODE method of accessing memory.
21824 @kindex use_ib_break
21825 Instructs the remote to set breakpoints by IB break.
21827 @item use_dbt_break
21828 @kindex use_dbt_break
21829 Instructs the remote to set breakpoints by DBT.
21835 The Motorola m68k configuration includes ColdFire support.
21838 @subsection MicroBlaze
21839 @cindex Xilinx MicroBlaze
21840 @cindex XMD, Xilinx Microprocessor Debugger
21842 The MicroBlaze is a soft-core processor supported on various Xilinx
21843 FPGAs, such as Spartan or Virtex series. Boards with these processors
21844 usually have JTAG ports which connect to a host system running the Xilinx
21845 Embedded Development Kit (EDK) or Software Development Kit (SDK).
21846 This host system is used to download the configuration bitstream to
21847 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
21848 communicates with the target board using the JTAG interface and
21849 presents a @code{gdbserver} interface to the board. By default
21850 @code{xmd} uses port @code{1234}. (While it is possible to change
21851 this default port, it requires the use of undocumented @code{xmd}
21852 commands. Contact Xilinx support if you need to do this.)
21854 Use these GDB commands to connect to the MicroBlaze target processor.
21857 @item target remote :1234
21858 Use this command to connect to the target if you are running @value{GDBN}
21859 on the same system as @code{xmd}.
21861 @item target remote @var{xmd-host}:1234
21862 Use this command to connect to the target if it is connected to @code{xmd}
21863 running on a different system named @var{xmd-host}.
21866 Use this command to download a program to the MicroBlaze target.
21868 @item set debug microblaze @var{n}
21869 Enable MicroBlaze-specific debugging messages if non-zero.
21871 @item show debug microblaze @var{n}
21872 Show MicroBlaze-specific debugging level.
21875 @node MIPS Embedded
21876 @subsection @acronym{MIPS} Embedded
21878 @cindex @acronym{MIPS} boards
21879 @value{GDBN} can use the @acronym{MIPS} remote debugging protocol to talk to a
21880 @acronym{MIPS} board attached to a serial line. This is available when
21881 you configure @value{GDBN} with @samp{--target=mips-elf}.
21884 Use these @value{GDBN} commands to specify the connection to your target board:
21887 @item target mips @var{port}
21888 @kindex target mips @var{port}
21889 To run a program on the board, start up @code{@value{GDBP}} with the
21890 name of your program as the argument. To connect to the board, use the
21891 command @samp{target mips @var{port}}, where @var{port} is the name of
21892 the serial port connected to the board. If the program has not already
21893 been downloaded to the board, you may use the @code{load} command to
21894 download it. You can then use all the usual @value{GDBN} commands.
21896 For example, this sequence connects to the target board through a serial
21897 port, and loads and runs a program called @var{prog} through the
21901 host$ @value{GDBP} @var{prog}
21902 @value{GDBN} is free software and @dots{}
21903 (@value{GDBP}) target mips /dev/ttyb
21904 (@value{GDBP}) load @var{prog}
21908 @item target mips @var{hostname}:@var{portnumber}
21909 On some @value{GDBN} host configurations, you can specify a TCP
21910 connection (for instance, to a serial line managed by a terminal
21911 concentrator) instead of a serial port, using the syntax
21912 @samp{@var{hostname}:@var{portnumber}}.
21914 @item target pmon @var{port}
21915 @kindex target pmon @var{port}
21918 @item target ddb @var{port}
21919 @kindex target ddb @var{port}
21920 NEC's DDB variant of PMON for Vr4300.
21922 @item target lsi @var{port}
21923 @kindex target lsi @var{port}
21924 LSI variant of PMON.
21930 @value{GDBN} also supports these special commands for @acronym{MIPS} targets:
21933 @item set mipsfpu double
21934 @itemx set mipsfpu single
21935 @itemx set mipsfpu none
21936 @itemx set mipsfpu auto
21937 @itemx show mipsfpu
21938 @kindex set mipsfpu
21939 @kindex show mipsfpu
21940 @cindex @acronym{MIPS} remote floating point
21941 @cindex floating point, @acronym{MIPS} remote
21942 If your target board does not support the @acronym{MIPS} floating point
21943 coprocessor, you should use the command @samp{set mipsfpu none} (if you
21944 need this, you may wish to put the command in your @value{GDBN} init
21945 file). This tells @value{GDBN} how to find the return value of
21946 functions which return floating point values. It also allows
21947 @value{GDBN} to avoid saving the floating point registers when calling
21948 functions on the board. If you are using a floating point coprocessor
21949 with only single precision floating point support, as on the @sc{r4650}
21950 processor, use the command @samp{set mipsfpu single}. The default
21951 double precision floating point coprocessor may be selected using
21952 @samp{set mipsfpu double}.
21954 In previous versions the only choices were double precision or no
21955 floating point, so @samp{set mipsfpu on} will select double precision
21956 and @samp{set mipsfpu off} will select no floating point.
21958 As usual, you can inquire about the @code{mipsfpu} variable with
21959 @samp{show mipsfpu}.
21961 @item set timeout @var{seconds}
21962 @itemx set retransmit-timeout @var{seconds}
21963 @itemx show timeout
21964 @itemx show retransmit-timeout
21965 @cindex @code{timeout}, @acronym{MIPS} protocol
21966 @cindex @code{retransmit-timeout}, @acronym{MIPS} protocol
21967 @kindex set timeout
21968 @kindex show timeout
21969 @kindex set retransmit-timeout
21970 @kindex show retransmit-timeout
21971 You can control the timeout used while waiting for a packet, in the @acronym{MIPS}
21972 remote protocol, with the @code{set timeout @var{seconds}} command. The
21973 default is 5 seconds. Similarly, you can control the timeout used while
21974 waiting for an acknowledgment of a packet with the @code{set
21975 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
21976 You can inspect both values with @code{show timeout} and @code{show
21977 retransmit-timeout}. (These commands are @emph{only} available when
21978 @value{GDBN} is configured for @samp{--target=mips-elf}.)
21980 The timeout set by @code{set timeout} does not apply when @value{GDBN}
21981 is waiting for your program to stop. In that case, @value{GDBN} waits
21982 forever because it has no way of knowing how long the program is going
21983 to run before stopping.
21985 @item set syn-garbage-limit @var{num}
21986 @kindex set syn-garbage-limit@r{, @acronym{MIPS} remote}
21987 @cindex synchronize with remote @acronym{MIPS} target
21988 Limit the maximum number of characters @value{GDBN} should ignore when
21989 it tries to synchronize with the remote target. The default is 10
21990 characters. Setting the limit to -1 means there's no limit.
21992 @item show syn-garbage-limit
21993 @kindex show syn-garbage-limit@r{, @acronym{MIPS} remote}
21994 Show the current limit on the number of characters to ignore when
21995 trying to synchronize with the remote system.
21997 @item set monitor-prompt @var{prompt}
21998 @kindex set monitor-prompt@r{, @acronym{MIPS} remote}
21999 @cindex remote monitor prompt
22000 Tell @value{GDBN} to expect the specified @var{prompt} string from the
22001 remote monitor. The default depends on the target:
22011 @item show monitor-prompt
22012 @kindex show monitor-prompt@r{, @acronym{MIPS} remote}
22013 Show the current strings @value{GDBN} expects as the prompt from the
22016 @item set monitor-warnings
22017 @kindex set monitor-warnings@r{, @acronym{MIPS} remote}
22018 Enable or disable monitor warnings about hardware breakpoints. This
22019 has effect only for the @code{lsi} target. When on, @value{GDBN} will
22020 display warning messages whose codes are returned by the @code{lsi}
22021 PMON monitor for breakpoint commands.
22023 @item show monitor-warnings
22024 @kindex show monitor-warnings@r{, @acronym{MIPS} remote}
22025 Show the current setting of printing monitor warnings.
22027 @item pmon @var{command}
22028 @kindex pmon@r{, @acronym{MIPS} remote}
22029 @cindex send PMON command
22030 This command allows sending an arbitrary @var{command} string to the
22031 monitor. The monitor must be in debug mode for this to work.
22034 @node PowerPC Embedded
22035 @subsection PowerPC Embedded
22037 @cindex DVC register
22038 @value{GDBN} supports using the DVC (Data Value Compare) register to
22039 implement in hardware simple hardware watchpoint conditions of the form:
22042 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
22043 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
22046 The DVC register will be automatically used when @value{GDBN} detects
22047 such pattern in a condition expression, and the created watchpoint uses one
22048 debug register (either the @code{exact-watchpoints} option is on and the
22049 variable is scalar, or the variable has a length of one byte). This feature
22050 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
22053 When running on PowerPC embedded processors, @value{GDBN} automatically uses
22054 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
22055 in which case watchpoints using only one debug register are created when
22056 watching variables of scalar types.
22058 You can create an artificial array to watch an arbitrary memory
22059 region using one of the following commands (@pxref{Expressions}):
22062 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
22063 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
22066 PowerPC embedded processors support masked watchpoints. See the discussion
22067 about the @code{mask} argument in @ref{Set Watchpoints}.
22069 @cindex ranged breakpoint
22070 PowerPC embedded processors support hardware accelerated
22071 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
22072 the inferior whenever it executes an instruction at any address within
22073 the range it specifies. To set a ranged breakpoint in @value{GDBN},
22074 use the @code{break-range} command.
22076 @value{GDBN} provides the following PowerPC-specific commands:
22079 @kindex break-range
22080 @item break-range @var{start-location}, @var{end-location}
22081 Set a breakpoint for an address range given by
22082 @var{start-location} and @var{end-location}, which can specify a function name,
22083 a line number, an offset of lines from the current line or from the start
22084 location, or an address of an instruction (see @ref{Specify Location},
22085 for a list of all the possible ways to specify a @var{location}.)
22086 The breakpoint will stop execution of the inferior whenever it
22087 executes an instruction at any address within the specified range,
22088 (including @var{start-location} and @var{end-location}.)
22090 @kindex set powerpc
22091 @item set powerpc soft-float
22092 @itemx show powerpc soft-float
22093 Force @value{GDBN} to use (or not use) a software floating point calling
22094 convention. By default, @value{GDBN} selects the calling convention based
22095 on the selected architecture and the provided executable file.
22097 @item set powerpc vector-abi
22098 @itemx show powerpc vector-abi
22099 Force @value{GDBN} to use the specified calling convention for vector
22100 arguments and return values. The valid options are @samp{auto};
22101 @samp{generic}, to avoid vector registers even if they are present;
22102 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
22103 registers. By default, @value{GDBN} selects the calling convention
22104 based on the selected architecture and the provided executable file.
22106 @item set powerpc exact-watchpoints
22107 @itemx show powerpc exact-watchpoints
22108 Allow @value{GDBN} to use only one debug register when watching a variable
22109 of scalar type, thus assuming that the variable is accessed through the
22110 address of its first byte.
22115 @subsection Atmel AVR
22118 When configured for debugging the Atmel AVR, @value{GDBN} supports the
22119 following AVR-specific commands:
22122 @item info io_registers
22123 @kindex info io_registers@r{, AVR}
22124 @cindex I/O registers (Atmel AVR)
22125 This command displays information about the AVR I/O registers. For
22126 each register, @value{GDBN} prints its number and value.
22133 When configured for debugging CRIS, @value{GDBN} provides the
22134 following CRIS-specific commands:
22137 @item set cris-version @var{ver}
22138 @cindex CRIS version
22139 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
22140 The CRIS version affects register names and sizes. This command is useful in
22141 case autodetection of the CRIS version fails.
22143 @item show cris-version
22144 Show the current CRIS version.
22146 @item set cris-dwarf2-cfi
22147 @cindex DWARF-2 CFI and CRIS
22148 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
22149 Change to @samp{off} when using @code{gcc-cris} whose version is below
22152 @item show cris-dwarf2-cfi
22153 Show the current state of using DWARF-2 CFI.
22155 @item set cris-mode @var{mode}
22157 Set the current CRIS mode to @var{mode}. It should only be changed when
22158 debugging in guru mode, in which case it should be set to
22159 @samp{guru} (the default is @samp{normal}).
22161 @item show cris-mode
22162 Show the current CRIS mode.
22166 @subsection Renesas Super-H
22169 For the Renesas Super-H processor, @value{GDBN} provides these
22173 @item set sh calling-convention @var{convention}
22174 @kindex set sh calling-convention
22175 Set the calling-convention used when calling functions from @value{GDBN}.
22176 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
22177 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
22178 convention. If the DWARF-2 information of the called function specifies
22179 that the function follows the Renesas calling convention, the function
22180 is called using the Renesas calling convention. If the calling convention
22181 is set to @samp{renesas}, the Renesas calling convention is always used,
22182 regardless of the DWARF-2 information. This can be used to override the
22183 default of @samp{gcc} if debug information is missing, or the compiler
22184 does not emit the DWARF-2 calling convention entry for a function.
22186 @item show sh calling-convention
22187 @kindex show sh calling-convention
22188 Show the current calling convention setting.
22193 @node Architectures
22194 @section Architectures
22196 This section describes characteristics of architectures that affect
22197 all uses of @value{GDBN} with the architecture, both native and cross.
22204 * HPPA:: HP PA architecture
22205 * SPU:: Cell Broadband Engine SPU architecture
22211 @subsection AArch64
22212 @cindex AArch64 support
22214 When @value{GDBN} is debugging the AArch64 architecture, it provides the
22215 following special commands:
22218 @item set debug aarch64
22219 @kindex set debug aarch64
22220 This command determines whether AArch64 architecture-specific debugging
22221 messages are to be displayed.
22223 @item show debug aarch64
22224 Show whether AArch64 debugging messages are displayed.
22229 @subsection x86 Architecture-specific Issues
22232 @item set struct-convention @var{mode}
22233 @kindex set struct-convention
22234 @cindex struct return convention
22235 @cindex struct/union returned in registers
22236 Set the convention used by the inferior to return @code{struct}s and
22237 @code{union}s from functions to @var{mode}. Possible values of
22238 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
22239 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
22240 are returned on the stack, while @code{"reg"} means that a
22241 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
22242 be returned in a register.
22244 @item show struct-convention
22245 @kindex show struct-convention
22246 Show the current setting of the convention to return @code{struct}s
22251 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
22252 @cindex Intel Memory Protection Extensions (MPX).
22254 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
22255 @footnote{The register named with capital letters represent the architecture
22256 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
22257 which are the lower bound and upper bound. Bounds are effective addresses or
22258 memory locations. The upper bounds are architecturally represented in 1's
22259 complement form. A bound having lower bound = 0, and upper bound = 0
22260 (1's complement of all bits set) will allow access to the entire address space.
22262 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
22263 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
22264 display the upper bound performing the complement of one operation on the
22265 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
22266 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
22267 can also be noted that the upper bounds are inclusive.
22269 As an example, assume that the register BND0 holds bounds for a pointer having
22270 access allowed for the range between 0x32 and 0x71. The values present on
22271 bnd0raw and bnd registers are presented as follows:
22274 bnd0raw = @{0x32, 0xffffffff8e@}
22275 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
22278 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
22279 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
22280 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
22281 Python, the display includes the memory size, in bits, accessible to
22284 Bounds can also be stored in bounds tables, which are stored in
22285 application memory. These tables store bounds for pointers by specifying
22286 the bounds pointer's value along with its bounds. Evaluating and changing
22287 bounds located in bound tables is therefore interesting while investigating
22288 bugs on MPX context. @value{GDBN} provides commands for this purpose:
22291 @item show mpx bound @var{pointer}
22292 @kindex show mpx bound
22293 Display bounds of the given @var{pointer}.
22295 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
22296 @kindex set mpx bound
22297 Set the bounds of a pointer in the bound table.
22298 This command takes three parameters: @var{pointer} is the pointers
22299 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
22300 for lower and upper bounds respectively.
22306 See the following section.
22309 @subsection @acronym{MIPS}
22311 @cindex stack on Alpha
22312 @cindex stack on @acronym{MIPS}
22313 @cindex Alpha stack
22314 @cindex @acronym{MIPS} stack
22315 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
22316 sometimes requires @value{GDBN} to search backward in the object code to
22317 find the beginning of a function.
22319 @cindex response time, @acronym{MIPS} debugging
22320 To improve response time (especially for embedded applications, where
22321 @value{GDBN} may be restricted to a slow serial line for this search)
22322 you may want to limit the size of this search, using one of these
22326 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
22327 @item set heuristic-fence-post @var{limit}
22328 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
22329 search for the beginning of a function. A value of @var{0} (the
22330 default) means there is no limit. However, except for @var{0}, the
22331 larger the limit the more bytes @code{heuristic-fence-post} must search
22332 and therefore the longer it takes to run. You should only need to use
22333 this command when debugging a stripped executable.
22335 @item show heuristic-fence-post
22336 Display the current limit.
22340 These commands are available @emph{only} when @value{GDBN} is configured
22341 for debugging programs on Alpha or @acronym{MIPS} processors.
22343 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
22347 @item set mips abi @var{arg}
22348 @kindex set mips abi
22349 @cindex set ABI for @acronym{MIPS}
22350 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
22351 values of @var{arg} are:
22355 The default ABI associated with the current binary (this is the
22365 @item show mips abi
22366 @kindex show mips abi
22367 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
22369 @item set mips compression @var{arg}
22370 @kindex set mips compression
22371 @cindex code compression, @acronym{MIPS}
22372 Tell @value{GDBN} which @acronym{MIPS} compressed
22373 @acronym{ISA, Instruction Set Architecture} encoding is used by the
22374 inferior. @value{GDBN} uses this for code disassembly and other
22375 internal interpretation purposes. This setting is only referred to
22376 when no executable has been associated with the debugging session or
22377 the executable does not provide information about the encoding it uses.
22378 Otherwise this setting is automatically updated from information
22379 provided by the executable.
22381 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
22382 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
22383 executables containing @acronym{MIPS16} code frequently are not
22384 identified as such.
22386 This setting is ``sticky''; that is, it retains its value across
22387 debugging sessions until reset either explicitly with this command or
22388 implicitly from an executable.
22390 The compiler and/or assembler typically add symbol table annotations to
22391 identify functions compiled for the @acronym{MIPS16} or
22392 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
22393 are present, @value{GDBN} uses them in preference to the global
22394 compressed @acronym{ISA} encoding setting.
22396 @item show mips compression
22397 @kindex show mips compression
22398 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
22399 @value{GDBN} to debug the inferior.
22402 @itemx show mipsfpu
22403 @xref{MIPS Embedded, set mipsfpu}.
22405 @item set mips mask-address @var{arg}
22406 @kindex set mips mask-address
22407 @cindex @acronym{MIPS} addresses, masking
22408 This command determines whether the most-significant 32 bits of 64-bit
22409 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
22410 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
22411 setting, which lets @value{GDBN} determine the correct value.
22413 @item show mips mask-address
22414 @kindex show mips mask-address
22415 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
22418 @item set remote-mips64-transfers-32bit-regs
22419 @kindex set remote-mips64-transfers-32bit-regs
22420 This command controls compatibility with 64-bit @acronym{MIPS} targets that
22421 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
22422 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
22423 and 64 bits for other registers, set this option to @samp{on}.
22425 @item show remote-mips64-transfers-32bit-regs
22426 @kindex show remote-mips64-transfers-32bit-regs
22427 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
22429 @item set debug mips
22430 @kindex set debug mips
22431 This command turns on and off debugging messages for the @acronym{MIPS}-specific
22432 target code in @value{GDBN}.
22434 @item show debug mips
22435 @kindex show debug mips
22436 Show the current setting of @acronym{MIPS} debugging messages.
22442 @cindex HPPA support
22444 When @value{GDBN} is debugging the HP PA architecture, it provides the
22445 following special commands:
22448 @item set debug hppa
22449 @kindex set debug hppa
22450 This command determines whether HPPA architecture-specific debugging
22451 messages are to be displayed.
22453 @item show debug hppa
22454 Show whether HPPA debugging messages are displayed.
22456 @item maint print unwind @var{address}
22457 @kindex maint print unwind@r{, HPPA}
22458 This command displays the contents of the unwind table entry at the
22459 given @var{address}.
22465 @subsection Cell Broadband Engine SPU architecture
22466 @cindex Cell Broadband Engine
22469 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
22470 it provides the following special commands:
22473 @item info spu event
22475 Display SPU event facility status. Shows current event mask
22476 and pending event status.
22478 @item info spu signal
22479 Display SPU signal notification facility status. Shows pending
22480 signal-control word and signal notification mode of both signal
22481 notification channels.
22483 @item info spu mailbox
22484 Display SPU mailbox facility status. Shows all pending entries,
22485 in order of processing, in each of the SPU Write Outbound,
22486 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
22489 Display MFC DMA status. Shows all pending commands in the MFC
22490 DMA queue. For each entry, opcode, tag, class IDs, effective
22491 and local store addresses and transfer size are shown.
22493 @item info spu proxydma
22494 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
22495 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
22496 and local store addresses and transfer size are shown.
22500 When @value{GDBN} is debugging a combined PowerPC/SPU application
22501 on the Cell Broadband Engine, it provides in addition the following
22505 @item set spu stop-on-load @var{arg}
22507 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
22508 will give control to the user when a new SPE thread enters its @code{main}
22509 function. The default is @code{off}.
22511 @item show spu stop-on-load
22513 Show whether to stop for new SPE threads.
22515 @item set spu auto-flush-cache @var{arg}
22516 Set whether to automatically flush the software-managed cache. When set to
22517 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
22518 cache to be flushed whenever SPE execution stops. This provides a consistent
22519 view of PowerPC memory that is accessed via the cache. If an application
22520 does not use the software-managed cache, this option has no effect.
22522 @item show spu auto-flush-cache
22523 Show whether to automatically flush the software-managed cache.
22528 @subsection PowerPC
22529 @cindex PowerPC architecture
22531 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
22532 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
22533 numbers stored in the floating point registers. These values must be stored
22534 in two consecutive registers, always starting at an even register like
22535 @code{f0} or @code{f2}.
22537 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
22538 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
22539 @code{f2} and @code{f3} for @code{$dl1} and so on.
22541 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
22542 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
22545 @subsection Nios II
22546 @cindex Nios II architecture
22548 When @value{GDBN} is debugging the Nios II architecture,
22549 it provides the following special commands:
22553 @item set debug nios2
22554 @kindex set debug nios2
22555 This command turns on and off debugging messages for the Nios II
22556 target code in @value{GDBN}.
22558 @item show debug nios2
22559 @kindex show debug nios2
22560 Show the current setting of Nios II debugging messages.
22563 @node Controlling GDB
22564 @chapter Controlling @value{GDBN}
22566 You can alter the way @value{GDBN} interacts with you by using the
22567 @code{set} command. For commands controlling how @value{GDBN} displays
22568 data, see @ref{Print Settings, ,Print Settings}. Other settings are
22573 * Editing:: Command editing
22574 * Command History:: Command history
22575 * Screen Size:: Screen size
22576 * Numbers:: Numbers
22577 * ABI:: Configuring the current ABI
22578 * Auto-loading:: Automatically loading associated files
22579 * Messages/Warnings:: Optional warnings and messages
22580 * Debugging Output:: Optional messages about internal happenings
22581 * Other Misc Settings:: Other Miscellaneous Settings
22589 @value{GDBN} indicates its readiness to read a command by printing a string
22590 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
22591 can change the prompt string with the @code{set prompt} command. For
22592 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
22593 the prompt in one of the @value{GDBN} sessions so that you can always tell
22594 which one you are talking to.
22596 @emph{Note:} @code{set prompt} does not add a space for you after the
22597 prompt you set. This allows you to set a prompt which ends in a space
22598 or a prompt that does not.
22602 @item set prompt @var{newprompt}
22603 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
22605 @kindex show prompt
22607 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
22610 Versions of @value{GDBN} that ship with Python scripting enabled have
22611 prompt extensions. The commands for interacting with these extensions
22615 @kindex set extended-prompt
22616 @item set extended-prompt @var{prompt}
22617 Set an extended prompt that allows for substitutions.
22618 @xref{gdb.prompt}, for a list of escape sequences that can be used for
22619 substitution. Any escape sequences specified as part of the prompt
22620 string are replaced with the corresponding strings each time the prompt
22626 set extended-prompt Current working directory: \w (gdb)
22629 Note that when an extended-prompt is set, it takes control of the
22630 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
22632 @kindex show extended-prompt
22633 @item show extended-prompt
22634 Prints the extended prompt. Any escape sequences specified as part of
22635 the prompt string with @code{set extended-prompt}, are replaced with the
22636 corresponding strings each time the prompt is displayed.
22640 @section Command Editing
22642 @cindex command line editing
22644 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
22645 @sc{gnu} library provides consistent behavior for programs which provide a
22646 command line interface to the user. Advantages are @sc{gnu} Emacs-style
22647 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
22648 substitution, and a storage and recall of command history across
22649 debugging sessions.
22651 You may control the behavior of command line editing in @value{GDBN} with the
22652 command @code{set}.
22655 @kindex set editing
22658 @itemx set editing on
22659 Enable command line editing (enabled by default).
22661 @item set editing off
22662 Disable command line editing.
22664 @kindex show editing
22666 Show whether command line editing is enabled.
22669 @ifset SYSTEM_READLINE
22670 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
22672 @ifclear SYSTEM_READLINE
22673 @xref{Command Line Editing},
22675 for more details about the Readline
22676 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
22677 encouraged to read that chapter.
22679 @node Command History
22680 @section Command History
22681 @cindex command history
22683 @value{GDBN} can keep track of the commands you type during your
22684 debugging sessions, so that you can be certain of precisely what
22685 happened. Use these commands to manage the @value{GDBN} command
22688 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
22689 package, to provide the history facility.
22690 @ifset SYSTEM_READLINE
22691 @xref{Using History Interactively, , , history, GNU History Library},
22693 @ifclear SYSTEM_READLINE
22694 @xref{Using History Interactively},
22696 for the detailed description of the History library.
22698 To issue a command to @value{GDBN} without affecting certain aspects of
22699 the state which is seen by users, prefix it with @samp{server }
22700 (@pxref{Server Prefix}). This
22701 means that this command will not affect the command history, nor will it
22702 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
22703 pressed on a line by itself.
22705 @cindex @code{server}, command prefix
22706 The server prefix does not affect the recording of values into the value
22707 history; to print a value without recording it into the value history,
22708 use the @code{output} command instead of the @code{print} command.
22710 Here is the description of @value{GDBN} commands related to command
22714 @cindex history substitution
22715 @cindex history file
22716 @kindex set history filename
22717 @cindex @env{GDBHISTFILE}, environment variable
22718 @item set history filename @var{fname}
22719 Set the name of the @value{GDBN} command history file to @var{fname}.
22720 This is the file where @value{GDBN} reads an initial command history
22721 list, and where it writes the command history from this session when it
22722 exits. You can access this list through history expansion or through
22723 the history command editing characters listed below. This file defaults
22724 to the value of the environment variable @code{GDBHISTFILE}, or to
22725 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
22728 @cindex save command history
22729 @kindex set history save
22730 @item set history save
22731 @itemx set history save on
22732 Record command history in a file, whose name may be specified with the
22733 @code{set history filename} command. By default, this option is disabled.
22735 @item set history save off
22736 Stop recording command history in a file.
22738 @cindex history size
22739 @kindex set history size
22740 @cindex @env{GDBHISTSIZE}, environment variable
22741 @item set history size @var{size}
22742 @itemx set history size unlimited
22743 Set the number of commands which @value{GDBN} keeps in its history list.
22744 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
22745 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
22746 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
22747 either a negative number or the empty string, then the number of commands
22748 @value{GDBN} keeps in the history list is unlimited.
22750 @cindex remove duplicate history
22751 @kindex set history remove-duplicates
22752 @item set history remove-duplicates @var{count}
22753 @itemx set history remove-duplicates unlimited
22754 Control the removal of duplicate history entries in the command history list.
22755 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
22756 history entries and remove the first entry that is a duplicate of the current
22757 entry being added to the command history list. If @var{count} is
22758 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
22759 removal of duplicate history entries is disabled.
22761 Only history entries added during the current session are considered for
22762 removal. This option is set to 0 by default.
22766 History expansion assigns special meaning to the character @kbd{!}.
22767 @ifset SYSTEM_READLINE
22768 @xref{Event Designators, , , history, GNU History Library},
22770 @ifclear SYSTEM_READLINE
22771 @xref{Event Designators},
22775 @cindex history expansion, turn on/off
22776 Since @kbd{!} is also the logical not operator in C, history expansion
22777 is off by default. If you decide to enable history expansion with the
22778 @code{set history expansion on} command, you may sometimes need to
22779 follow @kbd{!} (when it is used as logical not, in an expression) with
22780 a space or a tab to prevent it from being expanded. The readline
22781 history facilities do not attempt substitution on the strings
22782 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
22784 The commands to control history expansion are:
22787 @item set history expansion on
22788 @itemx set history expansion
22789 @kindex set history expansion
22790 Enable history expansion. History expansion is off by default.
22792 @item set history expansion off
22793 Disable history expansion.
22796 @kindex show history
22798 @itemx show history filename
22799 @itemx show history save
22800 @itemx show history size
22801 @itemx show history expansion
22802 These commands display the state of the @value{GDBN} history parameters.
22803 @code{show history} by itself displays all four states.
22808 @kindex show commands
22809 @cindex show last commands
22810 @cindex display command history
22811 @item show commands
22812 Display the last ten commands in the command history.
22814 @item show commands @var{n}
22815 Print ten commands centered on command number @var{n}.
22817 @item show commands +
22818 Print ten commands just after the commands last printed.
22822 @section Screen Size
22823 @cindex size of screen
22824 @cindex screen size
22827 @cindex pauses in output
22829 Certain commands to @value{GDBN} may produce large amounts of
22830 information output to the screen. To help you read all of it,
22831 @value{GDBN} pauses and asks you for input at the end of each page of
22832 output. Type @key{RET} when you want to continue the output, or @kbd{q}
22833 to discard the remaining output. Also, the screen width setting
22834 determines when to wrap lines of output. Depending on what is being
22835 printed, @value{GDBN} tries to break the line at a readable place,
22836 rather than simply letting it overflow onto the following line.
22838 Normally @value{GDBN} knows the size of the screen from the terminal
22839 driver software. For example, on Unix @value{GDBN} uses the termcap data base
22840 together with the value of the @code{TERM} environment variable and the
22841 @code{stty rows} and @code{stty cols} settings. If this is not correct,
22842 you can override it with the @code{set height} and @code{set
22849 @kindex show height
22850 @item set height @var{lpp}
22851 @itemx set height unlimited
22853 @itemx set width @var{cpl}
22854 @itemx set width unlimited
22856 These @code{set} commands specify a screen height of @var{lpp} lines and
22857 a screen width of @var{cpl} characters. The associated @code{show}
22858 commands display the current settings.
22860 If you specify a height of either @code{unlimited} or zero lines,
22861 @value{GDBN} does not pause during output no matter how long the
22862 output is. This is useful if output is to a file or to an editor
22865 Likewise, you can specify @samp{set width unlimited} or @samp{set
22866 width 0} to prevent @value{GDBN} from wrapping its output.
22868 @item set pagination on
22869 @itemx set pagination off
22870 @kindex set pagination
22871 Turn the output pagination on or off; the default is on. Turning
22872 pagination off is the alternative to @code{set height unlimited}. Note that
22873 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
22874 Options, -batch}) also automatically disables pagination.
22876 @item show pagination
22877 @kindex show pagination
22878 Show the current pagination mode.
22883 @cindex number representation
22884 @cindex entering numbers
22886 You can always enter numbers in octal, decimal, or hexadecimal in
22887 @value{GDBN} by the usual conventions: octal numbers begin with
22888 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
22889 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
22890 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
22891 10; likewise, the default display for numbers---when no particular
22892 format is specified---is base 10. You can change the default base for
22893 both input and output with the commands described below.
22896 @kindex set input-radix
22897 @item set input-radix @var{base}
22898 Set the default base for numeric input. Supported choices
22899 for @var{base} are decimal 8, 10, or 16. The base must itself be
22900 specified either unambiguously or using the current input radix; for
22904 set input-radix 012
22905 set input-radix 10.
22906 set input-radix 0xa
22910 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
22911 leaves the input radix unchanged, no matter what it was, since
22912 @samp{10}, being without any leading or trailing signs of its base, is
22913 interpreted in the current radix. Thus, if the current radix is 16,
22914 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
22917 @kindex set output-radix
22918 @item set output-radix @var{base}
22919 Set the default base for numeric display. Supported choices
22920 for @var{base} are decimal 8, 10, or 16. The base must itself be
22921 specified either unambiguously or using the current input radix.
22923 @kindex show input-radix
22924 @item show input-radix
22925 Display the current default base for numeric input.
22927 @kindex show output-radix
22928 @item show output-radix
22929 Display the current default base for numeric display.
22931 @item set radix @r{[}@var{base}@r{]}
22935 These commands set and show the default base for both input and output
22936 of numbers. @code{set radix} sets the radix of input and output to
22937 the same base; without an argument, it resets the radix back to its
22938 default value of 10.
22943 @section Configuring the Current ABI
22945 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
22946 application automatically. However, sometimes you need to override its
22947 conclusions. Use these commands to manage @value{GDBN}'s view of the
22953 @cindex Newlib OS ABI and its influence on the longjmp handling
22955 One @value{GDBN} configuration can debug binaries for multiple operating
22956 system targets, either via remote debugging or native emulation.
22957 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
22958 but you can override its conclusion using the @code{set osabi} command.
22959 One example where this is useful is in debugging of binaries which use
22960 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
22961 not have the same identifying marks that the standard C library for your
22964 When @value{GDBN} is debugging the AArch64 architecture, it provides a
22965 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
22966 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
22967 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
22971 Show the OS ABI currently in use.
22974 With no argument, show the list of registered available OS ABI's.
22976 @item set osabi @var{abi}
22977 Set the current OS ABI to @var{abi}.
22980 @cindex float promotion
22982 Generally, the way that an argument of type @code{float} is passed to a
22983 function depends on whether the function is prototyped. For a prototyped
22984 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
22985 according to the architecture's convention for @code{float}. For unprototyped
22986 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
22987 @code{double} and then passed.
22989 Unfortunately, some forms of debug information do not reliably indicate whether
22990 a function is prototyped. If @value{GDBN} calls a function that is not marked
22991 as prototyped, it consults @kbd{set coerce-float-to-double}.
22994 @kindex set coerce-float-to-double
22995 @item set coerce-float-to-double
22996 @itemx set coerce-float-to-double on
22997 Arguments of type @code{float} will be promoted to @code{double} when passed
22998 to an unprototyped function. This is the default setting.
23000 @item set coerce-float-to-double off
23001 Arguments of type @code{float} will be passed directly to unprototyped
23004 @kindex show coerce-float-to-double
23005 @item show coerce-float-to-double
23006 Show the current setting of promoting @code{float} to @code{double}.
23010 @kindex show cp-abi
23011 @value{GDBN} needs to know the ABI used for your program's C@t{++}
23012 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
23013 used to build your application. @value{GDBN} only fully supports
23014 programs with a single C@t{++} ABI; if your program contains code using
23015 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
23016 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
23017 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
23018 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
23019 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
23020 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
23025 Show the C@t{++} ABI currently in use.
23028 With no argument, show the list of supported C@t{++} ABI's.
23030 @item set cp-abi @var{abi}
23031 @itemx set cp-abi auto
23032 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
23036 @section Automatically loading associated files
23037 @cindex auto-loading
23039 @value{GDBN} sometimes reads files with commands and settings automatically,
23040 without being explicitly told so by the user. We call this feature
23041 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
23042 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
23043 results or introduce security risks (e.g., if the file comes from untrusted
23047 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
23048 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
23050 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
23051 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
23054 There are various kinds of files @value{GDBN} can automatically load.
23055 In addition to these files, @value{GDBN} supports auto-loading code written
23056 in various extension languages. @xref{Auto-loading extensions}.
23058 Note that loading of these associated files (including the local @file{.gdbinit}
23059 file) requires accordingly configured @code{auto-load safe-path}
23060 (@pxref{Auto-loading safe path}).
23062 For these reasons, @value{GDBN} includes commands and options to let you
23063 control when to auto-load files and which files should be auto-loaded.
23066 @anchor{set auto-load off}
23067 @kindex set auto-load off
23068 @item set auto-load off
23069 Globally disable loading of all auto-loaded files.
23070 You may want to use this command with the @samp{-iex} option
23071 (@pxref{Option -init-eval-command}) such as:
23073 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
23076 Be aware that system init file (@pxref{System-wide configuration})
23077 and init files from your home directory (@pxref{Home Directory Init File})
23078 still get read (as they come from generally trusted directories).
23079 To prevent @value{GDBN} from auto-loading even those init files, use the
23080 @option{-nx} option (@pxref{Mode Options}), in addition to
23081 @code{set auto-load no}.
23083 @anchor{show auto-load}
23084 @kindex show auto-load
23085 @item show auto-load
23086 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
23090 (gdb) show auto-load
23091 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
23092 libthread-db: Auto-loading of inferior specific libthread_db is on.
23093 local-gdbinit: Auto-loading of .gdbinit script from current directory
23095 python-scripts: Auto-loading of Python scripts is on.
23096 safe-path: List of directories from which it is safe to auto-load files
23097 is $debugdir:$datadir/auto-load.
23098 scripts-directory: List of directories from which to load auto-loaded scripts
23099 is $debugdir:$datadir/auto-load.
23102 @anchor{info auto-load}
23103 @kindex info auto-load
23104 @item info auto-load
23105 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
23109 (gdb) info auto-load
23112 Yes /home/user/gdb/gdb-gdb.gdb
23113 libthread-db: No auto-loaded libthread-db.
23114 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
23118 Yes /home/user/gdb/gdb-gdb.py
23122 These are @value{GDBN} control commands for the auto-loading:
23124 @multitable @columnfractions .5 .5
23125 @item @xref{set auto-load off}.
23126 @tab Disable auto-loading globally.
23127 @item @xref{show auto-load}.
23128 @tab Show setting of all kinds of files.
23129 @item @xref{info auto-load}.
23130 @tab Show state of all kinds of files.
23131 @item @xref{set auto-load gdb-scripts}.
23132 @tab Control for @value{GDBN} command scripts.
23133 @item @xref{show auto-load gdb-scripts}.
23134 @tab Show setting of @value{GDBN} command scripts.
23135 @item @xref{info auto-load gdb-scripts}.
23136 @tab Show state of @value{GDBN} command scripts.
23137 @item @xref{set auto-load python-scripts}.
23138 @tab Control for @value{GDBN} Python scripts.
23139 @item @xref{show auto-load python-scripts}.
23140 @tab Show setting of @value{GDBN} Python scripts.
23141 @item @xref{info auto-load python-scripts}.
23142 @tab Show state of @value{GDBN} Python scripts.
23143 @item @xref{set auto-load guile-scripts}.
23144 @tab Control for @value{GDBN} Guile scripts.
23145 @item @xref{show auto-load guile-scripts}.
23146 @tab Show setting of @value{GDBN} Guile scripts.
23147 @item @xref{info auto-load guile-scripts}.
23148 @tab Show state of @value{GDBN} Guile scripts.
23149 @item @xref{set auto-load scripts-directory}.
23150 @tab Control for @value{GDBN} auto-loaded scripts location.
23151 @item @xref{show auto-load scripts-directory}.
23152 @tab Show @value{GDBN} auto-loaded scripts location.
23153 @item @xref{add-auto-load-scripts-directory}.
23154 @tab Add directory for auto-loaded scripts location list.
23155 @item @xref{set auto-load local-gdbinit}.
23156 @tab Control for init file in the current directory.
23157 @item @xref{show auto-load local-gdbinit}.
23158 @tab Show setting of init file in the current directory.
23159 @item @xref{info auto-load local-gdbinit}.
23160 @tab Show state of init file in the current directory.
23161 @item @xref{set auto-load libthread-db}.
23162 @tab Control for thread debugging library.
23163 @item @xref{show auto-load libthread-db}.
23164 @tab Show setting of thread debugging library.
23165 @item @xref{info auto-load libthread-db}.
23166 @tab Show state of thread debugging library.
23167 @item @xref{set auto-load safe-path}.
23168 @tab Control directories trusted for automatic loading.
23169 @item @xref{show auto-load safe-path}.
23170 @tab Show directories trusted for automatic loading.
23171 @item @xref{add-auto-load-safe-path}.
23172 @tab Add directory trusted for automatic loading.
23175 @node Init File in the Current Directory
23176 @subsection Automatically loading init file in the current directory
23177 @cindex auto-loading init file in the current directory
23179 By default, @value{GDBN} reads and executes the canned sequences of commands
23180 from init file (if any) in the current working directory,
23181 see @ref{Init File in the Current Directory during Startup}.
23183 Note that loading of this local @file{.gdbinit} file also requires accordingly
23184 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
23187 @anchor{set auto-load local-gdbinit}
23188 @kindex set auto-load local-gdbinit
23189 @item set auto-load local-gdbinit [on|off]
23190 Enable or disable the auto-loading of canned sequences of commands
23191 (@pxref{Sequences}) found in init file in the current directory.
23193 @anchor{show auto-load local-gdbinit}
23194 @kindex show auto-load local-gdbinit
23195 @item show auto-load local-gdbinit
23196 Show whether auto-loading of canned sequences of commands from init file in the
23197 current directory is enabled or disabled.
23199 @anchor{info auto-load local-gdbinit}
23200 @kindex info auto-load local-gdbinit
23201 @item info auto-load local-gdbinit
23202 Print whether canned sequences of commands from init file in the
23203 current directory have been auto-loaded.
23206 @node libthread_db.so.1 file
23207 @subsection Automatically loading thread debugging library
23208 @cindex auto-loading libthread_db.so.1
23210 This feature is currently present only on @sc{gnu}/Linux native hosts.
23212 @value{GDBN} reads in some cases thread debugging library from places specific
23213 to the inferior (@pxref{set libthread-db-search-path}).
23215 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
23216 without checking this @samp{set auto-load libthread-db} switch as system
23217 libraries have to be trusted in general. In all other cases of
23218 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
23219 auto-load libthread-db} is enabled before trying to open such thread debugging
23222 Note that loading of this debugging library also requires accordingly configured
23223 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
23226 @anchor{set auto-load libthread-db}
23227 @kindex set auto-load libthread-db
23228 @item set auto-load libthread-db [on|off]
23229 Enable or disable the auto-loading of inferior specific thread debugging library.
23231 @anchor{show auto-load libthread-db}
23232 @kindex show auto-load libthread-db
23233 @item show auto-load libthread-db
23234 Show whether auto-loading of inferior specific thread debugging library is
23235 enabled or disabled.
23237 @anchor{info auto-load libthread-db}
23238 @kindex info auto-load libthread-db
23239 @item info auto-load libthread-db
23240 Print the list of all loaded inferior specific thread debugging libraries and
23241 for each such library print list of inferior @var{pid}s using it.
23244 @node Auto-loading safe path
23245 @subsection Security restriction for auto-loading
23246 @cindex auto-loading safe-path
23248 As the files of inferior can come from untrusted source (such as submitted by
23249 an application user) @value{GDBN} does not always load any files automatically.
23250 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
23251 directories trusted for loading files not explicitly requested by user.
23252 Each directory can also be a shell wildcard pattern.
23254 If the path is not set properly you will see a warning and the file will not
23259 Reading symbols from /home/user/gdb/gdb...done.
23260 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
23261 declined by your `auto-load safe-path' set
23262 to "$debugdir:$datadir/auto-load".
23263 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
23264 declined by your `auto-load safe-path' set
23265 to "$debugdir:$datadir/auto-load".
23269 To instruct @value{GDBN} to go ahead and use the init files anyway,
23270 invoke @value{GDBN} like this:
23273 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
23276 The list of trusted directories is controlled by the following commands:
23279 @anchor{set auto-load safe-path}
23280 @kindex set auto-load safe-path
23281 @item set auto-load safe-path @r{[}@var{directories}@r{]}
23282 Set the list of directories (and their subdirectories) trusted for automatic
23283 loading and execution of scripts. You can also enter a specific trusted file.
23284 Each directory can also be a shell wildcard pattern; wildcards do not match
23285 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
23286 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
23287 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
23288 its default value as specified during @value{GDBN} compilation.
23290 The list of directories uses path separator (@samp{:} on GNU and Unix
23291 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
23292 to the @env{PATH} environment variable.
23294 @anchor{show auto-load safe-path}
23295 @kindex show auto-load safe-path
23296 @item show auto-load safe-path
23297 Show the list of directories trusted for automatic loading and execution of
23300 @anchor{add-auto-load-safe-path}
23301 @kindex add-auto-load-safe-path
23302 @item add-auto-load-safe-path
23303 Add an entry (or list of entries) to the list of directories trusted for
23304 automatic loading and execution of scripts. Multiple entries may be delimited
23305 by the host platform path separator in use.
23308 This variable defaults to what @code{--with-auto-load-dir} has been configured
23309 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
23310 substitution applies the same as for @ref{set auto-load scripts-directory}.
23311 The default @code{set auto-load safe-path} value can be also overriden by
23312 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
23314 Setting this variable to @file{/} disables this security protection,
23315 corresponding @value{GDBN} configuration option is
23316 @option{--without-auto-load-safe-path}.
23317 This variable is supposed to be set to the system directories writable by the
23318 system superuser only. Users can add their source directories in init files in
23319 their home directories (@pxref{Home Directory Init File}). See also deprecated
23320 init file in the current directory
23321 (@pxref{Init File in the Current Directory during Startup}).
23323 To force @value{GDBN} to load the files it declined to load in the previous
23324 example, you could use one of the following ways:
23327 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
23328 Specify this trusted directory (or a file) as additional component of the list.
23329 You have to specify also any existing directories displayed by
23330 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
23332 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
23333 Specify this directory as in the previous case but just for a single
23334 @value{GDBN} session.
23336 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
23337 Disable auto-loading safety for a single @value{GDBN} session.
23338 This assumes all the files you debug during this @value{GDBN} session will come
23339 from trusted sources.
23341 @item @kbd{./configure --without-auto-load-safe-path}
23342 During compilation of @value{GDBN} you may disable any auto-loading safety.
23343 This assumes all the files you will ever debug with this @value{GDBN} come from
23347 On the other hand you can also explicitly forbid automatic files loading which
23348 also suppresses any such warning messages:
23351 @item @kbd{gdb -iex "set auto-load no" @dots{}}
23352 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
23354 @item @file{~/.gdbinit}: @samp{set auto-load no}
23355 Disable auto-loading globally for the user
23356 (@pxref{Home Directory Init File}). While it is improbable, you could also
23357 use system init file instead (@pxref{System-wide configuration}).
23360 This setting applies to the file names as entered by user. If no entry matches
23361 @value{GDBN} tries as a last resort to also resolve all the file names into
23362 their canonical form (typically resolving symbolic links) and compare the
23363 entries again. @value{GDBN} already canonicalizes most of the filenames on its
23364 own before starting the comparison so a canonical form of directories is
23365 recommended to be entered.
23367 @node Auto-loading verbose mode
23368 @subsection Displaying files tried for auto-load
23369 @cindex auto-loading verbose mode
23371 For better visibility of all the file locations where you can place scripts to
23372 be auto-loaded with inferior --- or to protect yourself against accidental
23373 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
23374 all the files attempted to be loaded. Both existing and non-existing files may
23377 For example the list of directories from which it is safe to auto-load files
23378 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
23379 may not be too obvious while setting it up.
23382 (gdb) set debug auto-load on
23383 (gdb) file ~/src/t/true
23384 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
23385 for objfile "/tmp/true".
23386 auto-load: Updating directories of "/usr:/opt".
23387 auto-load: Using directory "/usr".
23388 auto-load: Using directory "/opt".
23389 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
23390 by your `auto-load safe-path' set to "/usr:/opt".
23394 @anchor{set debug auto-load}
23395 @kindex set debug auto-load
23396 @item set debug auto-load [on|off]
23397 Set whether to print the filenames attempted to be auto-loaded.
23399 @anchor{show debug auto-load}
23400 @kindex show debug auto-load
23401 @item show debug auto-load
23402 Show whether printing of the filenames attempted to be auto-loaded is turned
23406 @node Messages/Warnings
23407 @section Optional Warnings and Messages
23409 @cindex verbose operation
23410 @cindex optional warnings
23411 By default, @value{GDBN} is silent about its inner workings. If you are
23412 running on a slow machine, you may want to use the @code{set verbose}
23413 command. This makes @value{GDBN} tell you when it does a lengthy
23414 internal operation, so you will not think it has crashed.
23416 Currently, the messages controlled by @code{set verbose} are those
23417 which announce that the symbol table for a source file is being read;
23418 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
23421 @kindex set verbose
23422 @item set verbose on
23423 Enables @value{GDBN} output of certain informational messages.
23425 @item set verbose off
23426 Disables @value{GDBN} output of certain informational messages.
23428 @kindex show verbose
23430 Displays whether @code{set verbose} is on or off.
23433 By default, if @value{GDBN} encounters bugs in the symbol table of an
23434 object file, it is silent; but if you are debugging a compiler, you may
23435 find this information useful (@pxref{Symbol Errors, ,Errors Reading
23440 @kindex set complaints
23441 @item set complaints @var{limit}
23442 Permits @value{GDBN} to output @var{limit} complaints about each type of
23443 unusual symbols before becoming silent about the problem. Set
23444 @var{limit} to zero to suppress all complaints; set it to a large number
23445 to prevent complaints from being suppressed.
23447 @kindex show complaints
23448 @item show complaints
23449 Displays how many symbol complaints @value{GDBN} is permitted to produce.
23453 @anchor{confirmation requests}
23454 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
23455 lot of stupid questions to confirm certain commands. For example, if
23456 you try to run a program which is already running:
23460 The program being debugged has been started already.
23461 Start it from the beginning? (y or n)
23464 If you are willing to unflinchingly face the consequences of your own
23465 commands, you can disable this ``feature'':
23469 @kindex set confirm
23471 @cindex confirmation
23472 @cindex stupid questions
23473 @item set confirm off
23474 Disables confirmation requests. Note that running @value{GDBN} with
23475 the @option{--batch} option (@pxref{Mode Options, -batch}) also
23476 automatically disables confirmation requests.
23478 @item set confirm on
23479 Enables confirmation requests (the default).
23481 @kindex show confirm
23483 Displays state of confirmation requests.
23487 @cindex command tracing
23488 If you need to debug user-defined commands or sourced files you may find it
23489 useful to enable @dfn{command tracing}. In this mode each command will be
23490 printed as it is executed, prefixed with one or more @samp{+} symbols, the
23491 quantity denoting the call depth of each command.
23494 @kindex set trace-commands
23495 @cindex command scripts, debugging
23496 @item set trace-commands on
23497 Enable command tracing.
23498 @item set trace-commands off
23499 Disable command tracing.
23500 @item show trace-commands
23501 Display the current state of command tracing.
23504 @node Debugging Output
23505 @section Optional Messages about Internal Happenings
23506 @cindex optional debugging messages
23508 @value{GDBN} has commands that enable optional debugging messages from
23509 various @value{GDBN} subsystems; normally these commands are of
23510 interest to @value{GDBN} maintainers, or when reporting a bug. This
23511 section documents those commands.
23514 @kindex set exec-done-display
23515 @item set exec-done-display
23516 Turns on or off the notification of asynchronous commands'
23517 completion. When on, @value{GDBN} will print a message when an
23518 asynchronous command finishes its execution. The default is off.
23519 @kindex show exec-done-display
23520 @item show exec-done-display
23521 Displays the current setting of asynchronous command completion
23524 @cindex ARM AArch64
23525 @item set debug aarch64
23526 Turns on or off display of debugging messages related to ARM AArch64.
23527 The default is off.
23529 @item show debug aarch64
23530 Displays the current state of displaying debugging messages related to
23532 @cindex gdbarch debugging info
23533 @cindex architecture debugging info
23534 @item set debug arch
23535 Turns on or off display of gdbarch debugging info. The default is off
23536 @item show debug arch
23537 Displays the current state of displaying gdbarch debugging info.
23538 @item set debug aix-solib
23539 @cindex AIX shared library debugging
23540 Control display of debugging messages from the AIX shared library
23541 support module. The default is off.
23542 @item show debug aix-thread
23543 Show the current state of displaying AIX shared library debugging messages.
23544 @item set debug aix-thread
23545 @cindex AIX threads
23546 Display debugging messages about inner workings of the AIX thread
23548 @item show debug aix-thread
23549 Show the current state of AIX thread debugging info display.
23550 @item set debug check-physname
23552 Check the results of the ``physname'' computation. When reading DWARF
23553 debugging information for C@t{++}, @value{GDBN} attempts to compute
23554 each entity's name. @value{GDBN} can do this computation in two
23555 different ways, depending on exactly what information is present.
23556 When enabled, this setting causes @value{GDBN} to compute the names
23557 both ways and display any discrepancies.
23558 @item show debug check-physname
23559 Show the current state of ``physname'' checking.
23560 @item set debug coff-pe-read
23561 @cindex COFF/PE exported symbols
23562 Control display of debugging messages related to reading of COFF/PE
23563 exported symbols. The default is off.
23564 @item show debug coff-pe-read
23565 Displays the current state of displaying debugging messages related to
23566 reading of COFF/PE exported symbols.
23567 @item set debug dwarf-die
23569 Dump DWARF DIEs after they are read in.
23570 The value is the number of nesting levels to print.
23571 A value of zero turns off the display.
23572 @item show debug dwarf-die
23573 Show the current state of DWARF DIE debugging.
23574 @item set debug dwarf-line
23575 @cindex DWARF Line Tables
23576 Turns on or off display of debugging messages related to reading
23577 DWARF line tables. The default is 0 (off).
23578 A value of 1 provides basic information.
23579 A value greater than 1 provides more verbose information.
23580 @item show debug dwarf-line
23581 Show the current state of DWARF line table debugging.
23582 @item set debug dwarf-read
23583 @cindex DWARF Reading
23584 Turns on or off display of debugging messages related to reading
23585 DWARF debug info. The default is 0 (off).
23586 A value of 1 provides basic information.
23587 A value greater than 1 provides more verbose information.
23588 @item show debug dwarf-read
23589 Show the current state of DWARF reader debugging.
23590 @item set debug displaced
23591 @cindex displaced stepping debugging info
23592 Turns on or off display of @value{GDBN} debugging info for the
23593 displaced stepping support. The default is off.
23594 @item show debug displaced
23595 Displays the current state of displaying @value{GDBN} debugging info
23596 related to displaced stepping.
23597 @item set debug event
23598 @cindex event debugging info
23599 Turns on or off display of @value{GDBN} event debugging info. The
23601 @item show debug event
23602 Displays the current state of displaying @value{GDBN} event debugging
23604 @item set debug expression
23605 @cindex expression debugging info
23606 Turns on or off display of debugging info about @value{GDBN}
23607 expression parsing. The default is off.
23608 @item show debug expression
23609 Displays the current state of displaying debugging info about
23610 @value{GDBN} expression parsing.
23611 @item set debug fbsd-lwp
23612 @cindex FreeBSD LWP debug messages
23613 Turns on or off debugging messages from the FreeBSD LWP debug support.
23614 @item show debug fbsd-lwp
23615 Show the current state of FreeBSD LWP debugging messages.
23616 @item set debug frame
23617 @cindex frame debugging info
23618 Turns on or off display of @value{GDBN} frame debugging info. The
23620 @item show debug frame
23621 Displays the current state of displaying @value{GDBN} frame debugging
23623 @item set debug gnu-nat
23624 @cindex @sc{gnu}/Hurd debug messages
23625 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
23626 @item show debug gnu-nat
23627 Show the current state of @sc{gnu}/Hurd debugging messages.
23628 @item set debug infrun
23629 @cindex inferior debugging info
23630 Turns on or off display of @value{GDBN} debugging info for running the inferior.
23631 The default is off. @file{infrun.c} contains GDB's runtime state machine used
23632 for implementing operations such as single-stepping the inferior.
23633 @item show debug infrun
23634 Displays the current state of @value{GDBN} inferior debugging.
23635 @item set debug jit
23636 @cindex just-in-time compilation, debugging messages
23637 Turn on or off debugging messages from JIT debug support.
23638 @item show debug jit
23639 Displays the current state of @value{GDBN} JIT debugging.
23640 @item set debug lin-lwp
23641 @cindex @sc{gnu}/Linux LWP debug messages
23642 @cindex Linux lightweight processes
23643 Turn on or off debugging messages from the Linux LWP debug support.
23644 @item show debug lin-lwp
23645 Show the current state of Linux LWP debugging messages.
23646 @item set debug linux-namespaces
23647 @cindex @sc{gnu}/Linux namespaces debug messages
23648 Turn on or off debugging messages from the Linux namespaces debug support.
23649 @item show debug linux-namespaces
23650 Show the current state of Linux namespaces debugging messages.
23651 @item set debug mach-o
23652 @cindex Mach-O symbols processing
23653 Control display of debugging messages related to Mach-O symbols
23654 processing. The default is off.
23655 @item show debug mach-o
23656 Displays the current state of displaying debugging messages related to
23657 reading of COFF/PE exported symbols.
23658 @item set debug notification
23659 @cindex remote async notification debugging info
23660 Turn on or off debugging messages about remote async notification.
23661 The default is off.
23662 @item show debug notification
23663 Displays the current state of remote async notification debugging messages.
23664 @item set debug observer
23665 @cindex observer debugging info
23666 Turns on or off display of @value{GDBN} observer debugging. This
23667 includes info such as the notification of observable events.
23668 @item show debug observer
23669 Displays the current state of observer debugging.
23670 @item set debug overload
23671 @cindex C@t{++} overload debugging info
23672 Turns on or off display of @value{GDBN} C@t{++} overload debugging
23673 info. This includes info such as ranking of functions, etc. The default
23675 @item show debug overload
23676 Displays the current state of displaying @value{GDBN} C@t{++} overload
23678 @cindex expression parser, debugging info
23679 @cindex debug expression parser
23680 @item set debug parser
23681 Turns on or off the display of expression parser debugging output.
23682 Internally, this sets the @code{yydebug} variable in the expression
23683 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
23684 details. The default is off.
23685 @item show debug parser
23686 Show the current state of expression parser debugging.
23687 @cindex packets, reporting on stdout
23688 @cindex serial connections, debugging
23689 @cindex debug remote protocol
23690 @cindex remote protocol debugging
23691 @cindex display remote packets
23692 @item set debug remote
23693 Turns on or off display of reports on all packets sent back and forth across
23694 the serial line to the remote machine. The info is printed on the
23695 @value{GDBN} standard output stream. The default is off.
23696 @item show debug remote
23697 Displays the state of display of remote packets.
23698 @item set debug serial
23699 Turns on or off display of @value{GDBN} serial debugging info. The
23701 @item show debug serial
23702 Displays the current state of displaying @value{GDBN} serial debugging
23704 @item set debug solib-frv
23705 @cindex FR-V shared-library debugging
23706 Turn on or off debugging messages for FR-V shared-library code.
23707 @item show debug solib-frv
23708 Display the current state of FR-V shared-library code debugging
23710 @item set debug symbol-lookup
23711 @cindex symbol lookup
23712 Turns on or off display of debugging messages related to symbol lookup.
23713 The default is 0 (off).
23714 A value of 1 provides basic information.
23715 A value greater than 1 provides more verbose information.
23716 @item show debug symbol-lookup
23717 Show the current state of symbol lookup debugging messages.
23718 @item set debug symfile
23719 @cindex symbol file functions
23720 Turns on or off display of debugging messages related to symbol file functions.
23721 The default is off. @xref{Files}.
23722 @item show debug symfile
23723 Show the current state of symbol file debugging messages.
23724 @item set debug symtab-create
23725 @cindex symbol table creation
23726 Turns on or off display of debugging messages related to symbol table creation.
23727 The default is 0 (off).
23728 A value of 1 provides basic information.
23729 A value greater than 1 provides more verbose information.
23730 @item show debug symtab-create
23731 Show the current state of symbol table creation debugging.
23732 @item set debug target
23733 @cindex target debugging info
23734 Turns on or off display of @value{GDBN} target debugging info. This info
23735 includes what is going on at the target level of GDB, as it happens. The
23736 default is 0. Set it to 1 to track events, and to 2 to also track the
23737 value of large memory transfers.
23738 @item show debug target
23739 Displays the current state of displaying @value{GDBN} target debugging
23741 @item set debug timestamp
23742 @cindex timestampping debugging info
23743 Turns on or off display of timestamps with @value{GDBN} debugging info.
23744 When enabled, seconds and microseconds are displayed before each debugging
23746 @item show debug timestamp
23747 Displays the current state of displaying timestamps with @value{GDBN}
23749 @item set debug varobj
23750 @cindex variable object debugging info
23751 Turns on or off display of @value{GDBN} variable object debugging
23752 info. The default is off.
23753 @item show debug varobj
23754 Displays the current state of displaying @value{GDBN} variable object
23756 @item set debug xml
23757 @cindex XML parser debugging
23758 Turn on or off debugging messages for built-in XML parsers.
23759 @item show debug xml
23760 Displays the current state of XML debugging messages.
23763 @node Other Misc Settings
23764 @section Other Miscellaneous Settings
23765 @cindex miscellaneous settings
23768 @kindex set interactive-mode
23769 @item set interactive-mode
23770 If @code{on}, forces @value{GDBN} to assume that GDB was started
23771 in a terminal. In practice, this means that @value{GDBN} should wait
23772 for the user to answer queries generated by commands entered at
23773 the command prompt. If @code{off}, forces @value{GDBN} to operate
23774 in the opposite mode, and it uses the default answers to all queries.
23775 If @code{auto} (the default), @value{GDBN} tries to determine whether
23776 its standard input is a terminal, and works in interactive-mode if it
23777 is, non-interactively otherwise.
23779 In the vast majority of cases, the debugger should be able to guess
23780 correctly which mode should be used. But this setting can be useful
23781 in certain specific cases, such as running a MinGW @value{GDBN}
23782 inside a cygwin window.
23784 @kindex show interactive-mode
23785 @item show interactive-mode
23786 Displays whether the debugger is operating in interactive mode or not.
23789 @node Extending GDB
23790 @chapter Extending @value{GDBN}
23791 @cindex extending GDB
23793 @value{GDBN} provides several mechanisms for extension.
23794 @value{GDBN} also provides the ability to automatically load
23795 extensions when it reads a file for debugging. This allows the
23796 user to automatically customize @value{GDBN} for the program
23800 * Sequences:: Canned Sequences of @value{GDBN} Commands
23801 * Python:: Extending @value{GDBN} using Python
23802 * Guile:: Extending @value{GDBN} using Guile
23803 * Auto-loading extensions:: Automatically loading extensions
23804 * Multiple Extension Languages:: Working with multiple extension languages
23805 * Aliases:: Creating new spellings of existing commands
23808 To facilitate the use of extension languages, @value{GDBN} is capable
23809 of evaluating the contents of a file. When doing so, @value{GDBN}
23810 can recognize which extension language is being used by looking at
23811 the filename extension. Files with an unrecognized filename extension
23812 are always treated as a @value{GDBN} Command Files.
23813 @xref{Command Files,, Command files}.
23815 You can control how @value{GDBN} evaluates these files with the following
23819 @kindex set script-extension
23820 @kindex show script-extension
23821 @item set script-extension off
23822 All scripts are always evaluated as @value{GDBN} Command Files.
23824 @item set script-extension soft
23825 The debugger determines the scripting language based on filename
23826 extension. If this scripting language is supported, @value{GDBN}
23827 evaluates the script using that language. Otherwise, it evaluates
23828 the file as a @value{GDBN} Command File.
23830 @item set script-extension strict
23831 The debugger determines the scripting language based on filename
23832 extension, and evaluates the script using that language. If the
23833 language is not supported, then the evaluation fails.
23835 @item show script-extension
23836 Display the current value of the @code{script-extension} option.
23841 @section Canned Sequences of Commands
23843 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
23844 Command Lists}), @value{GDBN} provides two ways to store sequences of
23845 commands for execution as a unit: user-defined commands and command
23849 * Define:: How to define your own commands
23850 * Hooks:: Hooks for user-defined commands
23851 * Command Files:: How to write scripts of commands to be stored in a file
23852 * Output:: Commands for controlled output
23853 * Auto-loading sequences:: Controlling auto-loaded command files
23857 @subsection User-defined Commands
23859 @cindex user-defined command
23860 @cindex arguments, to user-defined commands
23861 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
23862 which you assign a new name as a command. This is done with the
23863 @code{define} command. User commands may accept up to 10 arguments
23864 separated by whitespace. Arguments are accessed within the user command
23865 via @code{$arg0@dots{}$arg9}. A trivial example:
23869 print $arg0 + $arg1 + $arg2
23874 To execute the command use:
23881 This defines the command @code{adder}, which prints the sum of
23882 its three arguments. Note the arguments are text substitutions, so they may
23883 reference variables, use complex expressions, or even perform inferior
23886 @cindex argument count in user-defined commands
23887 @cindex how many arguments (user-defined commands)
23888 In addition, @code{$argc} may be used to find out how many arguments have
23889 been passed. This expands to a number in the range 0@dots{}10.
23894 print $arg0 + $arg1
23897 print $arg0 + $arg1 + $arg2
23905 @item define @var{commandname}
23906 Define a command named @var{commandname}. If there is already a command
23907 by that name, you are asked to confirm that you want to redefine it.
23908 The argument @var{commandname} may be a bare command name consisting of letters,
23909 numbers, dashes, and underscores. It may also start with any predefined
23910 prefix command. For example, @samp{define target my-target} creates
23911 a user-defined @samp{target my-target} command.
23913 The definition of the command is made up of other @value{GDBN} command lines,
23914 which are given following the @code{define} command. The end of these
23915 commands is marked by a line containing @code{end}.
23918 @kindex end@r{ (user-defined commands)}
23919 @item document @var{commandname}
23920 Document the user-defined command @var{commandname}, so that it can be
23921 accessed by @code{help}. The command @var{commandname} must already be
23922 defined. This command reads lines of documentation just as @code{define}
23923 reads the lines of the command definition, ending with @code{end}.
23924 After the @code{document} command is finished, @code{help} on command
23925 @var{commandname} displays the documentation you have written.
23927 You may use the @code{document} command again to change the
23928 documentation of a command. Redefining the command with @code{define}
23929 does not change the documentation.
23931 @kindex dont-repeat
23932 @cindex don't repeat command
23934 Used inside a user-defined command, this tells @value{GDBN} that this
23935 command should not be repeated when the user hits @key{RET}
23936 (@pxref{Command Syntax, repeat last command}).
23938 @kindex help user-defined
23939 @item help user-defined
23940 List all user-defined commands and all python commands defined in class
23941 COMAND_USER. The first line of the documentation or docstring is
23946 @itemx show user @var{commandname}
23947 Display the @value{GDBN} commands used to define @var{commandname} (but
23948 not its documentation). If no @var{commandname} is given, display the
23949 definitions for all user-defined commands.
23950 This does not work for user-defined python commands.
23952 @cindex infinite recursion in user-defined commands
23953 @kindex show max-user-call-depth
23954 @kindex set max-user-call-depth
23955 @item show max-user-call-depth
23956 @itemx set max-user-call-depth
23957 The value of @code{max-user-call-depth} controls how many recursion
23958 levels are allowed in user-defined commands before @value{GDBN} suspects an
23959 infinite recursion and aborts the command.
23960 This does not apply to user-defined python commands.
23963 In addition to the above commands, user-defined commands frequently
23964 use control flow commands, described in @ref{Command Files}.
23966 When user-defined commands are executed, the
23967 commands of the definition are not printed. An error in any command
23968 stops execution of the user-defined command.
23970 If used interactively, commands that would ask for confirmation proceed
23971 without asking when used inside a user-defined command. Many @value{GDBN}
23972 commands that normally print messages to say what they are doing omit the
23973 messages when used in a user-defined command.
23976 @subsection User-defined Command Hooks
23977 @cindex command hooks
23978 @cindex hooks, for commands
23979 @cindex hooks, pre-command
23982 You may define @dfn{hooks}, which are a special kind of user-defined
23983 command. Whenever you run the command @samp{foo}, if the user-defined
23984 command @samp{hook-foo} exists, it is executed (with no arguments)
23985 before that command.
23987 @cindex hooks, post-command
23989 A hook may also be defined which is run after the command you executed.
23990 Whenever you run the command @samp{foo}, if the user-defined command
23991 @samp{hookpost-foo} exists, it is executed (with no arguments) after
23992 that command. Post-execution hooks may exist simultaneously with
23993 pre-execution hooks, for the same command.
23995 It is valid for a hook to call the command which it hooks. If this
23996 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
23998 @c It would be nice if hookpost could be passed a parameter indicating
23999 @c if the command it hooks executed properly or not. FIXME!
24001 @kindex stop@r{, a pseudo-command}
24002 In addition, a pseudo-command, @samp{stop} exists. Defining
24003 (@samp{hook-stop}) makes the associated commands execute every time
24004 execution stops in your program: before breakpoint commands are run,
24005 displays are printed, or the stack frame is printed.
24007 For example, to ignore @code{SIGALRM} signals while
24008 single-stepping, but treat them normally during normal execution,
24013 handle SIGALRM nopass
24017 handle SIGALRM pass
24020 define hook-continue
24021 handle SIGALRM pass
24025 As a further example, to hook at the beginning and end of the @code{echo}
24026 command, and to add extra text to the beginning and end of the message,
24034 define hookpost-echo
24038 (@value{GDBP}) echo Hello World
24039 <<<---Hello World--->>>
24044 You can define a hook for any single-word command in @value{GDBN}, but
24045 not for command aliases; you should define a hook for the basic command
24046 name, e.g.@: @code{backtrace} rather than @code{bt}.
24047 @c FIXME! So how does Joe User discover whether a command is an alias
24049 You can hook a multi-word command by adding @code{hook-} or
24050 @code{hookpost-} to the last word of the command, e.g.@:
24051 @samp{define target hook-remote} to add a hook to @samp{target remote}.
24053 If an error occurs during the execution of your hook, execution of
24054 @value{GDBN} commands stops and @value{GDBN} issues a prompt
24055 (before the command that you actually typed had a chance to run).
24057 If you try to define a hook which does not match any known command, you
24058 get a warning from the @code{define} command.
24060 @node Command Files
24061 @subsection Command Files
24063 @cindex command files
24064 @cindex scripting commands
24065 A command file for @value{GDBN} is a text file made of lines that are
24066 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
24067 also be included. An empty line in a command file does nothing; it
24068 does not mean to repeat the last command, as it would from the
24071 You can request the execution of a command file with the @code{source}
24072 command. Note that the @code{source} command is also used to evaluate
24073 scripts that are not Command Files. The exact behavior can be configured
24074 using the @code{script-extension} setting.
24075 @xref{Extending GDB,, Extending GDB}.
24079 @cindex execute commands from a file
24080 @item source [-s] [-v] @var{filename}
24081 Execute the command file @var{filename}.
24084 The lines in a command file are generally executed sequentially,
24085 unless the order of execution is changed by one of the
24086 @emph{flow-control commands} described below. The commands are not
24087 printed as they are executed. An error in any command terminates
24088 execution of the command file and control is returned to the console.
24090 @value{GDBN} first searches for @var{filename} in the current directory.
24091 If the file is not found there, and @var{filename} does not specify a
24092 directory, then @value{GDBN} also looks for the file on the source search path
24093 (specified with the @samp{directory} command);
24094 except that @file{$cdir} is not searched because the compilation directory
24095 is not relevant to scripts.
24097 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
24098 on the search path even if @var{filename} specifies a directory.
24099 The search is done by appending @var{filename} to each element of the
24100 search path. So, for example, if @var{filename} is @file{mylib/myscript}
24101 and the search path contains @file{/home/user} then @value{GDBN} will
24102 look for the script @file{/home/user/mylib/myscript}.
24103 The search is also done if @var{filename} is an absolute path.
24104 For example, if @var{filename} is @file{/tmp/myscript} and
24105 the search path contains @file{/home/user} then @value{GDBN} will
24106 look for the script @file{/home/user/tmp/myscript}.
24107 For DOS-like systems, if @var{filename} contains a drive specification,
24108 it is stripped before concatenation. For example, if @var{filename} is
24109 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
24110 will look for the script @file{c:/tmp/myscript}.
24112 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
24113 each command as it is executed. The option must be given before
24114 @var{filename}, and is interpreted as part of the filename anywhere else.
24116 Commands that would ask for confirmation if used interactively proceed
24117 without asking when used in a command file. Many @value{GDBN} commands that
24118 normally print messages to say what they are doing omit the messages
24119 when called from command files.
24121 @value{GDBN} also accepts command input from standard input. In this
24122 mode, normal output goes to standard output and error output goes to
24123 standard error. Errors in a command file supplied on standard input do
24124 not terminate execution of the command file---execution continues with
24128 gdb < cmds > log 2>&1
24131 (The syntax above will vary depending on the shell used.) This example
24132 will execute commands from the file @file{cmds}. All output and errors
24133 would be directed to @file{log}.
24135 Since commands stored on command files tend to be more general than
24136 commands typed interactively, they frequently need to deal with
24137 complicated situations, such as different or unexpected values of
24138 variables and symbols, changes in how the program being debugged is
24139 built, etc. @value{GDBN} provides a set of flow-control commands to
24140 deal with these complexities. Using these commands, you can write
24141 complex scripts that loop over data structures, execute commands
24142 conditionally, etc.
24149 This command allows to include in your script conditionally executed
24150 commands. The @code{if} command takes a single argument, which is an
24151 expression to evaluate. It is followed by a series of commands that
24152 are executed only if the expression is true (its value is nonzero).
24153 There can then optionally be an @code{else} line, followed by a series
24154 of commands that are only executed if the expression was false. The
24155 end of the list is marked by a line containing @code{end}.
24159 This command allows to write loops. Its syntax is similar to
24160 @code{if}: the command takes a single argument, which is an expression
24161 to evaluate, and must be followed by the commands to execute, one per
24162 line, terminated by an @code{end}. These commands are called the
24163 @dfn{body} of the loop. The commands in the body of @code{while} are
24164 executed repeatedly as long as the expression evaluates to true.
24168 This command exits the @code{while} loop in whose body it is included.
24169 Execution of the script continues after that @code{while}s @code{end}
24172 @kindex loop_continue
24173 @item loop_continue
24174 This command skips the execution of the rest of the body of commands
24175 in the @code{while} loop in whose body it is included. Execution
24176 branches to the beginning of the @code{while} loop, where it evaluates
24177 the controlling expression.
24179 @kindex end@r{ (if/else/while commands)}
24181 Terminate the block of commands that are the body of @code{if},
24182 @code{else}, or @code{while} flow-control commands.
24187 @subsection Commands for Controlled Output
24189 During the execution of a command file or a user-defined command, normal
24190 @value{GDBN} output is suppressed; the only output that appears is what is
24191 explicitly printed by the commands in the definition. This section
24192 describes three commands useful for generating exactly the output you
24197 @item echo @var{text}
24198 @c I do not consider backslash-space a standard C escape sequence
24199 @c because it is not in ANSI.
24200 Print @var{text}. Nonprinting characters can be included in
24201 @var{text} using C escape sequences, such as @samp{\n} to print a
24202 newline. @strong{No newline is printed unless you specify one.}
24203 In addition to the standard C escape sequences, a backslash followed
24204 by a space stands for a space. This is useful for displaying a
24205 string with spaces at the beginning or the end, since leading and
24206 trailing spaces are otherwise trimmed from all arguments.
24207 To print @samp{@w{ }and foo =@w{ }}, use the command
24208 @samp{echo \@w{ }and foo = \@w{ }}.
24210 A backslash at the end of @var{text} can be used, as in C, to continue
24211 the command onto subsequent lines. For example,
24214 echo This is some text\n\
24215 which is continued\n\
24216 onto several lines.\n
24219 produces the same output as
24222 echo This is some text\n
24223 echo which is continued\n
24224 echo onto several lines.\n
24228 @item output @var{expression}
24229 Print the value of @var{expression} and nothing but that value: no
24230 newlines, no @samp{$@var{nn} = }. The value is not entered in the
24231 value history either. @xref{Expressions, ,Expressions}, for more information
24234 @item output/@var{fmt} @var{expression}
24235 Print the value of @var{expression} in format @var{fmt}. You can use
24236 the same formats as for @code{print}. @xref{Output Formats,,Output
24237 Formats}, for more information.
24240 @item printf @var{template}, @var{expressions}@dots{}
24241 Print the values of one or more @var{expressions} under the control of
24242 the string @var{template}. To print several values, make
24243 @var{expressions} be a comma-separated list of individual expressions,
24244 which may be either numbers or pointers. Their values are printed as
24245 specified by @var{template}, exactly as a C program would do by
24246 executing the code below:
24249 printf (@var{template}, @var{expressions}@dots{});
24252 As in @code{C} @code{printf}, ordinary characters in @var{template}
24253 are printed verbatim, while @dfn{conversion specification} introduced
24254 by the @samp{%} character cause subsequent @var{expressions} to be
24255 evaluated, their values converted and formatted according to type and
24256 style information encoded in the conversion specifications, and then
24259 For example, you can print two values in hex like this:
24262 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
24265 @code{printf} supports all the standard @code{C} conversion
24266 specifications, including the flags and modifiers between the @samp{%}
24267 character and the conversion letter, with the following exceptions:
24271 The argument-ordering modifiers, such as @samp{2$}, are not supported.
24274 The modifier @samp{*} is not supported for specifying precision or
24278 The @samp{'} flag (for separation of digits into groups according to
24279 @code{LC_NUMERIC'}) is not supported.
24282 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
24286 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
24289 The conversion letters @samp{a} and @samp{A} are not supported.
24293 Note that the @samp{ll} type modifier is supported only if the
24294 underlying @code{C} implementation used to build @value{GDBN} supports
24295 the @code{long long int} type, and the @samp{L} type modifier is
24296 supported only if @code{long double} type is available.
24298 As in @code{C}, @code{printf} supports simple backslash-escape
24299 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
24300 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
24301 single character. Octal and hexadecimal escape sequences are not
24304 Additionally, @code{printf} supports conversion specifications for DFP
24305 (@dfn{Decimal Floating Point}) types using the following length modifiers
24306 together with a floating point specifier.
24311 @samp{H} for printing @code{Decimal32} types.
24314 @samp{D} for printing @code{Decimal64} types.
24317 @samp{DD} for printing @code{Decimal128} types.
24320 If the underlying @code{C} implementation used to build @value{GDBN} has
24321 support for the three length modifiers for DFP types, other modifiers
24322 such as width and precision will also be available for @value{GDBN} to use.
24324 In case there is no such @code{C} support, no additional modifiers will be
24325 available and the value will be printed in the standard way.
24327 Here's an example of printing DFP types using the above conversion letters:
24329 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
24333 @item eval @var{template}, @var{expressions}@dots{}
24334 Convert the values of one or more @var{expressions} under the control of
24335 the string @var{template} to a command line, and call it.
24339 @node Auto-loading sequences
24340 @subsection Controlling auto-loading native @value{GDBN} scripts
24341 @cindex native script auto-loading
24343 When a new object file is read (for example, due to the @code{file}
24344 command, or because the inferior has loaded a shared library),
24345 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
24346 @xref{Auto-loading extensions}.
24348 Auto-loading can be enabled or disabled,
24349 and the list of auto-loaded scripts can be printed.
24352 @anchor{set auto-load gdb-scripts}
24353 @kindex set auto-load gdb-scripts
24354 @item set auto-load gdb-scripts [on|off]
24355 Enable or disable the auto-loading of canned sequences of commands scripts.
24357 @anchor{show auto-load gdb-scripts}
24358 @kindex show auto-load gdb-scripts
24359 @item show auto-load gdb-scripts
24360 Show whether auto-loading of canned sequences of commands scripts is enabled or
24363 @anchor{info auto-load gdb-scripts}
24364 @kindex info auto-load gdb-scripts
24365 @cindex print list of auto-loaded canned sequences of commands scripts
24366 @item info auto-load gdb-scripts [@var{regexp}]
24367 Print the list of all canned sequences of commands scripts that @value{GDBN}
24371 If @var{regexp} is supplied only canned sequences of commands scripts with
24372 matching names are printed.
24374 @c Python docs live in a separate file.
24375 @include python.texi
24377 @c Guile docs live in a separate file.
24378 @include guile.texi
24380 @node Auto-loading extensions
24381 @section Auto-loading extensions
24382 @cindex auto-loading extensions
24384 @value{GDBN} provides two mechanisms for automatically loading extensions
24385 when a new object file is read (for example, due to the @code{file}
24386 command, or because the inferior has loaded a shared library):
24387 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
24388 section of modern file formats like ELF.
24391 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
24392 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
24393 * Which flavor to choose?::
24396 The auto-loading feature is useful for supplying application-specific
24397 debugging commands and features.
24399 Auto-loading can be enabled or disabled,
24400 and the list of auto-loaded scripts can be printed.
24401 See the @samp{auto-loading} section of each extension language
24402 for more information.
24403 For @value{GDBN} command files see @ref{Auto-loading sequences}.
24404 For Python files see @ref{Python Auto-loading}.
24406 Note that loading of this script file also requires accordingly configured
24407 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24409 @node objfile-gdbdotext file
24410 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
24411 @cindex @file{@var{objfile}-gdb.gdb}
24412 @cindex @file{@var{objfile}-gdb.py}
24413 @cindex @file{@var{objfile}-gdb.scm}
24415 When a new object file is read, @value{GDBN} looks for a file named
24416 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
24417 where @var{objfile} is the object file's name and
24418 where @var{ext} is the file extension for the extension language:
24421 @item @file{@var{objfile}-gdb.gdb}
24422 GDB's own command language
24423 @item @file{@var{objfile}-gdb.py}
24425 @item @file{@var{objfile}-gdb.scm}
24429 @var{script-name} is formed by ensuring that the file name of @var{objfile}
24430 is absolute, following all symlinks, and resolving @code{.} and @code{..}
24431 components, and appending the @file{-gdb.@var{ext}} suffix.
24432 If this file exists and is readable, @value{GDBN} will evaluate it as a
24433 script in the specified extension language.
24435 If this file does not exist, then @value{GDBN} will look for
24436 @var{script-name} file in all of the directories as specified below.
24438 Note that loading of these files requires an accordingly configured
24439 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24441 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
24442 scripts normally according to its @file{.exe} filename. But if no scripts are
24443 found @value{GDBN} also tries script filenames matching the object file without
24444 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
24445 is attempted on any platform. This makes the script filenames compatible
24446 between Unix and MS-Windows hosts.
24449 @anchor{set auto-load scripts-directory}
24450 @kindex set auto-load scripts-directory
24451 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
24452 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
24453 may be delimited by the host platform path separator in use
24454 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
24456 Each entry here needs to be covered also by the security setting
24457 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
24459 @anchor{with-auto-load-dir}
24460 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
24461 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
24462 configuration option @option{--with-auto-load-dir}.
24464 Any reference to @file{$debugdir} will get replaced by
24465 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
24466 reference to @file{$datadir} will get replaced by @var{data-directory} which is
24467 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
24468 @file{$datadir} must be placed as a directory component --- either alone or
24469 delimited by @file{/} or @file{\} directory separators, depending on the host
24472 The list of directories uses path separator (@samp{:} on GNU and Unix
24473 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
24474 to the @env{PATH} environment variable.
24476 @anchor{show auto-load scripts-directory}
24477 @kindex show auto-load scripts-directory
24478 @item show auto-load scripts-directory
24479 Show @value{GDBN} auto-loaded scripts location.
24481 @anchor{add-auto-load-scripts-directory}
24482 @kindex add-auto-load-scripts-directory
24483 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
24484 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
24485 Multiple entries may be delimited by the host platform path separator in use.
24488 @value{GDBN} does not track which files it has already auto-loaded this way.
24489 @value{GDBN} will load the associated script every time the corresponding
24490 @var{objfile} is opened.
24491 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
24492 is evaluated more than once.
24494 @node dotdebug_gdb_scripts section
24495 @subsection The @code{.debug_gdb_scripts} section
24496 @cindex @code{.debug_gdb_scripts} section
24498 For systems using file formats like ELF and COFF,
24499 when @value{GDBN} loads a new object file
24500 it will look for a special section named @code{.debug_gdb_scripts}.
24501 If this section exists, its contents is a list of null-terminated entries
24502 specifying scripts to load. Each entry begins with a non-null prefix byte that
24503 specifies the kind of entry, typically the extension language and whether the
24504 script is in a file or inlined in @code{.debug_gdb_scripts}.
24506 The following entries are supported:
24509 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
24510 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
24511 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
24512 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
24515 @subsubsection Script File Entries
24517 If the entry specifies a file, @value{GDBN} will look for the file first
24518 in the current directory and then along the source search path
24519 (@pxref{Source Path, ,Specifying Source Directories}),
24520 except that @file{$cdir} is not searched, since the compilation
24521 directory is not relevant to scripts.
24523 File entries can be placed in section @code{.debug_gdb_scripts} with,
24524 for example, this GCC macro for Python scripts.
24527 /* Note: The "MS" section flags are to remove duplicates. */
24528 #define DEFINE_GDB_PY_SCRIPT(script_name) \
24530 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
24531 .byte 1 /* Python */\n\
24532 .asciz \"" script_name "\"\n\
24538 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
24539 Then one can reference the macro in a header or source file like this:
24542 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
24545 The script name may include directories if desired.
24547 Note that loading of this script file also requires accordingly configured
24548 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24550 If the macro invocation is put in a header, any application or library
24551 using this header will get a reference to the specified script,
24552 and with the use of @code{"MS"} attributes on the section, the linker
24553 will remove duplicates.
24555 @subsubsection Script Text Entries
24557 Script text entries allow to put the executable script in the entry
24558 itself instead of loading it from a file.
24559 The first line of the entry, everything after the prefix byte and up to
24560 the first newline (@code{0xa}) character, is the script name, and must not
24561 contain any kind of space character, e.g., spaces or tabs.
24562 The rest of the entry, up to the trailing null byte, is the script to
24563 execute in the specified language. The name needs to be unique among
24564 all script names, as @value{GDBN} executes each script only once based
24567 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
24571 #include "symcat.h"
24572 #include "gdb/section-scripts.h"
24574 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
24575 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
24576 ".ascii \"gdb.inlined-script\\n\"\n"
24577 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
24578 ".ascii \" def __init__ (self):\\n\"\n"
24579 ".ascii \" super (test_cmd, self).__init__ ("
24580 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
24581 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
24582 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
24583 ".ascii \"test_cmd ()\\n\"\n"
24589 Loading of inlined scripts requires a properly configured
24590 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24591 The path to specify in @code{auto-load safe-path} is the path of the file
24592 containing the @code{.debug_gdb_scripts} section.
24594 @node Which flavor to choose?
24595 @subsection Which flavor to choose?
24597 Given the multiple ways of auto-loading extensions, it might not always
24598 be clear which one to choose. This section provides some guidance.
24601 Benefits of the @file{-gdb.@var{ext}} way:
24605 Can be used with file formats that don't support multiple sections.
24608 Ease of finding scripts for public libraries.
24610 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
24611 in the source search path.
24612 For publicly installed libraries, e.g., @file{libstdc++}, there typically
24613 isn't a source directory in which to find the script.
24616 Doesn't require source code additions.
24620 Benefits of the @code{.debug_gdb_scripts} way:
24624 Works with static linking.
24626 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
24627 trigger their loading. When an application is statically linked the only
24628 objfile available is the executable, and it is cumbersome to attach all the
24629 scripts from all the input libraries to the executable's
24630 @file{-gdb.@var{ext}} script.
24633 Works with classes that are entirely inlined.
24635 Some classes can be entirely inlined, and thus there may not be an associated
24636 shared library to attach a @file{-gdb.@var{ext}} script to.
24639 Scripts needn't be copied out of the source tree.
24641 In some circumstances, apps can be built out of large collections of internal
24642 libraries, and the build infrastructure necessary to install the
24643 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
24644 cumbersome. It may be easier to specify the scripts in the
24645 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
24646 top of the source tree to the source search path.
24649 @node Multiple Extension Languages
24650 @section Multiple Extension Languages
24652 The Guile and Python extension languages do not share any state,
24653 and generally do not interfere with each other.
24654 There are some things to be aware of, however.
24656 @subsection Python comes first
24658 Python was @value{GDBN}'s first extension language, and to avoid breaking
24659 existing behaviour Python comes first. This is generally solved by the
24660 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
24661 extension languages, and when it makes a call to an extension language,
24662 (say to pretty-print a value), it tries each in turn until an extension
24663 language indicates it has performed the request (e.g., has returned the
24664 pretty-printed form of a value).
24665 This extends to errors while performing such requests: If an error happens
24666 while, for example, trying to pretty-print an object then the error is
24667 reported and any following extension languages are not tried.
24670 @section Creating new spellings of existing commands
24671 @cindex aliases for commands
24673 It is often useful to define alternate spellings of existing commands.
24674 For example, if a new @value{GDBN} command defined in Python has
24675 a long name to type, it is handy to have an abbreviated version of it
24676 that involves less typing.
24678 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
24679 of the @samp{step} command even though it is otherwise an ambiguous
24680 abbreviation of other commands like @samp{set} and @samp{show}.
24682 Aliases are also used to provide shortened or more common versions
24683 of multi-word commands. For example, @value{GDBN} provides the
24684 @samp{tty} alias of the @samp{set inferior-tty} command.
24686 You can define a new alias with the @samp{alias} command.
24691 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
24695 @var{ALIAS} specifies the name of the new alias.
24696 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
24699 @var{COMMAND} specifies the name of an existing command
24700 that is being aliased.
24702 The @samp{-a} option specifies that the new alias is an abbreviation
24703 of the command. Abbreviations are not shown in command
24704 lists displayed by the @samp{help} command.
24706 The @samp{--} option specifies the end of options,
24707 and is useful when @var{ALIAS} begins with a dash.
24709 Here is a simple example showing how to make an abbreviation
24710 of a command so that there is less to type.
24711 Suppose you were tired of typing @samp{disas}, the current
24712 shortest unambiguous abbreviation of the @samp{disassemble} command
24713 and you wanted an even shorter version named @samp{di}.
24714 The following will accomplish this.
24717 (gdb) alias -a di = disas
24720 Note that aliases are different from user-defined commands.
24721 With a user-defined command, you also need to write documentation
24722 for it with the @samp{document} command.
24723 An alias automatically picks up the documentation of the existing command.
24725 Here is an example where we make @samp{elms} an abbreviation of
24726 @samp{elements} in the @samp{set print elements} command.
24727 This is to show that you can make an abbreviation of any part
24731 (gdb) alias -a set print elms = set print elements
24732 (gdb) alias -a show print elms = show print elements
24733 (gdb) set p elms 20
24735 Limit on string chars or array elements to print is 200.
24738 Note that if you are defining an alias of a @samp{set} command,
24739 and you want to have an alias for the corresponding @samp{show}
24740 command, then you need to define the latter separately.
24742 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
24743 @var{ALIAS}, just as they are normally.
24746 (gdb) alias -a set pr elms = set p ele
24749 Finally, here is an example showing the creation of a one word
24750 alias for a more complex command.
24751 This creates alias @samp{spe} of the command @samp{set print elements}.
24754 (gdb) alias spe = set print elements
24759 @chapter Command Interpreters
24760 @cindex command interpreters
24762 @value{GDBN} supports multiple command interpreters, and some command
24763 infrastructure to allow users or user interface writers to switch
24764 between interpreters or run commands in other interpreters.
24766 @value{GDBN} currently supports two command interpreters, the console
24767 interpreter (sometimes called the command-line interpreter or @sc{cli})
24768 and the machine interface interpreter (or @sc{gdb/mi}). This manual
24769 describes both of these interfaces in great detail.
24771 By default, @value{GDBN} will start with the console interpreter.
24772 However, the user may choose to start @value{GDBN} with another
24773 interpreter by specifying the @option{-i} or @option{--interpreter}
24774 startup options. Defined interpreters include:
24778 @cindex console interpreter
24779 The traditional console or command-line interpreter. This is the most often
24780 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
24781 @value{GDBN} will use this interpreter.
24784 @cindex mi interpreter
24785 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
24786 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
24787 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
24791 @cindex mi2 interpreter
24792 The current @sc{gdb/mi} interface.
24795 @cindex mi1 interpreter
24796 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
24800 @cindex invoke another interpreter
24801 The interpreter being used by @value{GDBN} may not be dynamically
24802 switched at runtime. Although possible, this could lead to a very
24803 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
24804 enters the command "interpreter-set console" in a console view,
24805 @value{GDBN} would switch to using the console interpreter, rendering
24806 the IDE inoperable!
24808 @kindex interpreter-exec
24809 Although you may only choose a single interpreter at startup, you may execute
24810 commands in any interpreter from the current interpreter using the appropriate
24811 command. If you are running the console interpreter, simply use the
24812 @code{interpreter-exec} command:
24815 interpreter-exec mi "-data-list-register-names"
24818 @sc{gdb/mi} has a similar command, although it is only available in versions of
24819 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
24822 @chapter @value{GDBN} Text User Interface
24824 @cindex Text User Interface
24827 * TUI Overview:: TUI overview
24828 * TUI Keys:: TUI key bindings
24829 * TUI Single Key Mode:: TUI single key mode
24830 * TUI Commands:: TUI-specific commands
24831 * TUI Configuration:: TUI configuration variables
24834 The @value{GDBN} Text User Interface (TUI) is a terminal
24835 interface which uses the @code{curses} library to show the source
24836 file, the assembly output, the program registers and @value{GDBN}
24837 commands in separate text windows. The TUI mode is supported only
24838 on platforms where a suitable version of the @code{curses} library
24841 The TUI mode is enabled by default when you invoke @value{GDBN} as
24842 @samp{@value{GDBP} -tui}.
24843 You can also switch in and out of TUI mode while @value{GDBN} runs by
24844 using various TUI commands and key bindings, such as @command{tui
24845 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
24846 @ref{TUI Keys, ,TUI Key Bindings}.
24849 @section TUI Overview
24851 In TUI mode, @value{GDBN} can display several text windows:
24855 This window is the @value{GDBN} command window with the @value{GDBN}
24856 prompt and the @value{GDBN} output. The @value{GDBN} input is still
24857 managed using readline.
24860 The source window shows the source file of the program. The current
24861 line and active breakpoints are displayed in this window.
24864 The assembly window shows the disassembly output of the program.
24867 This window shows the processor registers. Registers are highlighted
24868 when their values change.
24871 The source and assembly windows show the current program position
24872 by highlighting the current line and marking it with a @samp{>} marker.
24873 Breakpoints are indicated with two markers. The first marker
24874 indicates the breakpoint type:
24878 Breakpoint which was hit at least once.
24881 Breakpoint which was never hit.
24884 Hardware breakpoint which was hit at least once.
24887 Hardware breakpoint which was never hit.
24890 The second marker indicates whether the breakpoint is enabled or not:
24894 Breakpoint is enabled.
24897 Breakpoint is disabled.
24900 The source, assembly and register windows are updated when the current
24901 thread changes, when the frame changes, or when the program counter
24904 These windows are not all visible at the same time. The command
24905 window is always visible. The others can be arranged in several
24916 source and assembly,
24919 source and registers, or
24922 assembly and registers.
24925 A status line above the command window shows the following information:
24929 Indicates the current @value{GDBN} target.
24930 (@pxref{Targets, ,Specifying a Debugging Target}).
24933 Gives the current process or thread number.
24934 When no process is being debugged, this field is set to @code{No process}.
24937 Gives the current function name for the selected frame.
24938 The name is demangled if demangling is turned on (@pxref{Print Settings}).
24939 When there is no symbol corresponding to the current program counter,
24940 the string @code{??} is displayed.
24943 Indicates the current line number for the selected frame.
24944 When the current line number is not known, the string @code{??} is displayed.
24947 Indicates the current program counter address.
24951 @section TUI Key Bindings
24952 @cindex TUI key bindings
24954 The TUI installs several key bindings in the readline keymaps
24955 @ifset SYSTEM_READLINE
24956 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
24958 @ifclear SYSTEM_READLINE
24959 (@pxref{Command Line Editing}).
24961 The following key bindings are installed for both TUI mode and the
24962 @value{GDBN} standard mode.
24971 Enter or leave the TUI mode. When leaving the TUI mode,
24972 the curses window management stops and @value{GDBN} operates using
24973 its standard mode, writing on the terminal directly. When reentering
24974 the TUI mode, control is given back to the curses windows.
24975 The screen is then refreshed.
24979 Use a TUI layout with only one window. The layout will
24980 either be @samp{source} or @samp{assembly}. When the TUI mode
24981 is not active, it will switch to the TUI mode.
24983 Think of this key binding as the Emacs @kbd{C-x 1} binding.
24987 Use a TUI layout with at least two windows. When the current
24988 layout already has two windows, the next layout with two windows is used.
24989 When a new layout is chosen, one window will always be common to the
24990 previous layout and the new one.
24992 Think of it as the Emacs @kbd{C-x 2} binding.
24996 Change the active window. The TUI associates several key bindings
24997 (like scrolling and arrow keys) with the active window. This command
24998 gives the focus to the next TUI window.
25000 Think of it as the Emacs @kbd{C-x o} binding.
25004 Switch in and out of the TUI SingleKey mode that binds single
25005 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
25008 The following key bindings only work in the TUI mode:
25013 Scroll the active window one page up.
25017 Scroll the active window one page down.
25021 Scroll the active window one line up.
25025 Scroll the active window one line down.
25029 Scroll the active window one column left.
25033 Scroll the active window one column right.
25037 Refresh the screen.
25040 Because the arrow keys scroll the active window in the TUI mode, they
25041 are not available for their normal use by readline unless the command
25042 window has the focus. When another window is active, you must use
25043 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
25044 and @kbd{C-f} to control the command window.
25046 @node TUI Single Key Mode
25047 @section TUI Single Key Mode
25048 @cindex TUI single key mode
25050 The TUI also provides a @dfn{SingleKey} mode, which binds several
25051 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
25052 switch into this mode, where the following key bindings are used:
25055 @kindex c @r{(SingleKey TUI key)}
25059 @kindex d @r{(SingleKey TUI key)}
25063 @kindex f @r{(SingleKey TUI key)}
25067 @kindex n @r{(SingleKey TUI key)}
25071 @kindex q @r{(SingleKey TUI key)}
25073 exit the SingleKey mode.
25075 @kindex r @r{(SingleKey TUI key)}
25079 @kindex s @r{(SingleKey TUI key)}
25083 @kindex u @r{(SingleKey TUI key)}
25087 @kindex v @r{(SingleKey TUI key)}
25091 @kindex w @r{(SingleKey TUI key)}
25096 Other keys temporarily switch to the @value{GDBN} command prompt.
25097 The key that was pressed is inserted in the editing buffer so that
25098 it is possible to type most @value{GDBN} commands without interaction
25099 with the TUI SingleKey mode. Once the command is entered the TUI
25100 SingleKey mode is restored. The only way to permanently leave
25101 this mode is by typing @kbd{q} or @kbd{C-x s}.
25105 @section TUI-specific Commands
25106 @cindex TUI commands
25108 The TUI has specific commands to control the text windows.
25109 These commands are always available, even when @value{GDBN} is not in
25110 the TUI mode. When @value{GDBN} is in the standard mode, most
25111 of these commands will automatically switch to the TUI mode.
25113 Note that if @value{GDBN}'s @code{stdout} is not connected to a
25114 terminal, or @value{GDBN} has been started with the machine interface
25115 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
25116 these commands will fail with an error, because it would not be
25117 possible or desirable to enable curses window management.
25122 Activate TUI mode. The last active TUI window layout will be used if
25123 TUI mode has prevsiouly been used in the current debugging session,
25124 otherwise a default layout is used.
25127 @kindex tui disable
25128 Disable TUI mode, returning to the console interpreter.
25132 List and give the size of all displayed windows.
25134 @item layout @var{name}
25136 Changes which TUI windows are displayed. In each layout the command
25137 window is always displayed, the @var{name} parameter controls which
25138 additional windows are displayed, and can be any of the following:
25142 Display the next layout.
25145 Display the previous layout.
25148 Display the source and command windows.
25151 Display the assembly and command windows.
25154 Display the source, assembly, and command windows.
25157 When in @code{src} layout display the register, source, and command
25158 windows. When in @code{asm} or @code{split} layout display the
25159 register, assembler, and command windows.
25162 @item focus @var{name}
25164 Changes which TUI window is currently active for scrolling. The
25165 @var{name} parameter can be any of the following:
25169 Make the next window active for scrolling.
25172 Make the previous window active for scrolling.
25175 Make the source window active for scrolling.
25178 Make the assembly window active for scrolling.
25181 Make the register window active for scrolling.
25184 Make the command window active for scrolling.
25189 Refresh the screen. This is similar to typing @kbd{C-L}.
25191 @item tui reg @var{group}
25193 Changes the register group displayed in the tui register window to
25194 @var{group}. If the register window is not currently displayed this
25195 command will cause the register window to be displayed. The list of
25196 register groups, as well as their order is target specific. The
25197 following groups are available on most targets:
25200 Repeatedly selecting this group will cause the display to cycle
25201 through all of the available register groups.
25204 Repeatedly selecting this group will cause the display to cycle
25205 through all of the available register groups in the reverse order to
25209 Display the general registers.
25211 Display the floating point registers.
25213 Display the system registers.
25215 Display the vector registers.
25217 Display all registers.
25222 Update the source window and the current execution point.
25224 @item winheight @var{name} +@var{count}
25225 @itemx winheight @var{name} -@var{count}
25227 Change the height of the window @var{name} by @var{count}
25228 lines. Positive counts increase the height, while negative counts
25229 decrease it. The @var{name} parameter can be one of @code{src} (the
25230 source window), @code{cmd} (the command window), @code{asm} (the
25231 disassembly window), or @code{regs} (the register display window).
25233 @item tabset @var{nchars}
25235 Set the width of tab stops to be @var{nchars} characters. This
25236 setting affects the display of TAB characters in the source and
25240 @node TUI Configuration
25241 @section TUI Configuration Variables
25242 @cindex TUI configuration variables
25244 Several configuration variables control the appearance of TUI windows.
25247 @item set tui border-kind @var{kind}
25248 @kindex set tui border-kind
25249 Select the border appearance for the source, assembly and register windows.
25250 The possible values are the following:
25253 Use a space character to draw the border.
25256 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
25259 Use the Alternate Character Set to draw the border. The border is
25260 drawn using character line graphics if the terminal supports them.
25263 @item set tui border-mode @var{mode}
25264 @kindex set tui border-mode
25265 @itemx set tui active-border-mode @var{mode}
25266 @kindex set tui active-border-mode
25267 Select the display attributes for the borders of the inactive windows
25268 or the active window. The @var{mode} can be one of the following:
25271 Use normal attributes to display the border.
25277 Use reverse video mode.
25280 Use half bright mode.
25282 @item half-standout
25283 Use half bright and standout mode.
25286 Use extra bright or bold mode.
25288 @item bold-standout
25289 Use extra bright or bold and standout mode.
25294 @chapter Using @value{GDBN} under @sc{gnu} Emacs
25297 @cindex @sc{gnu} Emacs
25298 A special interface allows you to use @sc{gnu} Emacs to view (and
25299 edit) the source files for the program you are debugging with
25302 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
25303 executable file you want to debug as an argument. This command starts
25304 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
25305 created Emacs buffer.
25306 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
25308 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
25313 All ``terminal'' input and output goes through an Emacs buffer, called
25316 This applies both to @value{GDBN} commands and their output, and to the input
25317 and output done by the program you are debugging.
25319 This is useful because it means that you can copy the text of previous
25320 commands and input them again; you can even use parts of the output
25323 All the facilities of Emacs' Shell mode are available for interacting
25324 with your program. In particular, you can send signals the usual
25325 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
25329 @value{GDBN} displays source code through Emacs.
25331 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
25332 source file for that frame and puts an arrow (@samp{=>}) at the
25333 left margin of the current line. Emacs uses a separate buffer for
25334 source display, and splits the screen to show both your @value{GDBN} session
25337 Explicit @value{GDBN} @code{list} or search commands still produce output as
25338 usual, but you probably have no reason to use them from Emacs.
25341 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
25342 a graphical mode, enabled by default, which provides further buffers
25343 that can control the execution and describe the state of your program.
25344 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
25346 If you specify an absolute file name when prompted for the @kbd{M-x
25347 gdb} argument, then Emacs sets your current working directory to where
25348 your program resides. If you only specify the file name, then Emacs
25349 sets your current working directory to the directory associated
25350 with the previous buffer. In this case, @value{GDBN} may find your
25351 program by searching your environment's @code{PATH} variable, but on
25352 some operating systems it might not find the source. So, although the
25353 @value{GDBN} input and output session proceeds normally, the auxiliary
25354 buffer does not display the current source and line of execution.
25356 The initial working directory of @value{GDBN} is printed on the top
25357 line of the GUD buffer and this serves as a default for the commands
25358 that specify files for @value{GDBN} to operate on. @xref{Files,
25359 ,Commands to Specify Files}.
25361 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
25362 need to call @value{GDBN} by a different name (for example, if you
25363 keep several configurations around, with different names) you can
25364 customize the Emacs variable @code{gud-gdb-command-name} to run the
25367 In the GUD buffer, you can use these special Emacs commands in
25368 addition to the standard Shell mode commands:
25372 Describe the features of Emacs' GUD Mode.
25375 Execute to another source line, like the @value{GDBN} @code{step} command; also
25376 update the display window to show the current file and location.
25379 Execute to next source line in this function, skipping all function
25380 calls, like the @value{GDBN} @code{next} command. Then update the display window
25381 to show the current file and location.
25384 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
25385 display window accordingly.
25388 Execute until exit from the selected stack frame, like the @value{GDBN}
25389 @code{finish} command.
25392 Continue execution of your program, like the @value{GDBN} @code{continue}
25396 Go up the number of frames indicated by the numeric argument
25397 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
25398 like the @value{GDBN} @code{up} command.
25401 Go down the number of frames indicated by the numeric argument, like the
25402 @value{GDBN} @code{down} command.
25405 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
25406 tells @value{GDBN} to set a breakpoint on the source line point is on.
25408 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
25409 separate frame which shows a backtrace when the GUD buffer is current.
25410 Move point to any frame in the stack and type @key{RET} to make it
25411 become the current frame and display the associated source in the
25412 source buffer. Alternatively, click @kbd{Mouse-2} to make the
25413 selected frame become the current one. In graphical mode, the
25414 speedbar displays watch expressions.
25416 If you accidentally delete the source-display buffer, an easy way to get
25417 it back is to type the command @code{f} in the @value{GDBN} buffer, to
25418 request a frame display; when you run under Emacs, this recreates
25419 the source buffer if necessary to show you the context of the current
25422 The source files displayed in Emacs are in ordinary Emacs buffers
25423 which are visiting the source files in the usual way. You can edit
25424 the files with these buffers if you wish; but keep in mind that @value{GDBN}
25425 communicates with Emacs in terms of line numbers. If you add or
25426 delete lines from the text, the line numbers that @value{GDBN} knows cease
25427 to correspond properly with the code.
25429 A more detailed description of Emacs' interaction with @value{GDBN} is
25430 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
25434 @chapter The @sc{gdb/mi} Interface
25436 @unnumberedsec Function and Purpose
25438 @cindex @sc{gdb/mi}, its purpose
25439 @sc{gdb/mi} is a line based machine oriented text interface to
25440 @value{GDBN} and is activated by specifying using the
25441 @option{--interpreter} command line option (@pxref{Mode Options}). It
25442 is specifically intended to support the development of systems which
25443 use the debugger as just one small component of a larger system.
25445 This chapter is a specification of the @sc{gdb/mi} interface. It is written
25446 in the form of a reference manual.
25448 Note that @sc{gdb/mi} is still under construction, so some of the
25449 features described below are incomplete and subject to change
25450 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
25452 @unnumberedsec Notation and Terminology
25454 @cindex notational conventions, for @sc{gdb/mi}
25455 This chapter uses the following notation:
25459 @code{|} separates two alternatives.
25462 @code{[ @var{something} ]} indicates that @var{something} is optional:
25463 it may or may not be given.
25466 @code{( @var{group} )*} means that @var{group} inside the parentheses
25467 may repeat zero or more times.
25470 @code{( @var{group} )+} means that @var{group} inside the parentheses
25471 may repeat one or more times.
25474 @code{"@var{string}"} means a literal @var{string}.
25478 @heading Dependencies
25482 * GDB/MI General Design::
25483 * GDB/MI Command Syntax::
25484 * GDB/MI Compatibility with CLI::
25485 * GDB/MI Development and Front Ends::
25486 * GDB/MI Output Records::
25487 * GDB/MI Simple Examples::
25488 * GDB/MI Command Description Format::
25489 * GDB/MI Breakpoint Commands::
25490 * GDB/MI Catchpoint Commands::
25491 * GDB/MI Program Context::
25492 * GDB/MI Thread Commands::
25493 * GDB/MI Ada Tasking Commands::
25494 * GDB/MI Program Execution::
25495 * GDB/MI Stack Manipulation::
25496 * GDB/MI Variable Objects::
25497 * GDB/MI Data Manipulation::
25498 * GDB/MI Tracepoint Commands::
25499 * GDB/MI Symbol Query::
25500 * GDB/MI File Commands::
25502 * GDB/MI Kod Commands::
25503 * GDB/MI Memory Overlay Commands::
25504 * GDB/MI Signal Handling Commands::
25506 * GDB/MI Target Manipulation::
25507 * GDB/MI File Transfer Commands::
25508 * GDB/MI Ada Exceptions Commands::
25509 * GDB/MI Support Commands::
25510 * GDB/MI Miscellaneous Commands::
25513 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25514 @node GDB/MI General Design
25515 @section @sc{gdb/mi} General Design
25516 @cindex GDB/MI General Design
25518 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
25519 parts---commands sent to @value{GDBN}, responses to those commands
25520 and notifications. Each command results in exactly one response,
25521 indicating either successful completion of the command, or an error.
25522 For the commands that do not resume the target, the response contains the
25523 requested information. For the commands that resume the target, the
25524 response only indicates whether the target was successfully resumed.
25525 Notifications is the mechanism for reporting changes in the state of the
25526 target, or in @value{GDBN} state, that cannot conveniently be associated with
25527 a command and reported as part of that command response.
25529 The important examples of notifications are:
25533 Exec notifications. These are used to report changes in
25534 target state---when a target is resumed, or stopped. It would not
25535 be feasible to include this information in response of resuming
25536 commands, because one resume commands can result in multiple events in
25537 different threads. Also, quite some time may pass before any event
25538 happens in the target, while a frontend needs to know whether the resuming
25539 command itself was successfully executed.
25542 Console output, and status notifications. Console output
25543 notifications are used to report output of CLI commands, as well as
25544 diagnostics for other commands. Status notifications are used to
25545 report the progress of a long-running operation. Naturally, including
25546 this information in command response would mean no output is produced
25547 until the command is finished, which is undesirable.
25550 General notifications. Commands may have various side effects on
25551 the @value{GDBN} or target state beyond their official purpose. For example,
25552 a command may change the selected thread. Although such changes can
25553 be included in command response, using notification allows for more
25554 orthogonal frontend design.
25558 There's no guarantee that whenever an MI command reports an error,
25559 @value{GDBN} or the target are in any specific state, and especially,
25560 the state is not reverted to the state before the MI command was
25561 processed. Therefore, whenever an MI command results in an error,
25562 we recommend that the frontend refreshes all the information shown in
25563 the user interface.
25567 * Context management::
25568 * Asynchronous and non-stop modes::
25572 @node Context management
25573 @subsection Context management
25575 @subsubsection Threads and Frames
25577 In most cases when @value{GDBN} accesses the target, this access is
25578 done in context of a specific thread and frame (@pxref{Frames}).
25579 Often, even when accessing global data, the target requires that a thread
25580 be specified. The CLI interface maintains the selected thread and frame,
25581 and supplies them to target on each command. This is convenient,
25582 because a command line user would not want to specify that information
25583 explicitly on each command, and because user interacts with
25584 @value{GDBN} via a single terminal, so no confusion is possible as
25585 to what thread and frame are the current ones.
25587 In the case of MI, the concept of selected thread and frame is less
25588 useful. First, a frontend can easily remember this information
25589 itself. Second, a graphical frontend can have more than one window,
25590 each one used for debugging a different thread, and the frontend might
25591 want to access additional threads for internal purposes. This
25592 increases the risk that by relying on implicitly selected thread, the
25593 frontend may be operating on a wrong one. Therefore, each MI command
25594 should explicitly specify which thread and frame to operate on. To
25595 make it possible, each MI command accepts the @samp{--thread} and
25596 @samp{--frame} options, the value to each is @value{GDBN} global
25597 identifier for thread and frame to operate on.
25599 Usually, each top-level window in a frontend allows the user to select
25600 a thread and a frame, and remembers the user selection for further
25601 operations. However, in some cases @value{GDBN} may suggest that the
25602 current thread be changed. For example, when stopping on a breakpoint
25603 it is reasonable to switch to the thread where breakpoint is hit. For
25604 another example, if the user issues the CLI @samp{thread} command via
25605 the frontend, it is desirable to change the frontend's selected thread to the
25606 one specified by user. @value{GDBN} communicates the suggestion to
25607 change current thread using the @samp{=thread-selected} notification.
25608 No such notification is available for the selected frame at the moment.
25610 Note that historically, MI shares the selected thread with CLI, so
25611 frontends used the @code{-thread-select} to execute commands in the
25612 right context. However, getting this to work right is cumbersome. The
25613 simplest way is for frontend to emit @code{-thread-select} command
25614 before every command. This doubles the number of commands that need
25615 to be sent. The alternative approach is to suppress @code{-thread-select}
25616 if the selected thread in @value{GDBN} is supposed to be identical to the
25617 thread the frontend wants to operate on. However, getting this
25618 optimization right can be tricky. In particular, if the frontend
25619 sends several commands to @value{GDBN}, and one of the commands changes the
25620 selected thread, then the behaviour of subsequent commands will
25621 change. So, a frontend should either wait for response from such
25622 problematic commands, or explicitly add @code{-thread-select} for
25623 all subsequent commands. No frontend is known to do this exactly
25624 right, so it is suggested to just always pass the @samp{--thread} and
25625 @samp{--frame} options.
25627 @subsubsection Language
25629 The execution of several commands depends on which language is selected.
25630 By default, the current language (@pxref{show language}) is used.
25631 But for commands known to be language-sensitive, it is recommended
25632 to use the @samp{--language} option. This option takes one argument,
25633 which is the name of the language to use while executing the command.
25637 -data-evaluate-expression --language c "sizeof (void*)"
25642 The valid language names are the same names accepted by the
25643 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
25644 @samp{local} or @samp{unknown}.
25646 @node Asynchronous and non-stop modes
25647 @subsection Asynchronous command execution and non-stop mode
25649 On some targets, @value{GDBN} is capable of processing MI commands
25650 even while the target is running. This is called @dfn{asynchronous
25651 command execution} (@pxref{Background Execution}). The frontend may
25652 specify a preferrence for asynchronous execution using the
25653 @code{-gdb-set mi-async 1} command, which should be emitted before
25654 either running the executable or attaching to the target. After the
25655 frontend has started the executable or attached to the target, it can
25656 find if asynchronous execution is enabled using the
25657 @code{-list-target-features} command.
25660 @item -gdb-set mi-async on
25661 @item -gdb-set mi-async off
25662 Set whether MI is in asynchronous mode.
25664 When @code{off}, which is the default, MI execution commands (e.g.,
25665 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
25666 for the program to stop before processing further commands.
25668 When @code{on}, MI execution commands are background execution
25669 commands (e.g., @code{-exec-continue} becomes the equivalent of the
25670 @code{c&} CLI command), and so @value{GDBN} is capable of processing
25671 MI commands even while the target is running.
25673 @item -gdb-show mi-async
25674 Show whether MI asynchronous mode is enabled.
25677 Note: In @value{GDBN} version 7.7 and earlier, this option was called
25678 @code{target-async} instead of @code{mi-async}, and it had the effect
25679 of both putting MI in asynchronous mode and making CLI background
25680 commands possible. CLI background commands are now always possible
25681 ``out of the box'' if the target supports them. The old spelling is
25682 kept as a deprecated alias for backwards compatibility.
25684 Even if @value{GDBN} can accept a command while target is running,
25685 many commands that access the target do not work when the target is
25686 running. Therefore, asynchronous command execution is most useful
25687 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
25688 it is possible to examine the state of one thread, while other threads
25691 When a given thread is running, MI commands that try to access the
25692 target in the context of that thread may not work, or may work only on
25693 some targets. In particular, commands that try to operate on thread's
25694 stack will not work, on any target. Commands that read memory, or
25695 modify breakpoints, may work or not work, depending on the target. Note
25696 that even commands that operate on global state, such as @code{print},
25697 @code{set}, and breakpoint commands, still access the target in the
25698 context of a specific thread, so frontend should try to find a
25699 stopped thread and perform the operation on that thread (using the
25700 @samp{--thread} option).
25702 Which commands will work in the context of a running thread is
25703 highly target dependent. However, the two commands
25704 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
25705 to find the state of a thread, will always work.
25707 @node Thread groups
25708 @subsection Thread groups
25709 @value{GDBN} may be used to debug several processes at the same time.
25710 On some platfroms, @value{GDBN} may support debugging of several
25711 hardware systems, each one having several cores with several different
25712 processes running on each core. This section describes the MI
25713 mechanism to support such debugging scenarios.
25715 The key observation is that regardless of the structure of the
25716 target, MI can have a global list of threads, because most commands that
25717 accept the @samp{--thread} option do not need to know what process that
25718 thread belongs to. Therefore, it is not necessary to introduce
25719 neither additional @samp{--process} option, nor an notion of the
25720 current process in the MI interface. The only strictly new feature
25721 that is required is the ability to find how the threads are grouped
25724 To allow the user to discover such grouping, and to support arbitrary
25725 hierarchy of machines/cores/processes, MI introduces the concept of a
25726 @dfn{thread group}. Thread group is a collection of threads and other
25727 thread groups. A thread group always has a string identifier, a type,
25728 and may have additional attributes specific to the type. A new
25729 command, @code{-list-thread-groups}, returns the list of top-level
25730 thread groups, which correspond to processes that @value{GDBN} is
25731 debugging at the moment. By passing an identifier of a thread group
25732 to the @code{-list-thread-groups} command, it is possible to obtain
25733 the members of specific thread group.
25735 To allow the user to easily discover processes, and other objects, he
25736 wishes to debug, a concept of @dfn{available thread group} is
25737 introduced. Available thread group is an thread group that
25738 @value{GDBN} is not debugging, but that can be attached to, using the
25739 @code{-target-attach} command. The list of available top-level thread
25740 groups can be obtained using @samp{-list-thread-groups --available}.
25741 In general, the content of a thread group may be only retrieved only
25742 after attaching to that thread group.
25744 Thread groups are related to inferiors (@pxref{Inferiors and
25745 Programs}). Each inferior corresponds to a thread group of a special
25746 type @samp{process}, and some additional operations are permitted on
25747 such thread groups.
25749 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25750 @node GDB/MI Command Syntax
25751 @section @sc{gdb/mi} Command Syntax
25754 * GDB/MI Input Syntax::
25755 * GDB/MI Output Syntax::
25758 @node GDB/MI Input Syntax
25759 @subsection @sc{gdb/mi} Input Syntax
25761 @cindex input syntax for @sc{gdb/mi}
25762 @cindex @sc{gdb/mi}, input syntax
25764 @item @var{command} @expansion{}
25765 @code{@var{cli-command} | @var{mi-command}}
25767 @item @var{cli-command} @expansion{}
25768 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
25769 @var{cli-command} is any existing @value{GDBN} CLI command.
25771 @item @var{mi-command} @expansion{}
25772 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
25773 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
25775 @item @var{token} @expansion{}
25776 "any sequence of digits"
25778 @item @var{option} @expansion{}
25779 @code{"-" @var{parameter} [ " " @var{parameter} ]}
25781 @item @var{parameter} @expansion{}
25782 @code{@var{non-blank-sequence} | @var{c-string}}
25784 @item @var{operation} @expansion{}
25785 @emph{any of the operations described in this chapter}
25787 @item @var{non-blank-sequence} @expansion{}
25788 @emph{anything, provided it doesn't contain special characters such as
25789 "-", @var{nl}, """ and of course " "}
25791 @item @var{c-string} @expansion{}
25792 @code{""" @var{seven-bit-iso-c-string-content} """}
25794 @item @var{nl} @expansion{}
25803 The CLI commands are still handled by the @sc{mi} interpreter; their
25804 output is described below.
25807 The @code{@var{token}}, when present, is passed back when the command
25811 Some @sc{mi} commands accept optional arguments as part of the parameter
25812 list. Each option is identified by a leading @samp{-} (dash) and may be
25813 followed by an optional argument parameter. Options occur first in the
25814 parameter list and can be delimited from normal parameters using
25815 @samp{--} (this is useful when some parameters begin with a dash).
25822 We want easy access to the existing CLI syntax (for debugging).
25825 We want it to be easy to spot a @sc{mi} operation.
25828 @node GDB/MI Output Syntax
25829 @subsection @sc{gdb/mi} Output Syntax
25831 @cindex output syntax of @sc{gdb/mi}
25832 @cindex @sc{gdb/mi}, output syntax
25833 The output from @sc{gdb/mi} consists of zero or more out-of-band records
25834 followed, optionally, by a single result record. This result record
25835 is for the most recent command. The sequence of output records is
25836 terminated by @samp{(gdb)}.
25838 If an input command was prefixed with a @code{@var{token}} then the
25839 corresponding output for that command will also be prefixed by that same
25843 @item @var{output} @expansion{}
25844 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
25846 @item @var{result-record} @expansion{}
25847 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
25849 @item @var{out-of-band-record} @expansion{}
25850 @code{@var{async-record} | @var{stream-record}}
25852 @item @var{async-record} @expansion{}
25853 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
25855 @item @var{exec-async-output} @expansion{}
25856 @code{[ @var{token} ] "*" @var{async-output nl}}
25858 @item @var{status-async-output} @expansion{}
25859 @code{[ @var{token} ] "+" @var{async-output nl}}
25861 @item @var{notify-async-output} @expansion{}
25862 @code{[ @var{token} ] "=" @var{async-output nl}}
25864 @item @var{async-output} @expansion{}
25865 @code{@var{async-class} ( "," @var{result} )*}
25867 @item @var{result-class} @expansion{}
25868 @code{"done" | "running" | "connected" | "error" | "exit"}
25870 @item @var{async-class} @expansion{}
25871 @code{"stopped" | @var{others}} (where @var{others} will be added
25872 depending on the needs---this is still in development).
25874 @item @var{result} @expansion{}
25875 @code{ @var{variable} "=" @var{value}}
25877 @item @var{variable} @expansion{}
25878 @code{ @var{string} }
25880 @item @var{value} @expansion{}
25881 @code{ @var{const} | @var{tuple} | @var{list} }
25883 @item @var{const} @expansion{}
25884 @code{@var{c-string}}
25886 @item @var{tuple} @expansion{}
25887 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
25889 @item @var{list} @expansion{}
25890 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
25891 @var{result} ( "," @var{result} )* "]" }
25893 @item @var{stream-record} @expansion{}
25894 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
25896 @item @var{console-stream-output} @expansion{}
25897 @code{"~" @var{c-string nl}}
25899 @item @var{target-stream-output} @expansion{}
25900 @code{"@@" @var{c-string nl}}
25902 @item @var{log-stream-output} @expansion{}
25903 @code{"&" @var{c-string nl}}
25905 @item @var{nl} @expansion{}
25908 @item @var{token} @expansion{}
25909 @emph{any sequence of digits}.
25917 All output sequences end in a single line containing a period.
25920 The @code{@var{token}} is from the corresponding request. Note that
25921 for all async output, while the token is allowed by the grammar and
25922 may be output by future versions of @value{GDBN} for select async
25923 output messages, it is generally omitted. Frontends should treat
25924 all async output as reporting general changes in the state of the
25925 target and there should be no need to associate async output to any
25929 @cindex status output in @sc{gdb/mi}
25930 @var{status-async-output} contains on-going status information about the
25931 progress of a slow operation. It can be discarded. All status output is
25932 prefixed by @samp{+}.
25935 @cindex async output in @sc{gdb/mi}
25936 @var{exec-async-output} contains asynchronous state change on the target
25937 (stopped, started, disappeared). All async output is prefixed by
25941 @cindex notify output in @sc{gdb/mi}
25942 @var{notify-async-output} contains supplementary information that the
25943 client should handle (e.g., a new breakpoint information). All notify
25944 output is prefixed by @samp{=}.
25947 @cindex console output in @sc{gdb/mi}
25948 @var{console-stream-output} is output that should be displayed as is in the
25949 console. It is the textual response to a CLI command. All the console
25950 output is prefixed by @samp{~}.
25953 @cindex target output in @sc{gdb/mi}
25954 @var{target-stream-output} is the output produced by the target program.
25955 All the target output is prefixed by @samp{@@}.
25958 @cindex log output in @sc{gdb/mi}
25959 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
25960 instance messages that should be displayed as part of an error log. All
25961 the log output is prefixed by @samp{&}.
25964 @cindex list output in @sc{gdb/mi}
25965 New @sc{gdb/mi} commands should only output @var{lists} containing
25971 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
25972 details about the various output records.
25974 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25975 @node GDB/MI Compatibility with CLI
25976 @section @sc{gdb/mi} Compatibility with CLI
25978 @cindex compatibility, @sc{gdb/mi} and CLI
25979 @cindex @sc{gdb/mi}, compatibility with CLI
25981 For the developers convenience CLI commands can be entered directly,
25982 but there may be some unexpected behaviour. For example, commands
25983 that query the user will behave as if the user replied yes, breakpoint
25984 command lists are not executed and some CLI commands, such as
25985 @code{if}, @code{when} and @code{define}, prompt for further input with
25986 @samp{>}, which is not valid MI output.
25988 This feature may be removed at some stage in the future and it is
25989 recommended that front ends use the @code{-interpreter-exec} command
25990 (@pxref{-interpreter-exec}).
25992 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25993 @node GDB/MI Development and Front Ends
25994 @section @sc{gdb/mi} Development and Front Ends
25995 @cindex @sc{gdb/mi} development
25997 The application which takes the MI output and presents the state of the
25998 program being debugged to the user is called a @dfn{front end}.
26000 Although @sc{gdb/mi} is still incomplete, it is currently being used
26001 by a variety of front ends to @value{GDBN}. This makes it difficult
26002 to introduce new functionality without breaking existing usage. This
26003 section tries to minimize the problems by describing how the protocol
26006 Some changes in MI need not break a carefully designed front end, and
26007 for these the MI version will remain unchanged. The following is a
26008 list of changes that may occur within one level, so front ends should
26009 parse MI output in a way that can handle them:
26013 New MI commands may be added.
26016 New fields may be added to the output of any MI command.
26019 The range of values for fields with specified values, e.g.,
26020 @code{in_scope} (@pxref{-var-update}) may be extended.
26022 @c The format of field's content e.g type prefix, may change so parse it
26023 @c at your own risk. Yes, in general?
26025 @c The order of fields may change? Shouldn't really matter but it might
26026 @c resolve inconsistencies.
26029 If the changes are likely to break front ends, the MI version level
26030 will be increased by one. This will allow the front end to parse the
26031 output according to the MI version. Apart from mi0, new versions of
26032 @value{GDBN} will not support old versions of MI and it will be the
26033 responsibility of the front end to work with the new one.
26035 @c Starting with mi3, add a new command -mi-version that prints the MI
26038 The best way to avoid unexpected changes in MI that might break your front
26039 end is to make your project known to @value{GDBN} developers and
26040 follow development on @email{gdb@@sourceware.org} and
26041 @email{gdb-patches@@sourceware.org}.
26042 @cindex mailing lists
26044 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26045 @node GDB/MI Output Records
26046 @section @sc{gdb/mi} Output Records
26049 * GDB/MI Result Records::
26050 * GDB/MI Stream Records::
26051 * GDB/MI Async Records::
26052 * GDB/MI Breakpoint Information::
26053 * GDB/MI Frame Information::
26054 * GDB/MI Thread Information::
26055 * GDB/MI Ada Exception Information::
26058 @node GDB/MI Result Records
26059 @subsection @sc{gdb/mi} Result Records
26061 @cindex result records in @sc{gdb/mi}
26062 @cindex @sc{gdb/mi}, result records
26063 In addition to a number of out-of-band notifications, the response to a
26064 @sc{gdb/mi} command includes one of the following result indications:
26068 @item "^done" [ "," @var{results} ]
26069 The synchronous operation was successful, @code{@var{results}} are the return
26074 This result record is equivalent to @samp{^done}. Historically, it
26075 was output instead of @samp{^done} if the command has resumed the
26076 target. This behaviour is maintained for backward compatibility, but
26077 all frontends should treat @samp{^done} and @samp{^running}
26078 identically and rely on the @samp{*running} output record to determine
26079 which threads are resumed.
26083 @value{GDBN} has connected to a remote target.
26085 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
26087 The operation failed. The @code{msg=@var{c-string}} variable contains
26088 the corresponding error message.
26090 If present, the @code{code=@var{c-string}} variable provides an error
26091 code on which consumers can rely on to detect the corresponding
26092 error condition. At present, only one error code is defined:
26095 @item "undefined-command"
26096 Indicates that the command causing the error does not exist.
26101 @value{GDBN} has terminated.
26105 @node GDB/MI Stream Records
26106 @subsection @sc{gdb/mi} Stream Records
26108 @cindex @sc{gdb/mi}, stream records
26109 @cindex stream records in @sc{gdb/mi}
26110 @value{GDBN} internally maintains a number of output streams: the console, the
26111 target, and the log. The output intended for each of these streams is
26112 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
26114 Each stream record begins with a unique @dfn{prefix character} which
26115 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
26116 Syntax}). In addition to the prefix, each stream record contains a
26117 @code{@var{string-output}}. This is either raw text (with an implicit new
26118 line) or a quoted C string (which does not contain an implicit newline).
26121 @item "~" @var{string-output}
26122 The console output stream contains text that should be displayed in the
26123 CLI console window. It contains the textual responses to CLI commands.
26125 @item "@@" @var{string-output}
26126 The target output stream contains any textual output from the running
26127 target. This is only present when GDB's event loop is truly
26128 asynchronous, which is currently only the case for remote targets.
26130 @item "&" @var{string-output}
26131 The log stream contains debugging messages being produced by @value{GDBN}'s
26135 @node GDB/MI Async Records
26136 @subsection @sc{gdb/mi} Async Records
26138 @cindex async records in @sc{gdb/mi}
26139 @cindex @sc{gdb/mi}, async records
26140 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
26141 additional changes that have occurred. Those changes can either be a
26142 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
26143 target activity (e.g., target stopped).
26145 The following is the list of possible async records:
26149 @item *running,thread-id="@var{thread}"
26150 The target is now running. The @var{thread} field can be the global
26151 thread ID of the the thread that is now running, and it can be
26152 @samp{all} if all threads are running. The frontend should assume
26153 that no interaction with a running thread is possible after this
26154 notification is produced. The frontend should not assume that this
26155 notification is output only once for any command. @value{GDBN} may
26156 emit this notification several times, either for different threads,
26157 because it cannot resume all threads together, or even for a single
26158 thread, if the thread must be stepped though some code before letting
26161 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
26162 The target has stopped. The @var{reason} field can have one of the
26166 @item breakpoint-hit
26167 A breakpoint was reached.
26168 @item watchpoint-trigger
26169 A watchpoint was triggered.
26170 @item read-watchpoint-trigger
26171 A read watchpoint was triggered.
26172 @item access-watchpoint-trigger
26173 An access watchpoint was triggered.
26174 @item function-finished
26175 An -exec-finish or similar CLI command was accomplished.
26176 @item location-reached
26177 An -exec-until or similar CLI command was accomplished.
26178 @item watchpoint-scope
26179 A watchpoint has gone out of scope.
26180 @item end-stepping-range
26181 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
26182 similar CLI command was accomplished.
26183 @item exited-signalled
26184 The inferior exited because of a signal.
26186 The inferior exited.
26187 @item exited-normally
26188 The inferior exited normally.
26189 @item signal-received
26190 A signal was received by the inferior.
26192 The inferior has stopped due to a library being loaded or unloaded.
26193 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
26194 set or when a @code{catch load} or @code{catch unload} catchpoint is
26195 in use (@pxref{Set Catchpoints}).
26197 The inferior has forked. This is reported when @code{catch fork}
26198 (@pxref{Set Catchpoints}) has been used.
26200 The inferior has vforked. This is reported in when @code{catch vfork}
26201 (@pxref{Set Catchpoints}) has been used.
26202 @item syscall-entry
26203 The inferior entered a system call. This is reported when @code{catch
26204 syscall} (@pxref{Set Catchpoints}) has been used.
26205 @item syscall-return
26206 The inferior returned from a system call. This is reported when
26207 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
26209 The inferior called @code{exec}. This is reported when @code{catch exec}
26210 (@pxref{Set Catchpoints}) has been used.
26213 The @var{id} field identifies the global thread ID of the thread
26214 that directly caused the stop -- for example by hitting a breakpoint.
26215 Depending on whether all-stop
26216 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
26217 stop all threads, or only the thread that directly triggered the stop.
26218 If all threads are stopped, the @var{stopped} field will have the
26219 value of @code{"all"}. Otherwise, the value of the @var{stopped}
26220 field will be a list of thread identifiers. Presently, this list will
26221 always include a single thread, but frontend should be prepared to see
26222 several threads in the list. The @var{core} field reports the
26223 processor core on which the stop event has happened. This field may be absent
26224 if such information is not available.
26226 @item =thread-group-added,id="@var{id}"
26227 @itemx =thread-group-removed,id="@var{id}"
26228 A thread group was either added or removed. The @var{id} field
26229 contains the @value{GDBN} identifier of the thread group. When a thread
26230 group is added, it generally might not be associated with a running
26231 process. When a thread group is removed, its id becomes invalid and
26232 cannot be used in any way.
26234 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
26235 A thread group became associated with a running program,
26236 either because the program was just started or the thread group
26237 was attached to a program. The @var{id} field contains the
26238 @value{GDBN} identifier of the thread group. The @var{pid} field
26239 contains process identifier, specific to the operating system.
26241 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
26242 A thread group is no longer associated with a running program,
26243 either because the program has exited, or because it was detached
26244 from. The @var{id} field contains the @value{GDBN} identifier of the
26245 thread group. The @var{code} field is the exit code of the inferior; it exists
26246 only when the inferior exited with some code.
26248 @item =thread-created,id="@var{id}",group-id="@var{gid}"
26249 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
26250 A thread either was created, or has exited. The @var{id} field
26251 contains the global @value{GDBN} identifier of the thread. The @var{gid}
26252 field identifies the thread group this thread belongs to.
26254 @item =thread-selected,id="@var{id}"
26255 Informs that the selected thread was changed as result of the last
26256 command. This notification is not emitted as result of @code{-thread-select}
26257 command but is emitted whenever an MI command that is not documented
26258 to change the selected thread actually changes it. In particular,
26259 invoking, directly or indirectly (via user-defined command), the CLI
26260 @code{thread} command, will generate this notification.
26262 We suggest that in response to this notification, front ends
26263 highlight the selected thread and cause subsequent commands to apply to
26266 @item =library-loaded,...
26267 Reports that a new library file was loaded by the program. This
26268 notification has 4 fields---@var{id}, @var{target-name},
26269 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
26270 opaque identifier of the library. For remote debugging case,
26271 @var{target-name} and @var{host-name} fields give the name of the
26272 library file on the target, and on the host respectively. For native
26273 debugging, both those fields have the same value. The
26274 @var{symbols-loaded} field is emitted only for backward compatibility
26275 and should not be relied on to convey any useful information. The
26276 @var{thread-group} field, if present, specifies the id of the thread
26277 group in whose context the library was loaded. If the field is
26278 absent, it means the library was loaded in the context of all present
26281 @item =library-unloaded,...
26282 Reports that a library was unloaded by the program. This notification
26283 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
26284 the same meaning as for the @code{=library-loaded} notification.
26285 The @var{thread-group} field, if present, specifies the id of the
26286 thread group in whose context the library was unloaded. If the field is
26287 absent, it means the library was unloaded in the context of all present
26290 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
26291 @itemx =traceframe-changed,end
26292 Reports that the trace frame was changed and its new number is
26293 @var{tfnum}. The number of the tracepoint associated with this trace
26294 frame is @var{tpnum}.
26296 @item =tsv-created,name=@var{name},initial=@var{initial}
26297 Reports that the new trace state variable @var{name} is created with
26298 initial value @var{initial}.
26300 @item =tsv-deleted,name=@var{name}
26301 @itemx =tsv-deleted
26302 Reports that the trace state variable @var{name} is deleted or all
26303 trace state variables are deleted.
26305 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
26306 Reports that the trace state variable @var{name} is modified with
26307 the initial value @var{initial}. The current value @var{current} of
26308 trace state variable is optional and is reported if the current
26309 value of trace state variable is known.
26311 @item =breakpoint-created,bkpt=@{...@}
26312 @itemx =breakpoint-modified,bkpt=@{...@}
26313 @itemx =breakpoint-deleted,id=@var{number}
26314 Reports that a breakpoint was created, modified, or deleted,
26315 respectively. Only user-visible breakpoints are reported to the MI
26318 The @var{bkpt} argument is of the same form as returned by the various
26319 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
26320 @var{number} is the ordinal number of the breakpoint.
26322 Note that if a breakpoint is emitted in the result record of a
26323 command, then it will not also be emitted in an async record.
26325 @item =record-started,thread-group="@var{id}"
26326 @itemx =record-stopped,thread-group="@var{id}"
26327 Execution log recording was either started or stopped on an
26328 inferior. The @var{id} is the @value{GDBN} identifier of the thread
26329 group corresponding to the affected inferior.
26331 @item =cmd-param-changed,param=@var{param},value=@var{value}
26332 Reports that a parameter of the command @code{set @var{param}} is
26333 changed to @var{value}. In the multi-word @code{set} command,
26334 the @var{param} is the whole parameter list to @code{set} command.
26335 For example, In command @code{set check type on}, @var{param}
26336 is @code{check type} and @var{value} is @code{on}.
26338 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
26339 Reports that bytes from @var{addr} to @var{data} + @var{len} were
26340 written in an inferior. The @var{id} is the identifier of the
26341 thread group corresponding to the affected inferior. The optional
26342 @code{type="code"} part is reported if the memory written to holds
26346 @node GDB/MI Breakpoint Information
26347 @subsection @sc{gdb/mi} Breakpoint Information
26349 When @value{GDBN} reports information about a breakpoint, a
26350 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
26355 The breakpoint number. For a breakpoint that represents one location
26356 of a multi-location breakpoint, this will be a dotted pair, like
26360 The type of the breakpoint. For ordinary breakpoints this will be
26361 @samp{breakpoint}, but many values are possible.
26364 If the type of the breakpoint is @samp{catchpoint}, then this
26365 indicates the exact type of catchpoint.
26368 This is the breakpoint disposition---either @samp{del}, meaning that
26369 the breakpoint will be deleted at the next stop, or @samp{keep},
26370 meaning that the breakpoint will not be deleted.
26373 This indicates whether the breakpoint is enabled, in which case the
26374 value is @samp{y}, or disabled, in which case the value is @samp{n}.
26375 Note that this is not the same as the field @code{enable}.
26378 The address of the breakpoint. This may be a hexidecimal number,
26379 giving the address; or the string @samp{<PENDING>}, for a pending
26380 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
26381 multiple locations. This field will not be present if no address can
26382 be determined. For example, a watchpoint does not have an address.
26385 If known, the function in which the breakpoint appears.
26386 If not known, this field is not present.
26389 The name of the source file which contains this function, if known.
26390 If not known, this field is not present.
26393 The full file name of the source file which contains this function, if
26394 known. If not known, this field is not present.
26397 The line number at which this breakpoint appears, if known.
26398 If not known, this field is not present.
26401 If the source file is not known, this field may be provided. If
26402 provided, this holds the address of the breakpoint, possibly followed
26406 If this breakpoint is pending, this field is present and holds the
26407 text used to set the breakpoint, as entered by the user.
26410 Where this breakpoint's condition is evaluated, either @samp{host} or
26414 If this is a thread-specific breakpoint, then this identifies the
26415 thread in which the breakpoint can trigger.
26418 If this breakpoint is restricted to a particular Ada task, then this
26419 field will hold the task identifier.
26422 If the breakpoint is conditional, this is the condition expression.
26425 The ignore count of the breakpoint.
26428 The enable count of the breakpoint.
26430 @item traceframe-usage
26433 @item static-tracepoint-marker-string-id
26434 For a static tracepoint, the name of the static tracepoint marker.
26437 For a masked watchpoint, this is the mask.
26440 A tracepoint's pass count.
26442 @item original-location
26443 The location of the breakpoint as originally specified by the user.
26444 This field is optional.
26447 The number of times the breakpoint has been hit.
26450 This field is only given for tracepoints. This is either @samp{y},
26451 meaning that the tracepoint is installed, or @samp{n}, meaning that it
26455 Some extra data, the exact contents of which are type-dependent.
26459 For example, here is what the output of @code{-break-insert}
26460 (@pxref{GDB/MI Breakpoint Commands}) might be:
26463 -> -break-insert main
26464 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
26465 enabled="y",addr="0x08048564",func="main",file="myprog.c",
26466 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
26471 @node GDB/MI Frame Information
26472 @subsection @sc{gdb/mi} Frame Information
26474 Response from many MI commands includes an information about stack
26475 frame. This information is a tuple that may have the following
26480 The level of the stack frame. The innermost frame has the level of
26481 zero. This field is always present.
26484 The name of the function corresponding to the frame. This field may
26485 be absent if @value{GDBN} is unable to determine the function name.
26488 The code address for the frame. This field is always present.
26491 The name of the source files that correspond to the frame's code
26492 address. This field may be absent.
26495 The source line corresponding to the frames' code address. This field
26499 The name of the binary file (either executable or shared library) the
26500 corresponds to the frame's code address. This field may be absent.
26504 @node GDB/MI Thread Information
26505 @subsection @sc{gdb/mi} Thread Information
26507 Whenever @value{GDBN} has to report an information about a thread, it
26508 uses a tuple with the following fields:
26512 The global numeric id assigned to the thread by @value{GDBN}. This field is
26516 Target-specific string identifying the thread. This field is always present.
26519 Additional information about the thread provided by the target.
26520 It is supposed to be human-readable and not interpreted by the
26521 frontend. This field is optional.
26524 Either @samp{stopped} or @samp{running}, depending on whether the
26525 thread is presently running. This field is always present.
26528 The value of this field is an integer number of the processor core the
26529 thread was last seen on. This field is optional.
26532 @node GDB/MI Ada Exception Information
26533 @subsection @sc{gdb/mi} Ada Exception Information
26535 Whenever a @code{*stopped} record is emitted because the program
26536 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
26537 @value{GDBN} provides the name of the exception that was raised via
26538 the @code{exception-name} field.
26540 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26541 @node GDB/MI Simple Examples
26542 @section Simple Examples of @sc{gdb/mi} Interaction
26543 @cindex @sc{gdb/mi}, simple examples
26545 This subsection presents several simple examples of interaction using
26546 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
26547 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
26548 the output received from @sc{gdb/mi}.
26550 Note the line breaks shown in the examples are here only for
26551 readability, they don't appear in the real output.
26553 @subheading Setting a Breakpoint
26555 Setting a breakpoint generates synchronous output which contains detailed
26556 information of the breakpoint.
26559 -> -break-insert main
26560 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
26561 enabled="y",addr="0x08048564",func="main",file="myprog.c",
26562 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
26567 @subheading Program Execution
26569 Program execution generates asynchronous records and MI gives the
26570 reason that execution stopped.
26576 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
26577 frame=@{addr="0x08048564",func="main",
26578 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
26579 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
26584 <- *stopped,reason="exited-normally"
26588 @subheading Quitting @value{GDBN}
26590 Quitting @value{GDBN} just prints the result class @samp{^exit}.
26598 Please note that @samp{^exit} is printed immediately, but it might
26599 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
26600 performs necessary cleanups, including killing programs being debugged
26601 or disconnecting from debug hardware, so the frontend should wait till
26602 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
26603 fails to exit in reasonable time.
26605 @subheading A Bad Command
26607 Here's what happens if you pass a non-existent command:
26611 <- ^error,msg="Undefined MI command: rubbish"
26616 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26617 @node GDB/MI Command Description Format
26618 @section @sc{gdb/mi} Command Description Format
26620 The remaining sections describe blocks of commands. Each block of
26621 commands is laid out in a fashion similar to this section.
26623 @subheading Motivation
26625 The motivation for this collection of commands.
26627 @subheading Introduction
26629 A brief introduction to this collection of commands as a whole.
26631 @subheading Commands
26633 For each command in the block, the following is described:
26635 @subsubheading Synopsis
26638 -command @var{args}@dots{}
26641 @subsubheading Result
26643 @subsubheading @value{GDBN} Command
26645 The corresponding @value{GDBN} CLI command(s), if any.
26647 @subsubheading Example
26649 Example(s) formatted for readability. Some of the described commands have
26650 not been implemented yet and these are labeled N.A.@: (not available).
26653 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26654 @node GDB/MI Breakpoint Commands
26655 @section @sc{gdb/mi} Breakpoint Commands
26657 @cindex breakpoint commands for @sc{gdb/mi}
26658 @cindex @sc{gdb/mi}, breakpoint commands
26659 This section documents @sc{gdb/mi} commands for manipulating
26662 @subheading The @code{-break-after} Command
26663 @findex -break-after
26665 @subsubheading Synopsis
26668 -break-after @var{number} @var{count}
26671 The breakpoint number @var{number} is not in effect until it has been
26672 hit @var{count} times. To see how this is reflected in the output of
26673 the @samp{-break-list} command, see the description of the
26674 @samp{-break-list} command below.
26676 @subsubheading @value{GDBN} Command
26678 The corresponding @value{GDBN} command is @samp{ignore}.
26680 @subsubheading Example
26685 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
26686 enabled="y",addr="0x000100d0",func="main",file="hello.c",
26687 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
26695 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
26696 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26697 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26698 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26699 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26700 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26701 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26702 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26703 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
26704 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
26709 @subheading The @code{-break-catch} Command
26710 @findex -break-catch
26713 @subheading The @code{-break-commands} Command
26714 @findex -break-commands
26716 @subsubheading Synopsis
26719 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
26722 Specifies the CLI commands that should be executed when breakpoint
26723 @var{number} is hit. The parameters @var{command1} to @var{commandN}
26724 are the commands. If no command is specified, any previously-set
26725 commands are cleared. @xref{Break Commands}. Typical use of this
26726 functionality is tracing a program, that is, printing of values of
26727 some variables whenever breakpoint is hit and then continuing.
26729 @subsubheading @value{GDBN} Command
26731 The corresponding @value{GDBN} command is @samp{commands}.
26733 @subsubheading Example
26738 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
26739 enabled="y",addr="0x000100d0",func="main",file="hello.c",
26740 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
26743 -break-commands 1 "print v" "continue"
26748 @subheading The @code{-break-condition} Command
26749 @findex -break-condition
26751 @subsubheading Synopsis
26754 -break-condition @var{number} @var{expr}
26757 Breakpoint @var{number} will stop the program only if the condition in
26758 @var{expr} is true. The condition becomes part of the
26759 @samp{-break-list} output (see the description of the @samp{-break-list}
26762 @subsubheading @value{GDBN} Command
26764 The corresponding @value{GDBN} command is @samp{condition}.
26766 @subsubheading Example
26770 -break-condition 1 1
26774 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
26775 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26776 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26777 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26778 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26779 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26780 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26781 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26782 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
26783 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
26787 @subheading The @code{-break-delete} Command
26788 @findex -break-delete
26790 @subsubheading Synopsis
26793 -break-delete ( @var{breakpoint} )+
26796 Delete the breakpoint(s) whose number(s) are specified in the argument
26797 list. This is obviously reflected in the breakpoint list.
26799 @subsubheading @value{GDBN} Command
26801 The corresponding @value{GDBN} command is @samp{delete}.
26803 @subsubheading Example
26811 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
26812 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26813 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26814 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26815 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26816 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26817 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26822 @subheading The @code{-break-disable} Command
26823 @findex -break-disable
26825 @subsubheading Synopsis
26828 -break-disable ( @var{breakpoint} )+
26831 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
26832 break list is now set to @samp{n} for the named @var{breakpoint}(s).
26834 @subsubheading @value{GDBN} Command
26836 The corresponding @value{GDBN} command is @samp{disable}.
26838 @subsubheading Example
26846 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
26847 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26848 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26849 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26850 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26851 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26852 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26853 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
26854 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
26855 line="5",thread-groups=["i1"],times="0"@}]@}
26859 @subheading The @code{-break-enable} Command
26860 @findex -break-enable
26862 @subsubheading Synopsis
26865 -break-enable ( @var{breakpoint} )+
26868 Enable (previously disabled) @var{breakpoint}(s).
26870 @subsubheading @value{GDBN} Command
26872 The corresponding @value{GDBN} command is @samp{enable}.
26874 @subsubheading Example
26882 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
26883 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26884 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26885 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26886 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26887 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26888 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26889 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
26890 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
26891 line="5",thread-groups=["i1"],times="0"@}]@}
26895 @subheading The @code{-break-info} Command
26896 @findex -break-info
26898 @subsubheading Synopsis
26901 -break-info @var{breakpoint}
26905 Get information about a single breakpoint.
26907 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
26908 Information}, for details on the format of each breakpoint in the
26911 @subsubheading @value{GDBN} Command
26913 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
26915 @subsubheading Example
26918 @subheading The @code{-break-insert} Command
26919 @findex -break-insert
26920 @anchor{-break-insert}
26922 @subsubheading Synopsis
26925 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
26926 [ -c @var{condition} ] [ -i @var{ignore-count} ]
26927 [ -p @var{thread-id} ] [ @var{location} ]
26931 If specified, @var{location}, can be one of:
26934 @item linespec location
26935 A linespec location. @xref{Linespec Locations}.
26937 @item explicit location
26938 An explicit location. @sc{gdb/mi} explicit locations are
26939 analogous to the CLI's explicit locations using the option names
26940 listed below. @xref{Explicit Locations}.
26943 @item --source @var{filename}
26944 The source file name of the location. This option requires the use
26945 of either @samp{--function} or @samp{--line}.
26947 @item --function @var{function}
26948 The name of a function or method.
26950 @item --label @var{label}
26951 The name of a label.
26953 @item --line @var{lineoffset}
26954 An absolute or relative line offset from the start of the location.
26957 @item address location
26958 An address location, *@var{address}. @xref{Address Locations}.
26962 The possible optional parameters of this command are:
26966 Insert a temporary breakpoint.
26968 Insert a hardware breakpoint.
26970 If @var{location} cannot be parsed (for example if it
26971 refers to unknown files or functions), create a pending
26972 breakpoint. Without this flag, @value{GDBN} will report
26973 an error, and won't create a breakpoint, if @var{location}
26976 Create a disabled breakpoint.
26978 Create a tracepoint. @xref{Tracepoints}. When this parameter
26979 is used together with @samp{-h}, a fast tracepoint is created.
26980 @item -c @var{condition}
26981 Make the breakpoint conditional on @var{condition}.
26982 @item -i @var{ignore-count}
26983 Initialize the @var{ignore-count}.
26984 @item -p @var{thread-id}
26985 Restrict the breakpoint to the thread with the specified global
26989 @subsubheading Result
26991 @xref{GDB/MI Breakpoint Information}, for details on the format of the
26992 resulting breakpoint.
26994 Note: this format is open to change.
26995 @c An out-of-band breakpoint instead of part of the result?
26997 @subsubheading @value{GDBN} Command
26999 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
27000 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
27002 @subsubheading Example
27007 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
27008 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
27011 -break-insert -t foo
27012 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
27013 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
27017 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27018 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27019 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27020 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27021 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27022 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27023 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27024 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27025 addr="0x0001072c", func="main",file="recursive2.c",
27026 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
27028 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
27029 addr="0x00010774",func="foo",file="recursive2.c",
27030 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
27033 @c -break-insert -r foo.*
27034 @c ~int foo(int, int);
27035 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
27036 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
27041 @subheading The @code{-dprintf-insert} Command
27042 @findex -dprintf-insert
27044 @subsubheading Synopsis
27047 -dprintf-insert [ -t ] [ -f ] [ -d ]
27048 [ -c @var{condition} ] [ -i @var{ignore-count} ]
27049 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
27054 If supplied, @var{location} may be specified the same way as for
27055 the @code{-break-insert} command. @xref{-break-insert}.
27057 The possible optional parameters of this command are:
27061 Insert a temporary breakpoint.
27063 If @var{location} cannot be parsed (for example, if it
27064 refers to unknown files or functions), create a pending
27065 breakpoint. Without this flag, @value{GDBN} will report
27066 an error, and won't create a breakpoint, if @var{location}
27069 Create a disabled breakpoint.
27070 @item -c @var{condition}
27071 Make the breakpoint conditional on @var{condition}.
27072 @item -i @var{ignore-count}
27073 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
27074 to @var{ignore-count}.
27075 @item -p @var{thread-id}
27076 Restrict the breakpoint to the thread with the specified global
27080 @subsubheading Result
27082 @xref{GDB/MI Breakpoint Information}, for details on the format of the
27083 resulting breakpoint.
27085 @c An out-of-band breakpoint instead of part of the result?
27087 @subsubheading @value{GDBN} Command
27089 The corresponding @value{GDBN} command is @samp{dprintf}.
27091 @subsubheading Example
27095 4-dprintf-insert foo "At foo entry\n"
27096 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
27097 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
27098 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
27099 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
27100 original-location="foo"@}
27102 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
27103 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
27104 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
27105 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
27106 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
27107 original-location="mi-dprintf.c:26"@}
27111 @subheading The @code{-break-list} Command
27112 @findex -break-list
27114 @subsubheading Synopsis
27120 Displays the list of inserted breakpoints, showing the following fields:
27124 number of the breakpoint
27126 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
27128 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
27131 is the breakpoint enabled or no: @samp{y} or @samp{n}
27133 memory location at which the breakpoint is set
27135 logical location of the breakpoint, expressed by function name, file
27137 @item Thread-groups
27138 list of thread groups to which this breakpoint applies
27140 number of times the breakpoint has been hit
27143 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
27144 @code{body} field is an empty list.
27146 @subsubheading @value{GDBN} Command
27148 The corresponding @value{GDBN} command is @samp{info break}.
27150 @subsubheading Example
27155 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27156 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27157 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27158 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27159 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27160 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27161 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27162 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27163 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
27165 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
27166 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
27167 line="13",thread-groups=["i1"],times="0"@}]@}
27171 Here's an example of the result when there are no breakpoints:
27176 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
27177 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27178 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27179 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27180 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27181 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27182 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27187 @subheading The @code{-break-passcount} Command
27188 @findex -break-passcount
27190 @subsubheading Synopsis
27193 -break-passcount @var{tracepoint-number} @var{passcount}
27196 Set the passcount for tracepoint @var{tracepoint-number} to
27197 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
27198 is not a tracepoint, error is emitted. This corresponds to CLI
27199 command @samp{passcount}.
27201 @subheading The @code{-break-watch} Command
27202 @findex -break-watch
27204 @subsubheading Synopsis
27207 -break-watch [ -a | -r ]
27210 Create a watchpoint. With the @samp{-a} option it will create an
27211 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
27212 read from or on a write to the memory location. With the @samp{-r}
27213 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
27214 trigger only when the memory location is accessed for reading. Without
27215 either of the options, the watchpoint created is a regular watchpoint,
27216 i.e., it will trigger when the memory location is accessed for writing.
27217 @xref{Set Watchpoints, , Setting Watchpoints}.
27219 Note that @samp{-break-list} will report a single list of watchpoints and
27220 breakpoints inserted.
27222 @subsubheading @value{GDBN} Command
27224 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
27227 @subsubheading Example
27229 Setting a watchpoint on a variable in the @code{main} function:
27234 ^done,wpt=@{number="2",exp="x"@}
27239 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
27240 value=@{old="-268439212",new="55"@},
27241 frame=@{func="main",args=[],file="recursive2.c",
27242 fullname="/home/foo/bar/recursive2.c",line="5"@}
27246 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
27247 the program execution twice: first for the variable changing value, then
27248 for the watchpoint going out of scope.
27253 ^done,wpt=@{number="5",exp="C"@}
27258 *stopped,reason="watchpoint-trigger",
27259 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
27260 frame=@{func="callee4",args=[],
27261 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27262 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
27267 *stopped,reason="watchpoint-scope",wpnum="5",
27268 frame=@{func="callee3",args=[@{name="strarg",
27269 value="0x11940 \"A string argument.\""@}],
27270 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27271 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
27275 Listing breakpoints and watchpoints, at different points in the program
27276 execution. Note that once the watchpoint goes out of scope, it is
27282 ^done,wpt=@{number="2",exp="C"@}
27285 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27286 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27287 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27288 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27289 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27290 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27291 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27292 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27293 addr="0x00010734",func="callee4",
27294 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27295 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
27297 bkpt=@{number="2",type="watchpoint",disp="keep",
27298 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
27303 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
27304 value=@{old="-276895068",new="3"@},
27305 frame=@{func="callee4",args=[],
27306 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27307 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
27310 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27311 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27312 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27313 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27314 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27315 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27316 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27317 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27318 addr="0x00010734",func="callee4",
27319 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27320 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
27322 bkpt=@{number="2",type="watchpoint",disp="keep",
27323 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
27327 ^done,reason="watchpoint-scope",wpnum="2",
27328 frame=@{func="callee3",args=[@{name="strarg",
27329 value="0x11940 \"A string argument.\""@}],
27330 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27331 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
27334 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27335 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27336 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27337 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27338 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27339 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27340 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27341 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27342 addr="0x00010734",func="callee4",
27343 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27344 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
27345 thread-groups=["i1"],times="1"@}]@}
27350 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27351 @node GDB/MI Catchpoint Commands
27352 @section @sc{gdb/mi} Catchpoint Commands
27354 This section documents @sc{gdb/mi} commands for manipulating
27358 * Shared Library GDB/MI Catchpoint Commands::
27359 * Ada Exception GDB/MI Catchpoint Commands::
27362 @node Shared Library GDB/MI Catchpoint Commands
27363 @subsection Shared Library @sc{gdb/mi} Catchpoints
27365 @subheading The @code{-catch-load} Command
27366 @findex -catch-load
27368 @subsubheading Synopsis
27371 -catch-load [ -t ] [ -d ] @var{regexp}
27374 Add a catchpoint for library load events. If the @samp{-t} option is used,
27375 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
27376 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
27377 in a disabled state. The @samp{regexp} argument is a regular
27378 expression used to match the name of the loaded library.
27381 @subsubheading @value{GDBN} Command
27383 The corresponding @value{GDBN} command is @samp{catch load}.
27385 @subsubheading Example
27388 -catch-load -t foo.so
27389 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
27390 what="load of library matching foo.so",catch-type="load",times="0"@}
27395 @subheading The @code{-catch-unload} Command
27396 @findex -catch-unload
27398 @subsubheading Synopsis
27401 -catch-unload [ -t ] [ -d ] @var{regexp}
27404 Add a catchpoint for library unload events. If the @samp{-t} option is
27405 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
27406 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
27407 created in a disabled state. The @samp{regexp} argument is a regular
27408 expression used to match the name of the unloaded library.
27410 @subsubheading @value{GDBN} Command
27412 The corresponding @value{GDBN} command is @samp{catch unload}.
27414 @subsubheading Example
27417 -catch-unload -d bar.so
27418 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
27419 what="load of library matching bar.so",catch-type="unload",times="0"@}
27423 @node Ada Exception GDB/MI Catchpoint Commands
27424 @subsection Ada Exception @sc{gdb/mi} Catchpoints
27426 The following @sc{gdb/mi} commands can be used to create catchpoints
27427 that stop the execution when Ada exceptions are being raised.
27429 @subheading The @code{-catch-assert} Command
27430 @findex -catch-assert
27432 @subsubheading Synopsis
27435 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
27438 Add a catchpoint for failed Ada assertions.
27440 The possible optional parameters for this command are:
27443 @item -c @var{condition}
27444 Make the catchpoint conditional on @var{condition}.
27446 Create a disabled catchpoint.
27448 Create a temporary catchpoint.
27451 @subsubheading @value{GDBN} Command
27453 The corresponding @value{GDBN} command is @samp{catch assert}.
27455 @subsubheading Example
27459 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
27460 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
27461 thread-groups=["i1"],times="0",
27462 original-location="__gnat_debug_raise_assert_failure"@}
27466 @subheading The @code{-catch-exception} Command
27467 @findex -catch-exception
27469 @subsubheading Synopsis
27472 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
27476 Add a catchpoint stopping when Ada exceptions are raised.
27477 By default, the command stops the program when any Ada exception
27478 gets raised. But it is also possible, by using some of the
27479 optional parameters described below, to create more selective
27482 The possible optional parameters for this command are:
27485 @item -c @var{condition}
27486 Make the catchpoint conditional on @var{condition}.
27488 Create a disabled catchpoint.
27489 @item -e @var{exception-name}
27490 Only stop when @var{exception-name} is raised. This option cannot
27491 be used combined with @samp{-u}.
27493 Create a temporary catchpoint.
27495 Stop only when an unhandled exception gets raised. This option
27496 cannot be used combined with @samp{-e}.
27499 @subsubheading @value{GDBN} Command
27501 The corresponding @value{GDBN} commands are @samp{catch exception}
27502 and @samp{catch exception unhandled}.
27504 @subsubheading Example
27507 -catch-exception -e Program_Error
27508 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
27509 enabled="y",addr="0x0000000000404874",
27510 what="`Program_Error' Ada exception", thread-groups=["i1"],
27511 times="0",original-location="__gnat_debug_raise_exception"@}
27515 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27516 @node GDB/MI Program Context
27517 @section @sc{gdb/mi} Program Context
27519 @subheading The @code{-exec-arguments} Command
27520 @findex -exec-arguments
27523 @subsubheading Synopsis
27526 -exec-arguments @var{args}
27529 Set the inferior program arguments, to be used in the next
27532 @subsubheading @value{GDBN} Command
27534 The corresponding @value{GDBN} command is @samp{set args}.
27536 @subsubheading Example
27540 -exec-arguments -v word
27547 @subheading The @code{-exec-show-arguments} Command
27548 @findex -exec-show-arguments
27550 @subsubheading Synopsis
27553 -exec-show-arguments
27556 Print the arguments of the program.
27558 @subsubheading @value{GDBN} Command
27560 The corresponding @value{GDBN} command is @samp{show args}.
27562 @subsubheading Example
27567 @subheading The @code{-environment-cd} Command
27568 @findex -environment-cd
27570 @subsubheading Synopsis
27573 -environment-cd @var{pathdir}
27576 Set @value{GDBN}'s working directory.
27578 @subsubheading @value{GDBN} Command
27580 The corresponding @value{GDBN} command is @samp{cd}.
27582 @subsubheading Example
27586 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
27592 @subheading The @code{-environment-directory} Command
27593 @findex -environment-directory
27595 @subsubheading Synopsis
27598 -environment-directory [ -r ] [ @var{pathdir} ]+
27601 Add directories @var{pathdir} to beginning of search path for source files.
27602 If the @samp{-r} option is used, the search path is reset to the default
27603 search path. If directories @var{pathdir} are supplied in addition to the
27604 @samp{-r} option, the search path is first reset and then addition
27606 Multiple directories may be specified, separated by blanks. Specifying
27607 multiple directories in a single command
27608 results in the directories added to the beginning of the
27609 search path in the same order they were presented in the command.
27610 If blanks are needed as
27611 part of a directory name, double-quotes should be used around
27612 the name. In the command output, the path will show up separated
27613 by the system directory-separator character. The directory-separator
27614 character must not be used
27615 in any directory name.
27616 If no directories are specified, the current search path is displayed.
27618 @subsubheading @value{GDBN} Command
27620 The corresponding @value{GDBN} command is @samp{dir}.
27622 @subsubheading Example
27626 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
27627 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
27629 -environment-directory ""
27630 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
27632 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
27633 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
27635 -environment-directory -r
27636 ^done,source-path="$cdir:$cwd"
27641 @subheading The @code{-environment-path} Command
27642 @findex -environment-path
27644 @subsubheading Synopsis
27647 -environment-path [ -r ] [ @var{pathdir} ]+
27650 Add directories @var{pathdir} to beginning of search path for object files.
27651 If the @samp{-r} option is used, the search path is reset to the original
27652 search path that existed at gdb start-up. If directories @var{pathdir} are
27653 supplied in addition to the
27654 @samp{-r} option, the search path is first reset and then addition
27656 Multiple directories may be specified, separated by blanks. Specifying
27657 multiple directories in a single command
27658 results in the directories added to the beginning of the
27659 search path in the same order they were presented in the command.
27660 If blanks are needed as
27661 part of a directory name, double-quotes should be used around
27662 the name. In the command output, the path will show up separated
27663 by the system directory-separator character. The directory-separator
27664 character must not be used
27665 in any directory name.
27666 If no directories are specified, the current path is displayed.
27669 @subsubheading @value{GDBN} Command
27671 The corresponding @value{GDBN} command is @samp{path}.
27673 @subsubheading Example
27678 ^done,path="/usr/bin"
27680 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
27681 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
27683 -environment-path -r /usr/local/bin
27684 ^done,path="/usr/local/bin:/usr/bin"
27689 @subheading The @code{-environment-pwd} Command
27690 @findex -environment-pwd
27692 @subsubheading Synopsis
27698 Show the current working directory.
27700 @subsubheading @value{GDBN} Command
27702 The corresponding @value{GDBN} command is @samp{pwd}.
27704 @subsubheading Example
27709 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
27713 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27714 @node GDB/MI Thread Commands
27715 @section @sc{gdb/mi} Thread Commands
27718 @subheading The @code{-thread-info} Command
27719 @findex -thread-info
27721 @subsubheading Synopsis
27724 -thread-info [ @var{thread-id} ]
27727 Reports information about either a specific thread, if the
27728 @var{thread-id} parameter is present, or about all threads.
27729 @var{thread-id} is the thread's global thread ID. When printing
27730 information about all threads, also reports the global ID of the
27733 @subsubheading @value{GDBN} Command
27735 The @samp{info thread} command prints the same information
27738 @subsubheading Result
27740 The result is a list of threads. The following attributes are
27741 defined for a given thread:
27745 This field exists only for the current thread. It has the value @samp{*}.
27748 The global identifier that @value{GDBN} uses to refer to the thread.
27751 The identifier that the target uses to refer to the thread.
27754 Extra information about the thread, in a target-specific format. This
27758 The name of the thread. If the user specified a name using the
27759 @code{thread name} command, then this name is given. Otherwise, if
27760 @value{GDBN} can extract the thread name from the target, then that
27761 name is given. If @value{GDBN} cannot find the thread name, then this
27765 The stack frame currently executing in the thread.
27768 The thread's state. The @samp{state} field may have the following
27773 The thread is stopped. Frame information is available for stopped
27777 The thread is running. There's no frame information for running
27783 If @value{GDBN} can find the CPU core on which this thread is running,
27784 then this field is the core identifier. This field is optional.
27788 @subsubheading Example
27793 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
27794 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
27795 args=[]@},state="running"@},
27796 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
27797 frame=@{level="0",addr="0x0804891f",func="foo",
27798 args=[@{name="i",value="10"@}],
27799 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},
27800 state="running"@}],
27801 current-thread-id="1"
27805 @subheading The @code{-thread-list-ids} Command
27806 @findex -thread-list-ids
27808 @subsubheading Synopsis
27814 Produces a list of the currently known global @value{GDBN} thread ids.
27815 At the end of the list it also prints the total number of such
27818 This command is retained for historical reasons, the
27819 @code{-thread-info} command should be used instead.
27821 @subsubheading @value{GDBN} Command
27823 Part of @samp{info threads} supplies the same information.
27825 @subsubheading Example
27830 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
27831 current-thread-id="1",number-of-threads="3"
27836 @subheading The @code{-thread-select} Command
27837 @findex -thread-select
27839 @subsubheading Synopsis
27842 -thread-select @var{thread-id}
27845 Make thread with global thread number @var{thread-id} the current
27846 thread. It prints the number of the new current thread, and the
27847 topmost frame for that thread.
27849 This command is deprecated in favor of explicitly using the
27850 @samp{--thread} option to each command.
27852 @subsubheading @value{GDBN} Command
27854 The corresponding @value{GDBN} command is @samp{thread}.
27856 @subsubheading Example
27863 *stopped,reason="end-stepping-range",thread-id="2",line="187",
27864 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
27868 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
27869 number-of-threads="3"
27872 ^done,new-thread-id="3",
27873 frame=@{level="0",func="vprintf",
27874 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
27875 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
27879 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27880 @node GDB/MI Ada Tasking Commands
27881 @section @sc{gdb/mi} Ada Tasking Commands
27883 @subheading The @code{-ada-task-info} Command
27884 @findex -ada-task-info
27886 @subsubheading Synopsis
27889 -ada-task-info [ @var{task-id} ]
27892 Reports information about either a specific Ada task, if the
27893 @var{task-id} parameter is present, or about all Ada tasks.
27895 @subsubheading @value{GDBN} Command
27897 The @samp{info tasks} command prints the same information
27898 about all Ada tasks (@pxref{Ada Tasks}).
27900 @subsubheading Result
27902 The result is a table of Ada tasks. The following columns are
27903 defined for each Ada task:
27907 This field exists only for the current thread. It has the value @samp{*}.
27910 The identifier that @value{GDBN} uses to refer to the Ada task.
27913 The identifier that the target uses to refer to the Ada task.
27916 The global thread identifier of the thread corresponding to the Ada
27919 This field should always exist, as Ada tasks are always implemented
27920 on top of a thread. But if @value{GDBN} cannot find this corresponding
27921 thread for any reason, the field is omitted.
27924 This field exists only when the task was created by another task.
27925 In this case, it provides the ID of the parent task.
27928 The base priority of the task.
27931 The current state of the task. For a detailed description of the
27932 possible states, see @ref{Ada Tasks}.
27935 The name of the task.
27939 @subsubheading Example
27943 ^done,tasks=@{nr_rows="3",nr_cols="8",
27944 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
27945 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
27946 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
27947 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
27948 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
27949 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
27950 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
27951 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
27952 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
27953 state="Child Termination Wait",name="main_task"@}]@}
27957 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27958 @node GDB/MI Program Execution
27959 @section @sc{gdb/mi} Program Execution
27961 These are the asynchronous commands which generate the out-of-band
27962 record @samp{*stopped}. Currently @value{GDBN} only really executes
27963 asynchronously with remote targets and this interaction is mimicked in
27966 @subheading The @code{-exec-continue} Command
27967 @findex -exec-continue
27969 @subsubheading Synopsis
27972 -exec-continue [--reverse] [--all|--thread-group N]
27975 Resumes the execution of the inferior program, which will continue
27976 to execute until it reaches a debugger stop event. If the
27977 @samp{--reverse} option is specified, execution resumes in reverse until
27978 it reaches a stop event. Stop events may include
27981 breakpoints or watchpoints
27983 signals or exceptions
27985 the end of the process (or its beginning under @samp{--reverse})
27987 the end or beginning of a replay log if one is being used.
27989 In all-stop mode (@pxref{All-Stop
27990 Mode}), may resume only one thread, or all threads, depending on the
27991 value of the @samp{scheduler-locking} variable. If @samp{--all} is
27992 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
27993 ignored in all-stop mode. If the @samp{--thread-group} options is
27994 specified, then all threads in that thread group are resumed.
27996 @subsubheading @value{GDBN} Command
27998 The corresponding @value{GDBN} corresponding is @samp{continue}.
28000 @subsubheading Example
28007 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
28008 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
28014 @subheading The @code{-exec-finish} Command
28015 @findex -exec-finish
28017 @subsubheading Synopsis
28020 -exec-finish [--reverse]
28023 Resumes the execution of the inferior program until the current
28024 function is exited. Displays the results returned by the function.
28025 If the @samp{--reverse} option is specified, resumes the reverse
28026 execution of the inferior program until the point where current
28027 function was called.
28029 @subsubheading @value{GDBN} Command
28031 The corresponding @value{GDBN} command is @samp{finish}.
28033 @subsubheading Example
28035 Function returning @code{void}.
28042 *stopped,reason="function-finished",frame=@{func="main",args=[],
28043 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
28047 Function returning other than @code{void}. The name of the internal
28048 @value{GDBN} variable storing the result is printed, together with the
28055 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
28056 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
28057 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28058 gdb-result-var="$1",return-value="0"
28063 @subheading The @code{-exec-interrupt} Command
28064 @findex -exec-interrupt
28066 @subsubheading Synopsis
28069 -exec-interrupt [--all|--thread-group N]
28072 Interrupts the background execution of the target. Note how the token
28073 associated with the stop message is the one for the execution command
28074 that has been interrupted. The token for the interrupt itself only
28075 appears in the @samp{^done} output. If the user is trying to
28076 interrupt a non-running program, an error message will be printed.
28078 Note that when asynchronous execution is enabled, this command is
28079 asynchronous just like other execution commands. That is, first the
28080 @samp{^done} response will be printed, and the target stop will be
28081 reported after that using the @samp{*stopped} notification.
28083 In non-stop mode, only the context thread is interrupted by default.
28084 All threads (in all inferiors) will be interrupted if the
28085 @samp{--all} option is specified. If the @samp{--thread-group}
28086 option is specified, all threads in that group will be interrupted.
28088 @subsubheading @value{GDBN} Command
28090 The corresponding @value{GDBN} command is @samp{interrupt}.
28092 @subsubheading Example
28103 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
28104 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
28105 fullname="/home/foo/bar/try.c",line="13"@}
28110 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
28114 @subheading The @code{-exec-jump} Command
28117 @subsubheading Synopsis
28120 -exec-jump @var{location}
28123 Resumes execution of the inferior program at the location specified by
28124 parameter. @xref{Specify Location}, for a description of the
28125 different forms of @var{location}.
28127 @subsubheading @value{GDBN} Command
28129 The corresponding @value{GDBN} command is @samp{jump}.
28131 @subsubheading Example
28134 -exec-jump foo.c:10
28135 *running,thread-id="all"
28140 @subheading The @code{-exec-next} Command
28143 @subsubheading Synopsis
28146 -exec-next [--reverse]
28149 Resumes execution of the inferior program, stopping when the beginning
28150 of the next source line is reached.
28152 If the @samp{--reverse} option is specified, resumes reverse execution
28153 of the inferior program, stopping at the beginning of the previous
28154 source line. If you issue this command on the first line of a
28155 function, it will take you back to the caller of that function, to the
28156 source line where the function was called.
28159 @subsubheading @value{GDBN} Command
28161 The corresponding @value{GDBN} command is @samp{next}.
28163 @subsubheading Example
28169 *stopped,reason="end-stepping-range",line="8",file="hello.c"
28174 @subheading The @code{-exec-next-instruction} Command
28175 @findex -exec-next-instruction
28177 @subsubheading Synopsis
28180 -exec-next-instruction [--reverse]
28183 Executes one machine instruction. If the instruction is a function
28184 call, continues until the function returns. If the program stops at an
28185 instruction in the middle of a source line, the address will be
28188 If the @samp{--reverse} option is specified, resumes reverse execution
28189 of the inferior program, stopping at the previous instruction. If the
28190 previously executed instruction was a return from another function,
28191 it will continue to execute in reverse until the call to that function
28192 (from the current stack frame) is reached.
28194 @subsubheading @value{GDBN} Command
28196 The corresponding @value{GDBN} command is @samp{nexti}.
28198 @subsubheading Example
28202 -exec-next-instruction
28206 *stopped,reason="end-stepping-range",
28207 addr="0x000100d4",line="5",file="hello.c"
28212 @subheading The @code{-exec-return} Command
28213 @findex -exec-return
28215 @subsubheading Synopsis
28221 Makes current function return immediately. Doesn't execute the inferior.
28222 Displays the new current frame.
28224 @subsubheading @value{GDBN} Command
28226 The corresponding @value{GDBN} command is @samp{return}.
28228 @subsubheading Example
28232 200-break-insert callee4
28233 200^done,bkpt=@{number="1",addr="0x00010734",
28234 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
28239 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
28240 frame=@{func="callee4",args=[],
28241 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28242 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
28248 111^done,frame=@{level="0",func="callee3",
28249 args=[@{name="strarg",
28250 value="0x11940 \"A string argument.\""@}],
28251 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28252 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
28257 @subheading The @code{-exec-run} Command
28260 @subsubheading Synopsis
28263 -exec-run [ --all | --thread-group N ] [ --start ]
28266 Starts execution of the inferior from the beginning. The inferior
28267 executes until either a breakpoint is encountered or the program
28268 exits. In the latter case the output will include an exit code, if
28269 the program has exited exceptionally.
28271 When neither the @samp{--all} nor the @samp{--thread-group} option
28272 is specified, the current inferior is started. If the
28273 @samp{--thread-group} option is specified, it should refer to a thread
28274 group of type @samp{process}, and that thread group will be started.
28275 If the @samp{--all} option is specified, then all inferiors will be started.
28277 Using the @samp{--start} option instructs the debugger to stop
28278 the execution at the start of the inferior's main subprogram,
28279 following the same behavior as the @code{start} command
28280 (@pxref{Starting}).
28282 @subsubheading @value{GDBN} Command
28284 The corresponding @value{GDBN} command is @samp{run}.
28286 @subsubheading Examples
28291 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
28296 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
28297 frame=@{func="main",args=[],file="recursive2.c",
28298 fullname="/home/foo/bar/recursive2.c",line="4"@}
28303 Program exited normally:
28311 *stopped,reason="exited-normally"
28316 Program exited exceptionally:
28324 *stopped,reason="exited",exit-code="01"
28328 Another way the program can terminate is if it receives a signal such as
28329 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
28333 *stopped,reason="exited-signalled",signal-name="SIGINT",
28334 signal-meaning="Interrupt"
28338 @c @subheading -exec-signal
28341 @subheading The @code{-exec-step} Command
28344 @subsubheading Synopsis
28347 -exec-step [--reverse]
28350 Resumes execution of the inferior program, stopping when the beginning
28351 of the next source line is reached, if the next source line is not a
28352 function call. If it is, stop at the first instruction of the called
28353 function. If the @samp{--reverse} option is specified, resumes reverse
28354 execution of the inferior program, stopping at the beginning of the
28355 previously executed source line.
28357 @subsubheading @value{GDBN} Command
28359 The corresponding @value{GDBN} command is @samp{step}.
28361 @subsubheading Example
28363 Stepping into a function:
28369 *stopped,reason="end-stepping-range",
28370 frame=@{func="foo",args=[@{name="a",value="10"@},
28371 @{name="b",value="0"@}],file="recursive2.c",
28372 fullname="/home/foo/bar/recursive2.c",line="11"@}
28382 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
28387 @subheading The @code{-exec-step-instruction} Command
28388 @findex -exec-step-instruction
28390 @subsubheading Synopsis
28393 -exec-step-instruction [--reverse]
28396 Resumes the inferior which executes one machine instruction. If the
28397 @samp{--reverse} option is specified, resumes reverse execution of the
28398 inferior program, stopping at the previously executed instruction.
28399 The output, once @value{GDBN} has stopped, will vary depending on
28400 whether we have stopped in the middle of a source line or not. In the
28401 former case, the address at which the program stopped will be printed
28404 @subsubheading @value{GDBN} Command
28406 The corresponding @value{GDBN} command is @samp{stepi}.
28408 @subsubheading Example
28412 -exec-step-instruction
28416 *stopped,reason="end-stepping-range",
28417 frame=@{func="foo",args=[],file="try.c",
28418 fullname="/home/foo/bar/try.c",line="10"@}
28420 -exec-step-instruction
28424 *stopped,reason="end-stepping-range",
28425 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
28426 fullname="/home/foo/bar/try.c",line="10"@}
28431 @subheading The @code{-exec-until} Command
28432 @findex -exec-until
28434 @subsubheading Synopsis
28437 -exec-until [ @var{location} ]
28440 Executes the inferior until the @var{location} specified in the
28441 argument is reached. If there is no argument, the inferior executes
28442 until a source line greater than the current one is reached. The
28443 reason for stopping in this case will be @samp{location-reached}.
28445 @subsubheading @value{GDBN} Command
28447 The corresponding @value{GDBN} command is @samp{until}.
28449 @subsubheading Example
28453 -exec-until recursive2.c:6
28457 *stopped,reason="location-reached",frame=@{func="main",args=[],
28458 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
28463 @subheading -file-clear
28464 Is this going away????
28467 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28468 @node GDB/MI Stack Manipulation
28469 @section @sc{gdb/mi} Stack Manipulation Commands
28471 @subheading The @code{-enable-frame-filters} Command
28472 @findex -enable-frame-filters
28475 -enable-frame-filters
28478 @value{GDBN} allows Python-based frame filters to affect the output of
28479 the MI commands relating to stack traces. As there is no way to
28480 implement this in a fully backward-compatible way, a front end must
28481 request that this functionality be enabled.
28483 Once enabled, this feature cannot be disabled.
28485 Note that if Python support has not been compiled into @value{GDBN},
28486 this command will still succeed (and do nothing).
28488 @subheading The @code{-stack-info-frame} Command
28489 @findex -stack-info-frame
28491 @subsubheading Synopsis
28497 Get info on the selected frame.
28499 @subsubheading @value{GDBN} Command
28501 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
28502 (without arguments).
28504 @subsubheading Example
28509 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
28510 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28511 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
28515 @subheading The @code{-stack-info-depth} Command
28516 @findex -stack-info-depth
28518 @subsubheading Synopsis
28521 -stack-info-depth [ @var{max-depth} ]
28524 Return the depth of the stack. If the integer argument @var{max-depth}
28525 is specified, do not count beyond @var{max-depth} frames.
28527 @subsubheading @value{GDBN} Command
28529 There's no equivalent @value{GDBN} command.
28531 @subsubheading Example
28533 For a stack with frame levels 0 through 11:
28540 -stack-info-depth 4
28543 -stack-info-depth 12
28546 -stack-info-depth 11
28549 -stack-info-depth 13
28554 @anchor{-stack-list-arguments}
28555 @subheading The @code{-stack-list-arguments} Command
28556 @findex -stack-list-arguments
28558 @subsubheading Synopsis
28561 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
28562 [ @var{low-frame} @var{high-frame} ]
28565 Display a list of the arguments for the frames between @var{low-frame}
28566 and @var{high-frame} (inclusive). If @var{low-frame} and
28567 @var{high-frame} are not provided, list the arguments for the whole
28568 call stack. If the two arguments are equal, show the single frame
28569 at the corresponding level. It is an error if @var{low-frame} is
28570 larger than the actual number of frames. On the other hand,
28571 @var{high-frame} may be larger than the actual number of frames, in
28572 which case only existing frames will be returned.
28574 If @var{print-values} is 0 or @code{--no-values}, print only the names of
28575 the variables; if it is 1 or @code{--all-values}, print also their
28576 values; and if it is 2 or @code{--simple-values}, print the name,
28577 type and value for simple data types, and the name and type for arrays,
28578 structures and unions. If the option @code{--no-frame-filters} is
28579 supplied, then Python frame filters will not be executed.
28581 If the @code{--skip-unavailable} option is specified, arguments that
28582 are not available are not listed. Partially available arguments
28583 are still displayed, however.
28585 Use of this command to obtain arguments in a single frame is
28586 deprecated in favor of the @samp{-stack-list-variables} command.
28588 @subsubheading @value{GDBN} Command
28590 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
28591 @samp{gdb_get_args} command which partially overlaps with the
28592 functionality of @samp{-stack-list-arguments}.
28594 @subsubheading Example
28601 frame=@{level="0",addr="0x00010734",func="callee4",
28602 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28603 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
28604 frame=@{level="1",addr="0x0001076c",func="callee3",
28605 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28606 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
28607 frame=@{level="2",addr="0x0001078c",func="callee2",
28608 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28609 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
28610 frame=@{level="3",addr="0x000107b4",func="callee1",
28611 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28612 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
28613 frame=@{level="4",addr="0x000107e0",func="main",
28614 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28615 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
28617 -stack-list-arguments 0
28620 frame=@{level="0",args=[]@},
28621 frame=@{level="1",args=[name="strarg"]@},
28622 frame=@{level="2",args=[name="intarg",name="strarg"]@},
28623 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
28624 frame=@{level="4",args=[]@}]
28626 -stack-list-arguments 1
28629 frame=@{level="0",args=[]@},
28631 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
28632 frame=@{level="2",args=[
28633 @{name="intarg",value="2"@},
28634 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
28635 @{frame=@{level="3",args=[
28636 @{name="intarg",value="2"@},
28637 @{name="strarg",value="0x11940 \"A string argument.\""@},
28638 @{name="fltarg",value="3.5"@}]@},
28639 frame=@{level="4",args=[]@}]
28641 -stack-list-arguments 0 2 2
28642 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
28644 -stack-list-arguments 1 2 2
28645 ^done,stack-args=[frame=@{level="2",
28646 args=[@{name="intarg",value="2"@},
28647 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
28651 @c @subheading -stack-list-exception-handlers
28654 @anchor{-stack-list-frames}
28655 @subheading The @code{-stack-list-frames} Command
28656 @findex -stack-list-frames
28658 @subsubheading Synopsis
28661 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
28664 List the frames currently on the stack. For each frame it displays the
28669 The frame number, 0 being the topmost frame, i.e., the innermost function.
28671 The @code{$pc} value for that frame.
28675 File name of the source file where the function lives.
28676 @item @var{fullname}
28677 The full file name of the source file where the function lives.
28679 Line number corresponding to the @code{$pc}.
28681 The shared library where this function is defined. This is only given
28682 if the frame's function is not known.
28685 If invoked without arguments, this command prints a backtrace for the
28686 whole stack. If given two integer arguments, it shows the frames whose
28687 levels are between the two arguments (inclusive). If the two arguments
28688 are equal, it shows the single frame at the corresponding level. It is
28689 an error if @var{low-frame} is larger than the actual number of
28690 frames. On the other hand, @var{high-frame} may be larger than the
28691 actual number of frames, in which case only existing frames will be
28692 returned. If the option @code{--no-frame-filters} is supplied, then
28693 Python frame filters will not be executed.
28695 @subsubheading @value{GDBN} Command
28697 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
28699 @subsubheading Example
28701 Full stack backtrace:
28707 [frame=@{level="0",addr="0x0001076c",func="foo",
28708 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
28709 frame=@{level="1",addr="0x000107a4",func="foo",
28710 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28711 frame=@{level="2",addr="0x000107a4",func="foo",
28712 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28713 frame=@{level="3",addr="0x000107a4",func="foo",
28714 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28715 frame=@{level="4",addr="0x000107a4",func="foo",
28716 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28717 frame=@{level="5",addr="0x000107a4",func="foo",
28718 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28719 frame=@{level="6",addr="0x000107a4",func="foo",
28720 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28721 frame=@{level="7",addr="0x000107a4",func="foo",
28722 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28723 frame=@{level="8",addr="0x000107a4",func="foo",
28724 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28725 frame=@{level="9",addr="0x000107a4",func="foo",
28726 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28727 frame=@{level="10",addr="0x000107a4",func="foo",
28728 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28729 frame=@{level="11",addr="0x00010738",func="main",
28730 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
28734 Show frames between @var{low_frame} and @var{high_frame}:
28738 -stack-list-frames 3 5
28740 [frame=@{level="3",addr="0x000107a4",func="foo",
28741 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28742 frame=@{level="4",addr="0x000107a4",func="foo",
28743 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28744 frame=@{level="5",addr="0x000107a4",func="foo",
28745 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
28749 Show a single frame:
28753 -stack-list-frames 3 3
28755 [frame=@{level="3",addr="0x000107a4",func="foo",
28756 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
28761 @subheading The @code{-stack-list-locals} Command
28762 @findex -stack-list-locals
28763 @anchor{-stack-list-locals}
28765 @subsubheading Synopsis
28768 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
28771 Display the local variable names for the selected frame. If
28772 @var{print-values} is 0 or @code{--no-values}, print only the names of
28773 the variables; if it is 1 or @code{--all-values}, print also their
28774 values; and if it is 2 or @code{--simple-values}, print the name,
28775 type and value for simple data types, and the name and type for arrays,
28776 structures and unions. In this last case, a frontend can immediately
28777 display the value of simple data types and create variable objects for
28778 other data types when the user wishes to explore their values in
28779 more detail. If the option @code{--no-frame-filters} is supplied, then
28780 Python frame filters will not be executed.
28782 If the @code{--skip-unavailable} option is specified, local variables
28783 that are not available are not listed. Partially available local
28784 variables are still displayed, however.
28786 This command is deprecated in favor of the
28787 @samp{-stack-list-variables} command.
28789 @subsubheading @value{GDBN} Command
28791 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
28793 @subsubheading Example
28797 -stack-list-locals 0
28798 ^done,locals=[name="A",name="B",name="C"]
28800 -stack-list-locals --all-values
28801 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
28802 @{name="C",value="@{1, 2, 3@}"@}]
28803 -stack-list-locals --simple-values
28804 ^done,locals=[@{name="A",type="int",value="1"@},
28805 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
28809 @anchor{-stack-list-variables}
28810 @subheading The @code{-stack-list-variables} Command
28811 @findex -stack-list-variables
28813 @subsubheading Synopsis
28816 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
28819 Display the names of local variables and function arguments for the selected frame. If
28820 @var{print-values} is 0 or @code{--no-values}, print only the names of
28821 the variables; if it is 1 or @code{--all-values}, print also their
28822 values; and if it is 2 or @code{--simple-values}, print the name,
28823 type and value for simple data types, and the name and type for arrays,
28824 structures and unions. If the option @code{--no-frame-filters} is
28825 supplied, then Python frame filters will not be executed.
28827 If the @code{--skip-unavailable} option is specified, local variables
28828 and arguments that are not available are not listed. Partially
28829 available arguments and local variables are still displayed, however.
28831 @subsubheading Example
28835 -stack-list-variables --thread 1 --frame 0 --all-values
28836 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
28841 @subheading The @code{-stack-select-frame} Command
28842 @findex -stack-select-frame
28844 @subsubheading Synopsis
28847 -stack-select-frame @var{framenum}
28850 Change the selected frame. Select a different frame @var{framenum} on
28853 This command in deprecated in favor of passing the @samp{--frame}
28854 option to every command.
28856 @subsubheading @value{GDBN} Command
28858 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
28859 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
28861 @subsubheading Example
28865 -stack-select-frame 2
28870 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28871 @node GDB/MI Variable Objects
28872 @section @sc{gdb/mi} Variable Objects
28876 @subheading Motivation for Variable Objects in @sc{gdb/mi}
28878 For the implementation of a variable debugger window (locals, watched
28879 expressions, etc.), we are proposing the adaptation of the existing code
28880 used by @code{Insight}.
28882 The two main reasons for that are:
28886 It has been proven in practice (it is already on its second generation).
28889 It will shorten development time (needless to say how important it is
28893 The original interface was designed to be used by Tcl code, so it was
28894 slightly changed so it could be used through @sc{gdb/mi}. This section
28895 describes the @sc{gdb/mi} operations that will be available and gives some
28896 hints about their use.
28898 @emph{Note}: In addition to the set of operations described here, we
28899 expect the @sc{gui} implementation of a variable window to require, at
28900 least, the following operations:
28903 @item @code{-gdb-show} @code{output-radix}
28904 @item @code{-stack-list-arguments}
28905 @item @code{-stack-list-locals}
28906 @item @code{-stack-select-frame}
28911 @subheading Introduction to Variable Objects
28913 @cindex variable objects in @sc{gdb/mi}
28915 Variable objects are "object-oriented" MI interface for examining and
28916 changing values of expressions. Unlike some other MI interfaces that
28917 work with expressions, variable objects are specifically designed for
28918 simple and efficient presentation in the frontend. A variable object
28919 is identified by string name. When a variable object is created, the
28920 frontend specifies the expression for that variable object. The
28921 expression can be a simple variable, or it can be an arbitrary complex
28922 expression, and can even involve CPU registers. After creating a
28923 variable object, the frontend can invoke other variable object
28924 operations---for example to obtain or change the value of a variable
28925 object, or to change display format.
28927 Variable objects have hierarchical tree structure. Any variable object
28928 that corresponds to a composite type, such as structure in C, has
28929 a number of child variable objects, for example corresponding to each
28930 element of a structure. A child variable object can itself have
28931 children, recursively. Recursion ends when we reach
28932 leaf variable objects, which always have built-in types. Child variable
28933 objects are created only by explicit request, so if a frontend
28934 is not interested in the children of a particular variable object, no
28935 child will be created.
28937 For a leaf variable object it is possible to obtain its value as a
28938 string, or set the value from a string. String value can be also
28939 obtained for a non-leaf variable object, but it's generally a string
28940 that only indicates the type of the object, and does not list its
28941 contents. Assignment to a non-leaf variable object is not allowed.
28943 A frontend does not need to read the values of all variable objects each time
28944 the program stops. Instead, MI provides an update command that lists all
28945 variable objects whose values has changed since the last update
28946 operation. This considerably reduces the amount of data that must
28947 be transferred to the frontend. As noted above, children variable
28948 objects are created on demand, and only leaf variable objects have a
28949 real value. As result, gdb will read target memory only for leaf
28950 variables that frontend has created.
28952 The automatic update is not always desirable. For example, a frontend
28953 might want to keep a value of some expression for future reference,
28954 and never update it. For another example, fetching memory is
28955 relatively slow for embedded targets, so a frontend might want
28956 to disable automatic update for the variables that are either not
28957 visible on the screen, or ``closed''. This is possible using so
28958 called ``frozen variable objects''. Such variable objects are never
28959 implicitly updated.
28961 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
28962 fixed variable object, the expression is parsed when the variable
28963 object is created, including associating identifiers to specific
28964 variables. The meaning of expression never changes. For a floating
28965 variable object the values of variables whose names appear in the
28966 expressions are re-evaluated every time in the context of the current
28967 frame. Consider this example:
28972 struct work_state state;
28979 If a fixed variable object for the @code{state} variable is created in
28980 this function, and we enter the recursive call, the variable
28981 object will report the value of @code{state} in the top-level
28982 @code{do_work} invocation. On the other hand, a floating variable
28983 object will report the value of @code{state} in the current frame.
28985 If an expression specified when creating a fixed variable object
28986 refers to a local variable, the variable object becomes bound to the
28987 thread and frame in which the variable object is created. When such
28988 variable object is updated, @value{GDBN} makes sure that the
28989 thread/frame combination the variable object is bound to still exists,
28990 and re-evaluates the variable object in context of that thread/frame.
28992 The following is the complete set of @sc{gdb/mi} operations defined to
28993 access this functionality:
28995 @multitable @columnfractions .4 .6
28996 @item @strong{Operation}
28997 @tab @strong{Description}
28999 @item @code{-enable-pretty-printing}
29000 @tab enable Python-based pretty-printing
29001 @item @code{-var-create}
29002 @tab create a variable object
29003 @item @code{-var-delete}
29004 @tab delete the variable object and/or its children
29005 @item @code{-var-set-format}
29006 @tab set the display format of this variable
29007 @item @code{-var-show-format}
29008 @tab show the display format of this variable
29009 @item @code{-var-info-num-children}
29010 @tab tells how many children this object has
29011 @item @code{-var-list-children}
29012 @tab return a list of the object's children
29013 @item @code{-var-info-type}
29014 @tab show the type of this variable object
29015 @item @code{-var-info-expression}
29016 @tab print parent-relative expression that this variable object represents
29017 @item @code{-var-info-path-expression}
29018 @tab print full expression that this variable object represents
29019 @item @code{-var-show-attributes}
29020 @tab is this variable editable? does it exist here?
29021 @item @code{-var-evaluate-expression}
29022 @tab get the value of this variable
29023 @item @code{-var-assign}
29024 @tab set the value of this variable
29025 @item @code{-var-update}
29026 @tab update the variable and its children
29027 @item @code{-var-set-frozen}
29028 @tab set frozeness attribute
29029 @item @code{-var-set-update-range}
29030 @tab set range of children to display on update
29033 In the next subsection we describe each operation in detail and suggest
29034 how it can be used.
29036 @subheading Description And Use of Operations on Variable Objects
29038 @subheading The @code{-enable-pretty-printing} Command
29039 @findex -enable-pretty-printing
29042 -enable-pretty-printing
29045 @value{GDBN} allows Python-based visualizers to affect the output of the
29046 MI variable object commands. However, because there was no way to
29047 implement this in a fully backward-compatible way, a front end must
29048 request that this functionality be enabled.
29050 Once enabled, this feature cannot be disabled.
29052 Note that if Python support has not been compiled into @value{GDBN},
29053 this command will still succeed (and do nothing).
29055 This feature is currently (as of @value{GDBN} 7.0) experimental, and
29056 may work differently in future versions of @value{GDBN}.
29058 @subheading The @code{-var-create} Command
29059 @findex -var-create
29061 @subsubheading Synopsis
29064 -var-create @{@var{name} | "-"@}
29065 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
29068 This operation creates a variable object, which allows the monitoring of
29069 a variable, the result of an expression, a memory cell or a CPU
29072 The @var{name} parameter is the string by which the object can be
29073 referenced. It must be unique. If @samp{-} is specified, the varobj
29074 system will generate a string ``varNNNNNN'' automatically. It will be
29075 unique provided that one does not specify @var{name} of that format.
29076 The command fails if a duplicate name is found.
29078 The frame under which the expression should be evaluated can be
29079 specified by @var{frame-addr}. A @samp{*} indicates that the current
29080 frame should be used. A @samp{@@} indicates that a floating variable
29081 object must be created.
29083 @var{expression} is any expression valid on the current language set (must not
29084 begin with a @samp{*}), or one of the following:
29088 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
29091 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
29094 @samp{$@var{regname}} --- a CPU register name
29097 @cindex dynamic varobj
29098 A varobj's contents may be provided by a Python-based pretty-printer. In this
29099 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
29100 have slightly different semantics in some cases. If the
29101 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
29102 will never create a dynamic varobj. This ensures backward
29103 compatibility for existing clients.
29105 @subsubheading Result
29107 This operation returns attributes of the newly-created varobj. These
29112 The name of the varobj.
29115 The number of children of the varobj. This number is not necessarily
29116 reliable for a dynamic varobj. Instead, you must examine the
29117 @samp{has_more} attribute.
29120 The varobj's scalar value. For a varobj whose type is some sort of
29121 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
29122 will not be interesting.
29125 The varobj's type. This is a string representation of the type, as
29126 would be printed by the @value{GDBN} CLI. If @samp{print object}
29127 (@pxref{Print Settings, set print object}) is set to @code{on}, the
29128 @emph{actual} (derived) type of the object is shown rather than the
29129 @emph{declared} one.
29132 If a variable object is bound to a specific thread, then this is the
29133 thread's global identifier.
29136 For a dynamic varobj, this indicates whether there appear to be any
29137 children available. For a non-dynamic varobj, this will be 0.
29140 This attribute will be present and have the value @samp{1} if the
29141 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29142 then this attribute will not be present.
29145 A dynamic varobj can supply a display hint to the front end. The
29146 value comes directly from the Python pretty-printer object's
29147 @code{display_hint} method. @xref{Pretty Printing API}.
29150 Typical output will look like this:
29153 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
29154 has_more="@var{has_more}"
29158 @subheading The @code{-var-delete} Command
29159 @findex -var-delete
29161 @subsubheading Synopsis
29164 -var-delete [ -c ] @var{name}
29167 Deletes a previously created variable object and all of its children.
29168 With the @samp{-c} option, just deletes the children.
29170 Returns an error if the object @var{name} is not found.
29173 @subheading The @code{-var-set-format} Command
29174 @findex -var-set-format
29176 @subsubheading Synopsis
29179 -var-set-format @var{name} @var{format-spec}
29182 Sets the output format for the value of the object @var{name} to be
29185 @anchor{-var-set-format}
29186 The syntax for the @var{format-spec} is as follows:
29189 @var{format-spec} @expansion{}
29190 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
29193 The natural format is the default format choosen automatically
29194 based on the variable type (like decimal for an @code{int}, hex
29195 for pointers, etc.).
29197 The zero-hexadecimal format has a representation similar to hexadecimal
29198 but with padding zeroes to the left of the value. For example, a 32-bit
29199 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
29200 zero-hexadecimal format.
29202 For a variable with children, the format is set only on the
29203 variable itself, and the children are not affected.
29205 @subheading The @code{-var-show-format} Command
29206 @findex -var-show-format
29208 @subsubheading Synopsis
29211 -var-show-format @var{name}
29214 Returns the format used to display the value of the object @var{name}.
29217 @var{format} @expansion{}
29222 @subheading The @code{-var-info-num-children} Command
29223 @findex -var-info-num-children
29225 @subsubheading Synopsis
29228 -var-info-num-children @var{name}
29231 Returns the number of children of a variable object @var{name}:
29237 Note that this number is not completely reliable for a dynamic varobj.
29238 It will return the current number of children, but more children may
29242 @subheading The @code{-var-list-children} Command
29243 @findex -var-list-children
29245 @subsubheading Synopsis
29248 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
29250 @anchor{-var-list-children}
29252 Return a list of the children of the specified variable object and
29253 create variable objects for them, if they do not already exist. With
29254 a single argument or if @var{print-values} has a value of 0 or
29255 @code{--no-values}, print only the names of the variables; if
29256 @var{print-values} is 1 or @code{--all-values}, also print their
29257 values; and if it is 2 or @code{--simple-values} print the name and
29258 value for simple data types and just the name for arrays, structures
29261 @var{from} and @var{to}, if specified, indicate the range of children
29262 to report. If @var{from} or @var{to} is less than zero, the range is
29263 reset and all children will be reported. Otherwise, children starting
29264 at @var{from} (zero-based) and up to and excluding @var{to} will be
29267 If a child range is requested, it will only affect the current call to
29268 @code{-var-list-children}, but not future calls to @code{-var-update}.
29269 For this, you must instead use @code{-var-set-update-range}. The
29270 intent of this approach is to enable a front end to implement any
29271 update approach it likes; for example, scrolling a view may cause the
29272 front end to request more children with @code{-var-list-children}, and
29273 then the front end could call @code{-var-set-update-range} with a
29274 different range to ensure that future updates are restricted to just
29277 For each child the following results are returned:
29282 Name of the variable object created for this child.
29285 The expression to be shown to the user by the front end to designate this child.
29286 For example this may be the name of a structure member.
29288 For a dynamic varobj, this value cannot be used to form an
29289 expression. There is no way to do this at all with a dynamic varobj.
29291 For C/C@t{++} structures there are several pseudo children returned to
29292 designate access qualifiers. For these pseudo children @var{exp} is
29293 @samp{public}, @samp{private}, or @samp{protected}. In this case the
29294 type and value are not present.
29296 A dynamic varobj will not report the access qualifying
29297 pseudo-children, regardless of the language. This information is not
29298 available at all with a dynamic varobj.
29301 Number of children this child has. For a dynamic varobj, this will be
29305 The type of the child. If @samp{print object}
29306 (@pxref{Print Settings, set print object}) is set to @code{on}, the
29307 @emph{actual} (derived) type of the object is shown rather than the
29308 @emph{declared} one.
29311 If values were requested, this is the value.
29314 If this variable object is associated with a thread, this is the
29315 thread's global thread id. Otherwise this result is not present.
29318 If the variable object is frozen, this variable will be present with a value of 1.
29321 A dynamic varobj can supply a display hint to the front end. The
29322 value comes directly from the Python pretty-printer object's
29323 @code{display_hint} method. @xref{Pretty Printing API}.
29326 This attribute will be present and have the value @samp{1} if the
29327 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29328 then this attribute will not be present.
29332 The result may have its own attributes:
29336 A dynamic varobj can supply a display hint to the front end. The
29337 value comes directly from the Python pretty-printer object's
29338 @code{display_hint} method. @xref{Pretty Printing API}.
29341 This is an integer attribute which is nonzero if there are children
29342 remaining after the end of the selected range.
29345 @subsubheading Example
29349 -var-list-children n
29350 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
29351 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
29353 -var-list-children --all-values n
29354 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
29355 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
29359 @subheading The @code{-var-info-type} Command
29360 @findex -var-info-type
29362 @subsubheading Synopsis
29365 -var-info-type @var{name}
29368 Returns the type of the specified variable @var{name}. The type is
29369 returned as a string in the same format as it is output by the
29373 type=@var{typename}
29377 @subheading The @code{-var-info-expression} Command
29378 @findex -var-info-expression
29380 @subsubheading Synopsis
29383 -var-info-expression @var{name}
29386 Returns a string that is suitable for presenting this
29387 variable object in user interface. The string is generally
29388 not valid expression in the current language, and cannot be evaluated.
29390 For example, if @code{a} is an array, and variable object
29391 @code{A} was created for @code{a}, then we'll get this output:
29394 (gdb) -var-info-expression A.1
29395 ^done,lang="C",exp="1"
29399 Here, the value of @code{lang} is the language name, which can be
29400 found in @ref{Supported Languages}.
29402 Note that the output of the @code{-var-list-children} command also
29403 includes those expressions, so the @code{-var-info-expression} command
29406 @subheading The @code{-var-info-path-expression} Command
29407 @findex -var-info-path-expression
29409 @subsubheading Synopsis
29412 -var-info-path-expression @var{name}
29415 Returns an expression that can be evaluated in the current
29416 context and will yield the same value that a variable object has.
29417 Compare this with the @code{-var-info-expression} command, which
29418 result can be used only for UI presentation. Typical use of
29419 the @code{-var-info-path-expression} command is creating a
29420 watchpoint from a variable object.
29422 This command is currently not valid for children of a dynamic varobj,
29423 and will give an error when invoked on one.
29425 For example, suppose @code{C} is a C@t{++} class, derived from class
29426 @code{Base}, and that the @code{Base} class has a member called
29427 @code{m_size}. Assume a variable @code{c} is has the type of
29428 @code{C} and a variable object @code{C} was created for variable
29429 @code{c}. Then, we'll get this output:
29431 (gdb) -var-info-path-expression C.Base.public.m_size
29432 ^done,path_expr=((Base)c).m_size)
29435 @subheading The @code{-var-show-attributes} Command
29436 @findex -var-show-attributes
29438 @subsubheading Synopsis
29441 -var-show-attributes @var{name}
29444 List attributes of the specified variable object @var{name}:
29447 status=@var{attr} [ ( ,@var{attr} )* ]
29451 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
29453 @subheading The @code{-var-evaluate-expression} Command
29454 @findex -var-evaluate-expression
29456 @subsubheading Synopsis
29459 -var-evaluate-expression [-f @var{format-spec}] @var{name}
29462 Evaluates the expression that is represented by the specified variable
29463 object and returns its value as a string. The format of the string
29464 can be specified with the @samp{-f} option. The possible values of
29465 this option are the same as for @code{-var-set-format}
29466 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
29467 the current display format will be used. The current display format
29468 can be changed using the @code{-var-set-format} command.
29474 Note that one must invoke @code{-var-list-children} for a variable
29475 before the value of a child variable can be evaluated.
29477 @subheading The @code{-var-assign} Command
29478 @findex -var-assign
29480 @subsubheading Synopsis
29483 -var-assign @var{name} @var{expression}
29486 Assigns the value of @var{expression} to the variable object specified
29487 by @var{name}. The object must be @samp{editable}. If the variable's
29488 value is altered by the assign, the variable will show up in any
29489 subsequent @code{-var-update} list.
29491 @subsubheading Example
29499 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
29503 @subheading The @code{-var-update} Command
29504 @findex -var-update
29506 @subsubheading Synopsis
29509 -var-update [@var{print-values}] @{@var{name} | "*"@}
29512 Reevaluate the expressions corresponding to the variable object
29513 @var{name} and all its direct and indirect children, and return the
29514 list of variable objects whose values have changed; @var{name} must
29515 be a root variable object. Here, ``changed'' means that the result of
29516 @code{-var-evaluate-expression} before and after the
29517 @code{-var-update} is different. If @samp{*} is used as the variable
29518 object names, all existing variable objects are updated, except
29519 for frozen ones (@pxref{-var-set-frozen}). The option
29520 @var{print-values} determines whether both names and values, or just
29521 names are printed. The possible values of this option are the same
29522 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
29523 recommended to use the @samp{--all-values} option, to reduce the
29524 number of MI commands needed on each program stop.
29526 With the @samp{*} parameter, if a variable object is bound to a
29527 currently running thread, it will not be updated, without any
29530 If @code{-var-set-update-range} was previously used on a varobj, then
29531 only the selected range of children will be reported.
29533 @code{-var-update} reports all the changed varobjs in a tuple named
29536 Each item in the change list is itself a tuple holding:
29540 The name of the varobj.
29543 If values were requested for this update, then this field will be
29544 present and will hold the value of the varobj.
29547 @anchor{-var-update}
29548 This field is a string which may take one of three values:
29552 The variable object's current value is valid.
29555 The variable object does not currently hold a valid value but it may
29556 hold one in the future if its associated expression comes back into
29560 The variable object no longer holds a valid value.
29561 This can occur when the executable file being debugged has changed,
29562 either through recompilation or by using the @value{GDBN} @code{file}
29563 command. The front end should normally choose to delete these variable
29567 In the future new values may be added to this list so the front should
29568 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
29571 This is only present if the varobj is still valid. If the type
29572 changed, then this will be the string @samp{true}; otherwise it will
29575 When a varobj's type changes, its children are also likely to have
29576 become incorrect. Therefore, the varobj's children are automatically
29577 deleted when this attribute is @samp{true}. Also, the varobj's update
29578 range, when set using the @code{-var-set-update-range} command, is
29582 If the varobj's type changed, then this field will be present and will
29585 @item new_num_children
29586 For a dynamic varobj, if the number of children changed, or if the
29587 type changed, this will be the new number of children.
29589 The @samp{numchild} field in other varobj responses is generally not
29590 valid for a dynamic varobj -- it will show the number of children that
29591 @value{GDBN} knows about, but because dynamic varobjs lazily
29592 instantiate their children, this will not reflect the number of
29593 children which may be available.
29595 The @samp{new_num_children} attribute only reports changes to the
29596 number of children known by @value{GDBN}. This is the only way to
29597 detect whether an update has removed children (which necessarily can
29598 only happen at the end of the update range).
29601 The display hint, if any.
29604 This is an integer value, which will be 1 if there are more children
29605 available outside the varobj's update range.
29608 This attribute will be present and have the value @samp{1} if the
29609 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29610 then this attribute will not be present.
29613 If new children were added to a dynamic varobj within the selected
29614 update range (as set by @code{-var-set-update-range}), then they will
29615 be listed in this attribute.
29618 @subsubheading Example
29625 -var-update --all-values var1
29626 ^done,changelist=[@{name="var1",value="3",in_scope="true",
29627 type_changed="false"@}]
29631 @subheading The @code{-var-set-frozen} Command
29632 @findex -var-set-frozen
29633 @anchor{-var-set-frozen}
29635 @subsubheading Synopsis
29638 -var-set-frozen @var{name} @var{flag}
29641 Set the frozenness flag on the variable object @var{name}. The
29642 @var{flag} parameter should be either @samp{1} to make the variable
29643 frozen or @samp{0} to make it unfrozen. If a variable object is
29644 frozen, then neither itself, nor any of its children, are
29645 implicitly updated by @code{-var-update} of
29646 a parent variable or by @code{-var-update *}. Only
29647 @code{-var-update} of the variable itself will update its value and
29648 values of its children. After a variable object is unfrozen, it is
29649 implicitly updated by all subsequent @code{-var-update} operations.
29650 Unfreezing a variable does not update it, only subsequent
29651 @code{-var-update} does.
29653 @subsubheading Example
29657 -var-set-frozen V 1
29662 @subheading The @code{-var-set-update-range} command
29663 @findex -var-set-update-range
29664 @anchor{-var-set-update-range}
29666 @subsubheading Synopsis
29669 -var-set-update-range @var{name} @var{from} @var{to}
29672 Set the range of children to be returned by future invocations of
29673 @code{-var-update}.
29675 @var{from} and @var{to} indicate the range of children to report. If
29676 @var{from} or @var{to} is less than zero, the range is reset and all
29677 children will be reported. Otherwise, children starting at @var{from}
29678 (zero-based) and up to and excluding @var{to} will be reported.
29680 @subsubheading Example
29684 -var-set-update-range V 1 2
29688 @subheading The @code{-var-set-visualizer} command
29689 @findex -var-set-visualizer
29690 @anchor{-var-set-visualizer}
29692 @subsubheading Synopsis
29695 -var-set-visualizer @var{name} @var{visualizer}
29698 Set a visualizer for the variable object @var{name}.
29700 @var{visualizer} is the visualizer to use. The special value
29701 @samp{None} means to disable any visualizer in use.
29703 If not @samp{None}, @var{visualizer} must be a Python expression.
29704 This expression must evaluate to a callable object which accepts a
29705 single argument. @value{GDBN} will call this object with the value of
29706 the varobj @var{name} as an argument (this is done so that the same
29707 Python pretty-printing code can be used for both the CLI and MI).
29708 When called, this object must return an object which conforms to the
29709 pretty-printing interface (@pxref{Pretty Printing API}).
29711 The pre-defined function @code{gdb.default_visualizer} may be used to
29712 select a visualizer by following the built-in process
29713 (@pxref{Selecting Pretty-Printers}). This is done automatically when
29714 a varobj is created, and so ordinarily is not needed.
29716 This feature is only available if Python support is enabled. The MI
29717 command @code{-list-features} (@pxref{GDB/MI Support Commands})
29718 can be used to check this.
29720 @subsubheading Example
29722 Resetting the visualizer:
29726 -var-set-visualizer V None
29730 Reselecting the default (type-based) visualizer:
29734 -var-set-visualizer V gdb.default_visualizer
29738 Suppose @code{SomeClass} is a visualizer class. A lambda expression
29739 can be used to instantiate this class for a varobj:
29743 -var-set-visualizer V "lambda val: SomeClass()"
29747 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29748 @node GDB/MI Data Manipulation
29749 @section @sc{gdb/mi} Data Manipulation
29751 @cindex data manipulation, in @sc{gdb/mi}
29752 @cindex @sc{gdb/mi}, data manipulation
29753 This section describes the @sc{gdb/mi} commands that manipulate data:
29754 examine memory and registers, evaluate expressions, etc.
29756 For details about what an addressable memory unit is,
29757 @pxref{addressable memory unit}.
29759 @c REMOVED FROM THE INTERFACE.
29760 @c @subheading -data-assign
29761 @c Change the value of a program variable. Plenty of side effects.
29762 @c @subsubheading GDB Command
29764 @c @subsubheading Example
29767 @subheading The @code{-data-disassemble} Command
29768 @findex -data-disassemble
29770 @subsubheading Synopsis
29774 [ -s @var{start-addr} -e @var{end-addr} ]
29775 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
29783 @item @var{start-addr}
29784 is the beginning address (or @code{$pc})
29785 @item @var{end-addr}
29787 @item @var{filename}
29788 is the name of the file to disassemble
29789 @item @var{linenum}
29790 is the line number to disassemble around
29792 is the number of disassembly lines to be produced. If it is -1,
29793 the whole function will be disassembled, in case no @var{end-addr} is
29794 specified. If @var{end-addr} is specified as a non-zero value, and
29795 @var{lines} is lower than the number of disassembly lines between
29796 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
29797 displayed; if @var{lines} is higher than the number of lines between
29798 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
29803 @item 0 disassembly only
29804 @item 1 mixed source and disassembly (deprecated)
29805 @item 2 disassembly with raw opcodes
29806 @item 3 mixed source and disassembly with raw opcodes (deprecated)
29807 @item 4 mixed source and disassembly
29808 @item 5 mixed source and disassembly with raw opcodes
29811 Modes 1 and 3 are deprecated. The output is ``source centric''
29812 which hasn't proved useful in practice.
29813 @xref{Machine Code}, for a discussion of the difference between
29814 @code{/m} and @code{/s} output of the @code{disassemble} command.
29817 @subsubheading Result
29819 The result of the @code{-data-disassemble} command will be a list named
29820 @samp{asm_insns}, the contents of this list depend on the @var{mode}
29821 used with the @code{-data-disassemble} command.
29823 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
29828 The address at which this instruction was disassembled.
29831 The name of the function this instruction is within.
29834 The decimal offset in bytes from the start of @samp{func-name}.
29837 The text disassembly for this @samp{address}.
29840 This field is only present for modes 2, 3 and 5. This contains the raw opcode
29841 bytes for the @samp{inst} field.
29845 For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
29846 @samp{src_and_asm_line}, each of which has the following fields:
29850 The line number within @samp{file}.
29853 The file name from the compilation unit. This might be an absolute
29854 file name or a relative file name depending on the compile command
29858 Absolute file name of @samp{file}. It is converted to a canonical form
29859 using the source file search path
29860 (@pxref{Source Path, ,Specifying Source Directories})
29861 and after resolving all the symbolic links.
29863 If the source file is not found this field will contain the path as
29864 present in the debug information.
29866 @item line_asm_insn
29867 This is a list of tuples containing the disassembly for @samp{line} in
29868 @samp{file}. The fields of each tuple are the same as for
29869 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
29870 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
29875 Note that whatever included in the @samp{inst} field, is not
29876 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
29879 @subsubheading @value{GDBN} Command
29881 The corresponding @value{GDBN} command is @samp{disassemble}.
29883 @subsubheading Example
29885 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
29889 -data-disassemble -s $pc -e "$pc + 20" -- 0
29892 @{address="0x000107c0",func-name="main",offset="4",
29893 inst="mov 2, %o0"@},
29894 @{address="0x000107c4",func-name="main",offset="8",
29895 inst="sethi %hi(0x11800), %o2"@},
29896 @{address="0x000107c8",func-name="main",offset="12",
29897 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
29898 @{address="0x000107cc",func-name="main",offset="16",
29899 inst="sethi %hi(0x11800), %o2"@},
29900 @{address="0x000107d0",func-name="main",offset="20",
29901 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
29905 Disassemble the whole @code{main} function. Line 32 is part of
29909 -data-disassemble -f basics.c -l 32 -- 0
29911 @{address="0x000107bc",func-name="main",offset="0",
29912 inst="save %sp, -112, %sp"@},
29913 @{address="0x000107c0",func-name="main",offset="4",
29914 inst="mov 2, %o0"@},
29915 @{address="0x000107c4",func-name="main",offset="8",
29916 inst="sethi %hi(0x11800), %o2"@},
29918 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
29919 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
29923 Disassemble 3 instructions from the start of @code{main}:
29927 -data-disassemble -f basics.c -l 32 -n 3 -- 0
29929 @{address="0x000107bc",func-name="main",offset="0",
29930 inst="save %sp, -112, %sp"@},
29931 @{address="0x000107c0",func-name="main",offset="4",
29932 inst="mov 2, %o0"@},
29933 @{address="0x000107c4",func-name="main",offset="8",
29934 inst="sethi %hi(0x11800), %o2"@}]
29938 Disassemble 3 instructions from the start of @code{main} in mixed mode:
29942 -data-disassemble -f basics.c -l 32 -n 3 -- 1
29944 src_and_asm_line=@{line="31",
29945 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
29946 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
29947 line_asm_insn=[@{address="0x000107bc",
29948 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
29949 src_and_asm_line=@{line="32",
29950 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
29951 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
29952 line_asm_insn=[@{address="0x000107c0",
29953 func-name="main",offset="4",inst="mov 2, %o0"@},
29954 @{address="0x000107c4",func-name="main",offset="8",
29955 inst="sethi %hi(0x11800), %o2"@}]@}]
29960 @subheading The @code{-data-evaluate-expression} Command
29961 @findex -data-evaluate-expression
29963 @subsubheading Synopsis
29966 -data-evaluate-expression @var{expr}
29969 Evaluate @var{expr} as an expression. The expression could contain an
29970 inferior function call. The function call will execute synchronously.
29971 If the expression contains spaces, it must be enclosed in double quotes.
29973 @subsubheading @value{GDBN} Command
29975 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
29976 @samp{call}. In @code{gdbtk} only, there's a corresponding
29977 @samp{gdb_eval} command.
29979 @subsubheading Example
29981 In the following example, the numbers that precede the commands are the
29982 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
29983 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
29987 211-data-evaluate-expression A
29990 311-data-evaluate-expression &A
29991 311^done,value="0xefffeb7c"
29993 411-data-evaluate-expression A+3
29996 511-data-evaluate-expression "A + 3"
30002 @subheading The @code{-data-list-changed-registers} Command
30003 @findex -data-list-changed-registers
30005 @subsubheading Synopsis
30008 -data-list-changed-registers
30011 Display a list of the registers that have changed.
30013 @subsubheading @value{GDBN} Command
30015 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
30016 has the corresponding command @samp{gdb_changed_register_list}.
30018 @subsubheading Example
30020 On a PPC MBX board:
30028 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
30029 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
30032 -data-list-changed-registers
30033 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
30034 "10","11","13","14","15","16","17","18","19","20","21","22","23",
30035 "24","25","26","27","28","30","31","64","65","66","67","69"]
30040 @subheading The @code{-data-list-register-names} Command
30041 @findex -data-list-register-names
30043 @subsubheading Synopsis
30046 -data-list-register-names [ ( @var{regno} )+ ]
30049 Show a list of register names for the current target. If no arguments
30050 are given, it shows a list of the names of all the registers. If
30051 integer numbers are given as arguments, it will print a list of the
30052 names of the registers corresponding to the arguments. To ensure
30053 consistency between a register name and its number, the output list may
30054 include empty register names.
30056 @subsubheading @value{GDBN} Command
30058 @value{GDBN} does not have a command which corresponds to
30059 @samp{-data-list-register-names}. In @code{gdbtk} there is a
30060 corresponding command @samp{gdb_regnames}.
30062 @subsubheading Example
30064 For the PPC MBX board:
30067 -data-list-register-names
30068 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
30069 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
30070 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
30071 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
30072 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
30073 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
30074 "", "pc","ps","cr","lr","ctr","xer"]
30076 -data-list-register-names 1 2 3
30077 ^done,register-names=["r1","r2","r3"]
30081 @subheading The @code{-data-list-register-values} Command
30082 @findex -data-list-register-values
30084 @subsubheading Synopsis
30087 -data-list-register-values
30088 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
30091 Display the registers' contents. The format according to which the
30092 registers' contents are to be returned is given by @var{fmt}, followed
30093 by an optional list of numbers specifying the registers to display. A
30094 missing list of numbers indicates that the contents of all the
30095 registers must be returned. The @code{--skip-unavailable} option
30096 indicates that only the available registers are to be returned.
30098 Allowed formats for @var{fmt} are:
30115 @subsubheading @value{GDBN} Command
30117 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
30118 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
30120 @subsubheading Example
30122 For a PPC MBX board (note: line breaks are for readability only, they
30123 don't appear in the actual output):
30127 -data-list-register-values r 64 65
30128 ^done,register-values=[@{number="64",value="0xfe00a300"@},
30129 @{number="65",value="0x00029002"@}]
30131 -data-list-register-values x
30132 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
30133 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
30134 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
30135 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
30136 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
30137 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
30138 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
30139 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
30140 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
30141 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
30142 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
30143 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
30144 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
30145 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
30146 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
30147 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
30148 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
30149 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
30150 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
30151 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
30152 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
30153 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
30154 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
30155 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
30156 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
30157 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
30158 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
30159 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
30160 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
30161 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
30162 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
30163 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
30164 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
30165 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
30166 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
30167 @{number="69",value="0x20002b03"@}]
30172 @subheading The @code{-data-read-memory} Command
30173 @findex -data-read-memory
30175 This command is deprecated, use @code{-data-read-memory-bytes} instead.
30177 @subsubheading Synopsis
30180 -data-read-memory [ -o @var{byte-offset} ]
30181 @var{address} @var{word-format} @var{word-size}
30182 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
30189 @item @var{address}
30190 An expression specifying the address of the first memory word to be
30191 read. Complex expressions containing embedded white space should be
30192 quoted using the C convention.
30194 @item @var{word-format}
30195 The format to be used to print the memory words. The notation is the
30196 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
30199 @item @var{word-size}
30200 The size of each memory word in bytes.
30202 @item @var{nr-rows}
30203 The number of rows in the output table.
30205 @item @var{nr-cols}
30206 The number of columns in the output table.
30209 If present, indicates that each row should include an @sc{ascii} dump. The
30210 value of @var{aschar} is used as a padding character when a byte is not a
30211 member of the printable @sc{ascii} character set (printable @sc{ascii}
30212 characters are those whose code is between 32 and 126, inclusively).
30214 @item @var{byte-offset}
30215 An offset to add to the @var{address} before fetching memory.
30218 This command displays memory contents as a table of @var{nr-rows} by
30219 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
30220 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
30221 (returned as @samp{total-bytes}). Should less than the requested number
30222 of bytes be returned by the target, the missing words are identified
30223 using @samp{N/A}. The number of bytes read from the target is returned
30224 in @samp{nr-bytes} and the starting address used to read memory in
30227 The address of the next/previous row or page is available in
30228 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
30231 @subsubheading @value{GDBN} Command
30233 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
30234 @samp{gdb_get_mem} memory read command.
30236 @subsubheading Example
30238 Read six bytes of memory starting at @code{bytes+6} but then offset by
30239 @code{-6} bytes. Format as three rows of two columns. One byte per
30240 word. Display each word in hex.
30244 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
30245 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
30246 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
30247 prev-page="0x0000138a",memory=[
30248 @{addr="0x00001390",data=["0x00","0x01"]@},
30249 @{addr="0x00001392",data=["0x02","0x03"]@},
30250 @{addr="0x00001394",data=["0x04","0x05"]@}]
30254 Read two bytes of memory starting at address @code{shorts + 64} and
30255 display as a single word formatted in decimal.
30259 5-data-read-memory shorts+64 d 2 1 1
30260 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
30261 next-row="0x00001512",prev-row="0x0000150e",
30262 next-page="0x00001512",prev-page="0x0000150e",memory=[
30263 @{addr="0x00001510",data=["128"]@}]
30267 Read thirty two bytes of memory starting at @code{bytes+16} and format
30268 as eight rows of four columns. Include a string encoding with @samp{x}
30269 used as the non-printable character.
30273 4-data-read-memory bytes+16 x 1 8 4 x
30274 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
30275 next-row="0x000013c0",prev-row="0x0000139c",
30276 next-page="0x000013c0",prev-page="0x00001380",memory=[
30277 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
30278 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
30279 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
30280 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
30281 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
30282 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
30283 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
30284 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
30288 @subheading The @code{-data-read-memory-bytes} Command
30289 @findex -data-read-memory-bytes
30291 @subsubheading Synopsis
30294 -data-read-memory-bytes [ -o @var{offset} ]
30295 @var{address} @var{count}
30302 @item @var{address}
30303 An expression specifying the address of the first addressable memory unit
30304 to be read. Complex expressions containing embedded white space should be
30305 quoted using the C convention.
30308 The number of addressable memory units to read. This should be an integer
30312 The offset relative to @var{address} at which to start reading. This
30313 should be an integer literal. This option is provided so that a frontend
30314 is not required to first evaluate address and then perform address
30315 arithmetics itself.
30319 This command attempts to read all accessible memory regions in the
30320 specified range. First, all regions marked as unreadable in the memory
30321 map (if one is defined) will be skipped. @xref{Memory Region
30322 Attributes}. Second, @value{GDBN} will attempt to read the remaining
30323 regions. For each one, if reading full region results in an errors,
30324 @value{GDBN} will try to read a subset of the region.
30326 In general, every single memory unit in the region may be readable or not,
30327 and the only way to read every readable unit is to try a read at
30328 every address, which is not practical. Therefore, @value{GDBN} will
30329 attempt to read all accessible memory units at either beginning or the end
30330 of the region, using a binary division scheme. This heuristic works
30331 well for reading accross a memory map boundary. Note that if a region
30332 has a readable range that is neither at the beginning or the end,
30333 @value{GDBN} will not read it.
30335 The result record (@pxref{GDB/MI Result Records}) that is output of
30336 the command includes a field named @samp{memory} whose content is a
30337 list of tuples. Each tuple represent a successfully read memory block
30338 and has the following fields:
30342 The start address of the memory block, as hexadecimal literal.
30345 The end address of the memory block, as hexadecimal literal.
30348 The offset of the memory block, as hexadecimal literal, relative to
30349 the start address passed to @code{-data-read-memory-bytes}.
30352 The contents of the memory block, in hex.
30358 @subsubheading @value{GDBN} Command
30360 The corresponding @value{GDBN} command is @samp{x}.
30362 @subsubheading Example
30366 -data-read-memory-bytes &a 10
30367 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
30369 contents="01000000020000000300"@}]
30374 @subheading The @code{-data-write-memory-bytes} Command
30375 @findex -data-write-memory-bytes
30377 @subsubheading Synopsis
30380 -data-write-memory-bytes @var{address} @var{contents}
30381 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
30388 @item @var{address}
30389 An expression specifying the address of the first addressable memory unit
30390 to be written. Complex expressions containing embedded white space should
30391 be quoted using the C convention.
30393 @item @var{contents}
30394 The hex-encoded data to write. It is an error if @var{contents} does
30395 not represent an integral number of addressable memory units.
30398 Optional argument indicating the number of addressable memory units to be
30399 written. If @var{count} is greater than @var{contents}' length,
30400 @value{GDBN} will repeatedly write @var{contents} until it fills
30401 @var{count} memory units.
30405 @subsubheading @value{GDBN} Command
30407 There's no corresponding @value{GDBN} command.
30409 @subsubheading Example
30413 -data-write-memory-bytes &a "aabbccdd"
30420 -data-write-memory-bytes &a "aabbccdd" 16e
30425 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30426 @node GDB/MI Tracepoint Commands
30427 @section @sc{gdb/mi} Tracepoint Commands
30429 The commands defined in this section implement MI support for
30430 tracepoints. For detailed introduction, see @ref{Tracepoints}.
30432 @subheading The @code{-trace-find} Command
30433 @findex -trace-find
30435 @subsubheading Synopsis
30438 -trace-find @var{mode} [@var{parameters}@dots{}]
30441 Find a trace frame using criteria defined by @var{mode} and
30442 @var{parameters}. The following table lists permissible
30443 modes and their parameters. For details of operation, see @ref{tfind}.
30448 No parameters are required. Stops examining trace frames.
30451 An integer is required as parameter. Selects tracepoint frame with
30454 @item tracepoint-number
30455 An integer is required as parameter. Finds next
30456 trace frame that corresponds to tracepoint with the specified number.
30459 An address is required as parameter. Finds
30460 next trace frame that corresponds to any tracepoint at the specified
30463 @item pc-inside-range
30464 Two addresses are required as parameters. Finds next trace
30465 frame that corresponds to a tracepoint at an address inside the
30466 specified range. Both bounds are considered to be inside the range.
30468 @item pc-outside-range
30469 Two addresses are required as parameters. Finds
30470 next trace frame that corresponds to a tracepoint at an address outside
30471 the specified range. Both bounds are considered to be inside the range.
30474 Line specification is required as parameter. @xref{Specify Location}.
30475 Finds next trace frame that corresponds to a tracepoint at
30476 the specified location.
30480 If @samp{none} was passed as @var{mode}, the response does not
30481 have fields. Otherwise, the response may have the following fields:
30485 This field has either @samp{0} or @samp{1} as the value, depending
30486 on whether a matching tracepoint was found.
30489 The index of the found traceframe. This field is present iff
30490 the @samp{found} field has value of @samp{1}.
30493 The index of the found tracepoint. This field is present iff
30494 the @samp{found} field has value of @samp{1}.
30497 The information about the frame corresponding to the found trace
30498 frame. This field is present only if a trace frame was found.
30499 @xref{GDB/MI Frame Information}, for description of this field.
30503 @subsubheading @value{GDBN} Command
30505 The corresponding @value{GDBN} command is @samp{tfind}.
30507 @subheading -trace-define-variable
30508 @findex -trace-define-variable
30510 @subsubheading Synopsis
30513 -trace-define-variable @var{name} [ @var{value} ]
30516 Create trace variable @var{name} if it does not exist. If
30517 @var{value} is specified, sets the initial value of the specified
30518 trace variable to that value. Note that the @var{name} should start
30519 with the @samp{$} character.
30521 @subsubheading @value{GDBN} Command
30523 The corresponding @value{GDBN} command is @samp{tvariable}.
30525 @subheading The @code{-trace-frame-collected} Command
30526 @findex -trace-frame-collected
30528 @subsubheading Synopsis
30531 -trace-frame-collected
30532 [--var-print-values @var{var_pval}]
30533 [--comp-print-values @var{comp_pval}]
30534 [--registers-format @var{regformat}]
30535 [--memory-contents]
30538 This command returns the set of collected objects, register names,
30539 trace state variable names, memory ranges and computed expressions
30540 that have been collected at a particular trace frame. The optional
30541 parameters to the command affect the output format in different ways.
30542 See the output description table below for more details.
30544 The reported names can be used in the normal manner to create
30545 varobjs and inspect the objects themselves. The items returned by
30546 this command are categorized so that it is clear which is a variable,
30547 which is a register, which is a trace state variable, which is a
30548 memory range and which is a computed expression.
30550 For instance, if the actions were
30552 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
30553 collect *(int*)0xaf02bef0@@40
30557 the object collected in its entirety would be @code{myVar}. The
30558 object @code{myArray} would be partially collected, because only the
30559 element at index @code{myIndex} would be collected. The remaining
30560 objects would be computed expressions.
30562 An example output would be:
30566 -trace-frame-collected
30568 explicit-variables=[@{name="myVar",value="1"@}],
30569 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
30570 @{name="myObj.field",value="0"@},
30571 @{name="myPtr->field",value="1"@},
30572 @{name="myCount + 2",value="3"@},
30573 @{name="$tvar1 + 1",value="43970027"@}],
30574 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
30575 @{number="1",value="0x0"@},
30576 @{number="2",value="0x4"@},
30578 @{number="125",value="0x0"@}],
30579 tvars=[@{name="$tvar1",current="43970026"@}],
30580 memory=[@{address="0x0000000000602264",length="4"@},
30581 @{address="0x0000000000615bc0",length="4"@}]
30588 @item explicit-variables
30589 The set of objects that have been collected in their entirety (as
30590 opposed to collecting just a few elements of an array or a few struct
30591 members). For each object, its name and value are printed.
30592 The @code{--var-print-values} option affects how or whether the value
30593 field is output. If @var{var_pval} is 0, then print only the names;
30594 if it is 1, print also their values; and if it is 2, print the name,
30595 type and value for simple data types, and the name and type for
30596 arrays, structures and unions.
30598 @item computed-expressions
30599 The set of computed expressions that have been collected at the
30600 current trace frame. The @code{--comp-print-values} option affects
30601 this set like the @code{--var-print-values} option affects the
30602 @code{explicit-variables} set. See above.
30605 The registers that have been collected at the current trace frame.
30606 For each register collected, the name and current value are returned.
30607 The value is formatted according to the @code{--registers-format}
30608 option. See the @command{-data-list-register-values} command for a
30609 list of the allowed formats. The default is @samp{x}.
30612 The trace state variables that have been collected at the current
30613 trace frame. For each trace state variable collected, the name and
30614 current value are returned.
30617 The set of memory ranges that have been collected at the current trace
30618 frame. Its content is a list of tuples. Each tuple represents a
30619 collected memory range and has the following fields:
30623 The start address of the memory range, as hexadecimal literal.
30626 The length of the memory range, as decimal literal.
30629 The contents of the memory block, in hex. This field is only present
30630 if the @code{--memory-contents} option is specified.
30636 @subsubheading @value{GDBN} Command
30638 There is no corresponding @value{GDBN} command.
30640 @subsubheading Example
30642 @subheading -trace-list-variables
30643 @findex -trace-list-variables
30645 @subsubheading Synopsis
30648 -trace-list-variables
30651 Return a table of all defined trace variables. Each element of the
30652 table has the following fields:
30656 The name of the trace variable. This field is always present.
30659 The initial value. This is a 64-bit signed integer. This
30660 field is always present.
30663 The value the trace variable has at the moment. This is a 64-bit
30664 signed integer. This field is absent iff current value is
30665 not defined, for example if the trace was never run, or is
30670 @subsubheading @value{GDBN} Command
30672 The corresponding @value{GDBN} command is @samp{tvariables}.
30674 @subsubheading Example
30678 -trace-list-variables
30679 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
30680 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
30681 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
30682 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
30683 body=[variable=@{name="$trace_timestamp",initial="0"@}
30684 variable=@{name="$foo",initial="10",current="15"@}]@}
30688 @subheading -trace-save
30689 @findex -trace-save
30691 @subsubheading Synopsis
30694 -trace-save [-r ] @var{filename}
30697 Saves the collected trace data to @var{filename}. Without the
30698 @samp{-r} option, the data is downloaded from the target and saved
30699 in a local file. With the @samp{-r} option the target is asked
30700 to perform the save.
30702 @subsubheading @value{GDBN} Command
30704 The corresponding @value{GDBN} command is @samp{tsave}.
30707 @subheading -trace-start
30708 @findex -trace-start
30710 @subsubheading Synopsis
30716 Starts a tracing experiments. The result of this command does not
30719 @subsubheading @value{GDBN} Command
30721 The corresponding @value{GDBN} command is @samp{tstart}.
30723 @subheading -trace-status
30724 @findex -trace-status
30726 @subsubheading Synopsis
30732 Obtains the status of a tracing experiment. The result may include
30733 the following fields:
30738 May have a value of either @samp{0}, when no tracing operations are
30739 supported, @samp{1}, when all tracing operations are supported, or
30740 @samp{file} when examining trace file. In the latter case, examining
30741 of trace frame is possible but new tracing experiement cannot be
30742 started. This field is always present.
30745 May have a value of either @samp{0} or @samp{1} depending on whether
30746 tracing experiement is in progress on target. This field is present
30747 if @samp{supported} field is not @samp{0}.
30750 Report the reason why the tracing was stopped last time. This field
30751 may be absent iff tracing was never stopped on target yet. The
30752 value of @samp{request} means the tracing was stopped as result of
30753 the @code{-trace-stop} command. The value of @samp{overflow} means
30754 the tracing buffer is full. The value of @samp{disconnection} means
30755 tracing was automatically stopped when @value{GDBN} has disconnected.
30756 The value of @samp{passcount} means tracing was stopped when a
30757 tracepoint was passed a maximal number of times for that tracepoint.
30758 This field is present if @samp{supported} field is not @samp{0}.
30760 @item stopping-tracepoint
30761 The number of tracepoint whose passcount as exceeded. This field is
30762 present iff the @samp{stop-reason} field has the value of
30766 @itemx frames-created
30767 The @samp{frames} field is a count of the total number of trace frames
30768 in the trace buffer, while @samp{frames-created} is the total created
30769 during the run, including ones that were discarded, such as when a
30770 circular trace buffer filled up. Both fields are optional.
30774 These fields tell the current size of the tracing buffer and the
30775 remaining space. These fields are optional.
30778 The value of the circular trace buffer flag. @code{1} means that the
30779 trace buffer is circular and old trace frames will be discarded if
30780 necessary to make room, @code{0} means that the trace buffer is linear
30784 The value of the disconnected tracing flag. @code{1} means that
30785 tracing will continue after @value{GDBN} disconnects, @code{0} means
30786 that the trace run will stop.
30789 The filename of the trace file being examined. This field is
30790 optional, and only present when examining a trace file.
30794 @subsubheading @value{GDBN} Command
30796 The corresponding @value{GDBN} command is @samp{tstatus}.
30798 @subheading -trace-stop
30799 @findex -trace-stop
30801 @subsubheading Synopsis
30807 Stops a tracing experiment. The result of this command has the same
30808 fields as @code{-trace-status}, except that the @samp{supported} and
30809 @samp{running} fields are not output.
30811 @subsubheading @value{GDBN} Command
30813 The corresponding @value{GDBN} command is @samp{tstop}.
30816 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30817 @node GDB/MI Symbol Query
30818 @section @sc{gdb/mi} Symbol Query Commands
30822 @subheading The @code{-symbol-info-address} Command
30823 @findex -symbol-info-address
30825 @subsubheading Synopsis
30828 -symbol-info-address @var{symbol}
30831 Describe where @var{symbol} is stored.
30833 @subsubheading @value{GDBN} Command
30835 The corresponding @value{GDBN} command is @samp{info address}.
30837 @subsubheading Example
30841 @subheading The @code{-symbol-info-file} Command
30842 @findex -symbol-info-file
30844 @subsubheading Synopsis
30850 Show the file for the symbol.
30852 @subsubheading @value{GDBN} Command
30854 There's no equivalent @value{GDBN} command. @code{gdbtk} has
30855 @samp{gdb_find_file}.
30857 @subsubheading Example
30861 @subheading The @code{-symbol-info-function} Command
30862 @findex -symbol-info-function
30864 @subsubheading Synopsis
30867 -symbol-info-function
30870 Show which function the symbol lives in.
30872 @subsubheading @value{GDBN} Command
30874 @samp{gdb_get_function} in @code{gdbtk}.
30876 @subsubheading Example
30880 @subheading The @code{-symbol-info-line} Command
30881 @findex -symbol-info-line
30883 @subsubheading Synopsis
30889 Show the core addresses of the code for a source line.
30891 @subsubheading @value{GDBN} Command
30893 The corresponding @value{GDBN} command is @samp{info line}.
30894 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
30896 @subsubheading Example
30900 @subheading The @code{-symbol-info-symbol} Command
30901 @findex -symbol-info-symbol
30903 @subsubheading Synopsis
30906 -symbol-info-symbol @var{addr}
30909 Describe what symbol is at location @var{addr}.
30911 @subsubheading @value{GDBN} Command
30913 The corresponding @value{GDBN} command is @samp{info symbol}.
30915 @subsubheading Example
30919 @subheading The @code{-symbol-list-functions} Command
30920 @findex -symbol-list-functions
30922 @subsubheading Synopsis
30925 -symbol-list-functions
30928 List the functions in the executable.
30930 @subsubheading @value{GDBN} Command
30932 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
30933 @samp{gdb_search} in @code{gdbtk}.
30935 @subsubheading Example
30940 @subheading The @code{-symbol-list-lines} Command
30941 @findex -symbol-list-lines
30943 @subsubheading Synopsis
30946 -symbol-list-lines @var{filename}
30949 Print the list of lines that contain code and their associated program
30950 addresses for the given source filename. The entries are sorted in
30951 ascending PC order.
30953 @subsubheading @value{GDBN} Command
30955 There is no corresponding @value{GDBN} command.
30957 @subsubheading Example
30960 -symbol-list-lines basics.c
30961 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
30967 @subheading The @code{-symbol-list-types} Command
30968 @findex -symbol-list-types
30970 @subsubheading Synopsis
30976 List all the type names.
30978 @subsubheading @value{GDBN} Command
30980 The corresponding commands are @samp{info types} in @value{GDBN},
30981 @samp{gdb_search} in @code{gdbtk}.
30983 @subsubheading Example
30987 @subheading The @code{-symbol-list-variables} Command
30988 @findex -symbol-list-variables
30990 @subsubheading Synopsis
30993 -symbol-list-variables
30996 List all the global and static variable names.
30998 @subsubheading @value{GDBN} Command
31000 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
31002 @subsubheading Example
31006 @subheading The @code{-symbol-locate} Command
31007 @findex -symbol-locate
31009 @subsubheading Synopsis
31015 @subsubheading @value{GDBN} Command
31017 @samp{gdb_loc} in @code{gdbtk}.
31019 @subsubheading Example
31023 @subheading The @code{-symbol-type} Command
31024 @findex -symbol-type
31026 @subsubheading Synopsis
31029 -symbol-type @var{variable}
31032 Show type of @var{variable}.
31034 @subsubheading @value{GDBN} Command
31036 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
31037 @samp{gdb_obj_variable}.
31039 @subsubheading Example
31044 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31045 @node GDB/MI File Commands
31046 @section @sc{gdb/mi} File Commands
31048 This section describes the GDB/MI commands to specify executable file names
31049 and to read in and obtain symbol table information.
31051 @subheading The @code{-file-exec-and-symbols} Command
31052 @findex -file-exec-and-symbols
31054 @subsubheading Synopsis
31057 -file-exec-and-symbols @var{file}
31060 Specify the executable file to be debugged. This file is the one from
31061 which the symbol table is also read. If no file is specified, the
31062 command clears the executable and symbol information. If breakpoints
31063 are set when using this command with no arguments, @value{GDBN} will produce
31064 error messages. Otherwise, no output is produced, except a completion
31067 @subsubheading @value{GDBN} Command
31069 The corresponding @value{GDBN} command is @samp{file}.
31071 @subsubheading Example
31075 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31081 @subheading The @code{-file-exec-file} Command
31082 @findex -file-exec-file
31084 @subsubheading Synopsis
31087 -file-exec-file @var{file}
31090 Specify the executable file to be debugged. Unlike
31091 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
31092 from this file. If used without argument, @value{GDBN} clears the information
31093 about the executable file. No output is produced, except a completion
31096 @subsubheading @value{GDBN} Command
31098 The corresponding @value{GDBN} command is @samp{exec-file}.
31100 @subsubheading Example
31104 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31111 @subheading The @code{-file-list-exec-sections} Command
31112 @findex -file-list-exec-sections
31114 @subsubheading Synopsis
31117 -file-list-exec-sections
31120 List the sections of the current executable file.
31122 @subsubheading @value{GDBN} Command
31124 The @value{GDBN} command @samp{info file} shows, among the rest, the same
31125 information as this command. @code{gdbtk} has a corresponding command
31126 @samp{gdb_load_info}.
31128 @subsubheading Example
31133 @subheading The @code{-file-list-exec-source-file} Command
31134 @findex -file-list-exec-source-file
31136 @subsubheading Synopsis
31139 -file-list-exec-source-file
31142 List the line number, the current source file, and the absolute path
31143 to the current source file for the current executable. The macro
31144 information field has a value of @samp{1} or @samp{0} depending on
31145 whether or not the file includes preprocessor macro information.
31147 @subsubheading @value{GDBN} Command
31149 The @value{GDBN} equivalent is @samp{info source}
31151 @subsubheading Example
31155 123-file-list-exec-source-file
31156 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
31161 @subheading The @code{-file-list-exec-source-files} Command
31162 @findex -file-list-exec-source-files
31164 @subsubheading Synopsis
31167 -file-list-exec-source-files
31170 List the source files for the current executable.
31172 It will always output both the filename and fullname (absolute file
31173 name) of a source file.
31175 @subsubheading @value{GDBN} Command
31177 The @value{GDBN} equivalent is @samp{info sources}.
31178 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
31180 @subsubheading Example
31183 -file-list-exec-source-files
31185 @{file=foo.c,fullname=/home/foo.c@},
31186 @{file=/home/bar.c,fullname=/home/bar.c@},
31187 @{file=gdb_could_not_find_fullpath.c@}]
31192 @subheading The @code{-file-list-shared-libraries} Command
31193 @findex -file-list-shared-libraries
31195 @subsubheading Synopsis
31198 -file-list-shared-libraries
31201 List the shared libraries in the program.
31203 @subsubheading @value{GDBN} Command
31205 The corresponding @value{GDBN} command is @samp{info shared}.
31207 @subsubheading Example
31211 @subheading The @code{-file-list-symbol-files} Command
31212 @findex -file-list-symbol-files
31214 @subsubheading Synopsis
31217 -file-list-symbol-files
31222 @subsubheading @value{GDBN} Command
31224 The corresponding @value{GDBN} command is @samp{info file} (part of it).
31226 @subsubheading Example
31231 @subheading The @code{-file-symbol-file} Command
31232 @findex -file-symbol-file
31234 @subsubheading Synopsis
31237 -file-symbol-file @var{file}
31240 Read symbol table info from the specified @var{file} argument. When
31241 used without arguments, clears @value{GDBN}'s symbol table info. No output is
31242 produced, except for a completion notification.
31244 @subsubheading @value{GDBN} Command
31246 The corresponding @value{GDBN} command is @samp{symbol-file}.
31248 @subsubheading Example
31252 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31258 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31259 @node GDB/MI Memory Overlay Commands
31260 @section @sc{gdb/mi} Memory Overlay Commands
31262 The memory overlay commands are not implemented.
31264 @c @subheading -overlay-auto
31266 @c @subheading -overlay-list-mapping-state
31268 @c @subheading -overlay-list-overlays
31270 @c @subheading -overlay-map
31272 @c @subheading -overlay-off
31274 @c @subheading -overlay-on
31276 @c @subheading -overlay-unmap
31278 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31279 @node GDB/MI Signal Handling Commands
31280 @section @sc{gdb/mi} Signal Handling Commands
31282 Signal handling commands are not implemented.
31284 @c @subheading -signal-handle
31286 @c @subheading -signal-list-handle-actions
31288 @c @subheading -signal-list-signal-types
31292 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31293 @node GDB/MI Target Manipulation
31294 @section @sc{gdb/mi} Target Manipulation Commands
31297 @subheading The @code{-target-attach} Command
31298 @findex -target-attach
31300 @subsubheading Synopsis
31303 -target-attach @var{pid} | @var{gid} | @var{file}
31306 Attach to a process @var{pid} or a file @var{file} outside of
31307 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
31308 group, the id previously returned by
31309 @samp{-list-thread-groups --available} must be used.
31311 @subsubheading @value{GDBN} Command
31313 The corresponding @value{GDBN} command is @samp{attach}.
31315 @subsubheading Example
31319 =thread-created,id="1"
31320 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
31326 @subheading The @code{-target-compare-sections} Command
31327 @findex -target-compare-sections
31329 @subsubheading Synopsis
31332 -target-compare-sections [ @var{section} ]
31335 Compare data of section @var{section} on target to the exec file.
31336 Without the argument, all sections are compared.
31338 @subsubheading @value{GDBN} Command
31340 The @value{GDBN} equivalent is @samp{compare-sections}.
31342 @subsubheading Example
31347 @subheading The @code{-target-detach} Command
31348 @findex -target-detach
31350 @subsubheading Synopsis
31353 -target-detach [ @var{pid} | @var{gid} ]
31356 Detach from the remote target which normally resumes its execution.
31357 If either @var{pid} or @var{gid} is specified, detaches from either
31358 the specified process, or specified thread group. There's no output.
31360 @subsubheading @value{GDBN} Command
31362 The corresponding @value{GDBN} command is @samp{detach}.
31364 @subsubheading Example
31374 @subheading The @code{-target-disconnect} Command
31375 @findex -target-disconnect
31377 @subsubheading Synopsis
31383 Disconnect from the remote target. There's no output and the target is
31384 generally not resumed.
31386 @subsubheading @value{GDBN} Command
31388 The corresponding @value{GDBN} command is @samp{disconnect}.
31390 @subsubheading Example
31400 @subheading The @code{-target-download} Command
31401 @findex -target-download
31403 @subsubheading Synopsis
31409 Loads the executable onto the remote target.
31410 It prints out an update message every half second, which includes the fields:
31414 The name of the section.
31416 The size of what has been sent so far for that section.
31418 The size of the section.
31420 The total size of what was sent so far (the current and the previous sections).
31422 The size of the overall executable to download.
31426 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
31427 @sc{gdb/mi} Output Syntax}).
31429 In addition, it prints the name and size of the sections, as they are
31430 downloaded. These messages include the following fields:
31434 The name of the section.
31436 The size of the section.
31438 The size of the overall executable to download.
31442 At the end, a summary is printed.
31444 @subsubheading @value{GDBN} Command
31446 The corresponding @value{GDBN} command is @samp{load}.
31448 @subsubheading Example
31450 Note: each status message appears on a single line. Here the messages
31451 have been broken down so that they can fit onto a page.
31456 +download,@{section=".text",section-size="6668",total-size="9880"@}
31457 +download,@{section=".text",section-sent="512",section-size="6668",
31458 total-sent="512",total-size="9880"@}
31459 +download,@{section=".text",section-sent="1024",section-size="6668",
31460 total-sent="1024",total-size="9880"@}
31461 +download,@{section=".text",section-sent="1536",section-size="6668",
31462 total-sent="1536",total-size="9880"@}
31463 +download,@{section=".text",section-sent="2048",section-size="6668",
31464 total-sent="2048",total-size="9880"@}
31465 +download,@{section=".text",section-sent="2560",section-size="6668",
31466 total-sent="2560",total-size="9880"@}
31467 +download,@{section=".text",section-sent="3072",section-size="6668",
31468 total-sent="3072",total-size="9880"@}
31469 +download,@{section=".text",section-sent="3584",section-size="6668",
31470 total-sent="3584",total-size="9880"@}
31471 +download,@{section=".text",section-sent="4096",section-size="6668",
31472 total-sent="4096",total-size="9880"@}
31473 +download,@{section=".text",section-sent="4608",section-size="6668",
31474 total-sent="4608",total-size="9880"@}
31475 +download,@{section=".text",section-sent="5120",section-size="6668",
31476 total-sent="5120",total-size="9880"@}
31477 +download,@{section=".text",section-sent="5632",section-size="6668",
31478 total-sent="5632",total-size="9880"@}
31479 +download,@{section=".text",section-sent="6144",section-size="6668",
31480 total-sent="6144",total-size="9880"@}
31481 +download,@{section=".text",section-sent="6656",section-size="6668",
31482 total-sent="6656",total-size="9880"@}
31483 +download,@{section=".init",section-size="28",total-size="9880"@}
31484 +download,@{section=".fini",section-size="28",total-size="9880"@}
31485 +download,@{section=".data",section-size="3156",total-size="9880"@}
31486 +download,@{section=".data",section-sent="512",section-size="3156",
31487 total-sent="7236",total-size="9880"@}
31488 +download,@{section=".data",section-sent="1024",section-size="3156",
31489 total-sent="7748",total-size="9880"@}
31490 +download,@{section=".data",section-sent="1536",section-size="3156",
31491 total-sent="8260",total-size="9880"@}
31492 +download,@{section=".data",section-sent="2048",section-size="3156",
31493 total-sent="8772",total-size="9880"@}
31494 +download,@{section=".data",section-sent="2560",section-size="3156",
31495 total-sent="9284",total-size="9880"@}
31496 +download,@{section=".data",section-sent="3072",section-size="3156",
31497 total-sent="9796",total-size="9880"@}
31498 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
31505 @subheading The @code{-target-exec-status} Command
31506 @findex -target-exec-status
31508 @subsubheading Synopsis
31511 -target-exec-status
31514 Provide information on the state of the target (whether it is running or
31515 not, for instance).
31517 @subsubheading @value{GDBN} Command
31519 There's no equivalent @value{GDBN} command.
31521 @subsubheading Example
31525 @subheading The @code{-target-list-available-targets} Command
31526 @findex -target-list-available-targets
31528 @subsubheading Synopsis
31531 -target-list-available-targets
31534 List the possible targets to connect to.
31536 @subsubheading @value{GDBN} Command
31538 The corresponding @value{GDBN} command is @samp{help target}.
31540 @subsubheading Example
31544 @subheading The @code{-target-list-current-targets} Command
31545 @findex -target-list-current-targets
31547 @subsubheading Synopsis
31550 -target-list-current-targets
31553 Describe the current target.
31555 @subsubheading @value{GDBN} Command
31557 The corresponding information is printed by @samp{info file} (among
31560 @subsubheading Example
31564 @subheading The @code{-target-list-parameters} Command
31565 @findex -target-list-parameters
31567 @subsubheading Synopsis
31570 -target-list-parameters
31576 @subsubheading @value{GDBN} Command
31580 @subsubheading Example
31584 @subheading The @code{-target-select} Command
31585 @findex -target-select
31587 @subsubheading Synopsis
31590 -target-select @var{type} @var{parameters @dots{}}
31593 Connect @value{GDBN} to the remote target. This command takes two args:
31597 The type of target, for instance @samp{remote}, etc.
31598 @item @var{parameters}
31599 Device names, host names and the like. @xref{Target Commands, ,
31600 Commands for Managing Targets}, for more details.
31603 The output is a connection notification, followed by the address at
31604 which the target program is, in the following form:
31607 ^connected,addr="@var{address}",func="@var{function name}",
31608 args=[@var{arg list}]
31611 @subsubheading @value{GDBN} Command
31613 The corresponding @value{GDBN} command is @samp{target}.
31615 @subsubheading Example
31619 -target-select remote /dev/ttya
31620 ^connected,addr="0xfe00a300",func="??",args=[]
31624 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31625 @node GDB/MI File Transfer Commands
31626 @section @sc{gdb/mi} File Transfer Commands
31629 @subheading The @code{-target-file-put} Command
31630 @findex -target-file-put
31632 @subsubheading Synopsis
31635 -target-file-put @var{hostfile} @var{targetfile}
31638 Copy file @var{hostfile} from the host system (the machine running
31639 @value{GDBN}) to @var{targetfile} on the target system.
31641 @subsubheading @value{GDBN} Command
31643 The corresponding @value{GDBN} command is @samp{remote put}.
31645 @subsubheading Example
31649 -target-file-put localfile remotefile
31655 @subheading The @code{-target-file-get} Command
31656 @findex -target-file-get
31658 @subsubheading Synopsis
31661 -target-file-get @var{targetfile} @var{hostfile}
31664 Copy file @var{targetfile} from the target system to @var{hostfile}
31665 on the host system.
31667 @subsubheading @value{GDBN} Command
31669 The corresponding @value{GDBN} command is @samp{remote get}.
31671 @subsubheading Example
31675 -target-file-get remotefile localfile
31681 @subheading The @code{-target-file-delete} Command
31682 @findex -target-file-delete
31684 @subsubheading Synopsis
31687 -target-file-delete @var{targetfile}
31690 Delete @var{targetfile} from the target system.
31692 @subsubheading @value{GDBN} Command
31694 The corresponding @value{GDBN} command is @samp{remote delete}.
31696 @subsubheading Example
31700 -target-file-delete remotefile
31706 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31707 @node GDB/MI Ada Exceptions Commands
31708 @section Ada Exceptions @sc{gdb/mi} Commands
31710 @subheading The @code{-info-ada-exceptions} Command
31711 @findex -info-ada-exceptions
31713 @subsubheading Synopsis
31716 -info-ada-exceptions [ @var{regexp}]
31719 List all Ada exceptions defined within the program being debugged.
31720 With a regular expression @var{regexp}, only those exceptions whose
31721 names match @var{regexp} are listed.
31723 @subsubheading @value{GDBN} Command
31725 The corresponding @value{GDBN} command is @samp{info exceptions}.
31727 @subsubheading Result
31729 The result is a table of Ada exceptions. The following columns are
31730 defined for each exception:
31734 The name of the exception.
31737 The address of the exception.
31741 @subsubheading Example
31744 -info-ada-exceptions aint
31745 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
31746 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
31747 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
31748 body=[@{name="constraint_error",address="0x0000000000613da0"@},
31749 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
31752 @subheading Catching Ada Exceptions
31754 The commands describing how to ask @value{GDBN} to stop when a program
31755 raises an exception are described at @ref{Ada Exception GDB/MI
31756 Catchpoint Commands}.
31759 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31760 @node GDB/MI Support Commands
31761 @section @sc{gdb/mi} Support Commands
31763 Since new commands and features get regularly added to @sc{gdb/mi},
31764 some commands are available to help front-ends query the debugger
31765 about support for these capabilities. Similarly, it is also possible
31766 to query @value{GDBN} about target support of certain features.
31768 @subheading The @code{-info-gdb-mi-command} Command
31769 @cindex @code{-info-gdb-mi-command}
31770 @findex -info-gdb-mi-command
31772 @subsubheading Synopsis
31775 -info-gdb-mi-command @var{cmd_name}
31778 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
31780 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
31781 is technically not part of the command name (@pxref{GDB/MI Input
31782 Syntax}), and thus should be omitted in @var{cmd_name}. However,
31783 for ease of use, this command also accepts the form with the leading
31786 @subsubheading @value{GDBN} Command
31788 There is no corresponding @value{GDBN} command.
31790 @subsubheading Result
31792 The result is a tuple. There is currently only one field:
31796 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
31797 @code{"false"} otherwise.
31801 @subsubheading Example
31803 Here is an example where the @sc{gdb/mi} command does not exist:
31806 -info-gdb-mi-command unsupported-command
31807 ^done,command=@{exists="false"@}
31811 And here is an example where the @sc{gdb/mi} command is known
31815 -info-gdb-mi-command symbol-list-lines
31816 ^done,command=@{exists="true"@}
31819 @subheading The @code{-list-features} Command
31820 @findex -list-features
31821 @cindex supported @sc{gdb/mi} features, list
31823 Returns a list of particular features of the MI protocol that
31824 this version of gdb implements. A feature can be a command,
31825 or a new field in an output of some command, or even an
31826 important bugfix. While a frontend can sometimes detect presence
31827 of a feature at runtime, it is easier to perform detection at debugger
31830 The command returns a list of strings, with each string naming an
31831 available feature. Each returned string is just a name, it does not
31832 have any internal structure. The list of possible feature names
31838 (gdb) -list-features
31839 ^done,result=["feature1","feature2"]
31842 The current list of features is:
31845 @item frozen-varobjs
31846 Indicates support for the @code{-var-set-frozen} command, as well
31847 as possible presense of the @code{frozen} field in the output
31848 of @code{-varobj-create}.
31849 @item pending-breakpoints
31850 Indicates support for the @option{-f} option to the @code{-break-insert}
31853 Indicates Python scripting support, Python-based
31854 pretty-printing commands, and possible presence of the
31855 @samp{display_hint} field in the output of @code{-var-list-children}
31857 Indicates support for the @code{-thread-info} command.
31858 @item data-read-memory-bytes
31859 Indicates support for the @code{-data-read-memory-bytes} and the
31860 @code{-data-write-memory-bytes} commands.
31861 @item breakpoint-notifications
31862 Indicates that changes to breakpoints and breakpoints created via the
31863 CLI will be announced via async records.
31864 @item ada-task-info
31865 Indicates support for the @code{-ada-task-info} command.
31866 @item language-option
31867 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
31868 option (@pxref{Context management}).
31869 @item info-gdb-mi-command
31870 Indicates support for the @code{-info-gdb-mi-command} command.
31871 @item undefined-command-error-code
31872 Indicates support for the "undefined-command" error code in error result
31873 records, produced when trying to execute an undefined @sc{gdb/mi} command
31874 (@pxref{GDB/MI Result Records}).
31875 @item exec-run-start-option
31876 Indicates that the @code{-exec-run} command supports the @option{--start}
31877 option (@pxref{GDB/MI Program Execution}).
31880 @subheading The @code{-list-target-features} Command
31881 @findex -list-target-features
31883 Returns a list of particular features that are supported by the
31884 target. Those features affect the permitted MI commands, but
31885 unlike the features reported by the @code{-list-features} command, the
31886 features depend on which target GDB is using at the moment. Whenever
31887 a target can change, due to commands such as @code{-target-select},
31888 @code{-target-attach} or @code{-exec-run}, the list of target features
31889 may change, and the frontend should obtain it again.
31893 (gdb) -list-target-features
31894 ^done,result=["async"]
31897 The current list of features is:
31901 Indicates that the target is capable of asynchronous command
31902 execution, which means that @value{GDBN} will accept further commands
31903 while the target is running.
31906 Indicates that the target is capable of reverse execution.
31907 @xref{Reverse Execution}, for more information.
31911 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31912 @node GDB/MI Miscellaneous Commands
31913 @section Miscellaneous @sc{gdb/mi} Commands
31915 @c @subheading -gdb-complete
31917 @subheading The @code{-gdb-exit} Command
31920 @subsubheading Synopsis
31926 Exit @value{GDBN} immediately.
31928 @subsubheading @value{GDBN} Command
31930 Approximately corresponds to @samp{quit}.
31932 @subsubheading Example
31942 @subheading The @code{-exec-abort} Command
31943 @findex -exec-abort
31945 @subsubheading Synopsis
31951 Kill the inferior running program.
31953 @subsubheading @value{GDBN} Command
31955 The corresponding @value{GDBN} command is @samp{kill}.
31957 @subsubheading Example
31962 @subheading The @code{-gdb-set} Command
31965 @subsubheading Synopsis
31971 Set an internal @value{GDBN} variable.
31972 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
31974 @subsubheading @value{GDBN} Command
31976 The corresponding @value{GDBN} command is @samp{set}.
31978 @subsubheading Example
31988 @subheading The @code{-gdb-show} Command
31991 @subsubheading Synopsis
31997 Show the current value of a @value{GDBN} variable.
31999 @subsubheading @value{GDBN} Command
32001 The corresponding @value{GDBN} command is @samp{show}.
32003 @subsubheading Example
32012 @c @subheading -gdb-source
32015 @subheading The @code{-gdb-version} Command
32016 @findex -gdb-version
32018 @subsubheading Synopsis
32024 Show version information for @value{GDBN}. Used mostly in testing.
32026 @subsubheading @value{GDBN} Command
32028 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
32029 default shows this information when you start an interactive session.
32031 @subsubheading Example
32033 @c This example modifies the actual output from GDB to avoid overfull
32039 ~Copyright 2000 Free Software Foundation, Inc.
32040 ~GDB is free software, covered by the GNU General Public License, and
32041 ~you are welcome to change it and/or distribute copies of it under
32042 ~ certain conditions.
32043 ~Type "show copying" to see the conditions.
32044 ~There is absolutely no warranty for GDB. Type "show warranty" for
32046 ~This GDB was configured as
32047 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
32052 @subheading The @code{-list-thread-groups} Command
32053 @findex -list-thread-groups
32055 @subheading Synopsis
32058 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
32061 Lists thread groups (@pxref{Thread groups}). When a single thread
32062 group is passed as the argument, lists the children of that group.
32063 When several thread group are passed, lists information about those
32064 thread groups. Without any parameters, lists information about all
32065 top-level thread groups.
32067 Normally, thread groups that are being debugged are reported.
32068 With the @samp{--available} option, @value{GDBN} reports thread groups
32069 available on the target.
32071 The output of this command may have either a @samp{threads} result or
32072 a @samp{groups} result. The @samp{thread} result has a list of tuples
32073 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
32074 Information}). The @samp{groups} result has a list of tuples as value,
32075 each tuple describing a thread group. If top-level groups are
32076 requested (that is, no parameter is passed), or when several groups
32077 are passed, the output always has a @samp{groups} result. The format
32078 of the @samp{group} result is described below.
32080 To reduce the number of roundtrips it's possible to list thread groups
32081 together with their children, by passing the @samp{--recurse} option
32082 and the recursion depth. Presently, only recursion depth of 1 is
32083 permitted. If this option is present, then every reported thread group
32084 will also include its children, either as @samp{group} or
32085 @samp{threads} field.
32087 In general, any combination of option and parameters is permitted, with
32088 the following caveats:
32092 When a single thread group is passed, the output will typically
32093 be the @samp{threads} result. Because threads may not contain
32094 anything, the @samp{recurse} option will be ignored.
32097 When the @samp{--available} option is passed, limited information may
32098 be available. In particular, the list of threads of a process might
32099 be inaccessible. Further, specifying specific thread groups might
32100 not give any performance advantage over listing all thread groups.
32101 The frontend should assume that @samp{-list-thread-groups --available}
32102 is always an expensive operation and cache the results.
32106 The @samp{groups} result is a list of tuples, where each tuple may
32107 have the following fields:
32111 Identifier of the thread group. This field is always present.
32112 The identifier is an opaque string; frontends should not try to
32113 convert it to an integer, even though it might look like one.
32116 The type of the thread group. At present, only @samp{process} is a
32120 The target-specific process identifier. This field is only present
32121 for thread groups of type @samp{process} and only if the process exists.
32124 The exit code of this group's last exited thread, formatted in octal.
32125 This field is only present for thread groups of type @samp{process} and
32126 only if the process is not running.
32129 The number of children this thread group has. This field may be
32130 absent for an available thread group.
32133 This field has a list of tuples as value, each tuple describing a
32134 thread. It may be present if the @samp{--recurse} option is
32135 specified, and it's actually possible to obtain the threads.
32138 This field is a list of integers, each identifying a core that one
32139 thread of the group is running on. This field may be absent if
32140 such information is not available.
32143 The name of the executable file that corresponds to this thread group.
32144 The field is only present for thread groups of type @samp{process},
32145 and only if there is a corresponding executable file.
32149 @subheading Example
32153 -list-thread-groups
32154 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
32155 -list-thread-groups 17
32156 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
32157 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
32158 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
32159 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
32160 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
32161 -list-thread-groups --available
32162 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
32163 -list-thread-groups --available --recurse 1
32164 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
32165 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
32166 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
32167 -list-thread-groups --available --recurse 1 17 18
32168 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
32169 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
32170 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
32173 @subheading The @code{-info-os} Command
32176 @subsubheading Synopsis
32179 -info-os [ @var{type} ]
32182 If no argument is supplied, the command returns a table of available
32183 operating-system-specific information types. If one of these types is
32184 supplied as an argument @var{type}, then the command returns a table
32185 of data of that type.
32187 The types of information available depend on the target operating
32190 @subsubheading @value{GDBN} Command
32192 The corresponding @value{GDBN} command is @samp{info os}.
32194 @subsubheading Example
32196 When run on a @sc{gnu}/Linux system, the output will look something
32202 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
32203 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
32204 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
32205 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
32206 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
32208 item=@{col0="files",col1="Listing of all file descriptors",
32209 col2="File descriptors"@},
32210 item=@{col0="modules",col1="Listing of all loaded kernel modules",
32211 col2="Kernel modules"@},
32212 item=@{col0="msg",col1="Listing of all message queues",
32213 col2="Message queues"@},
32214 item=@{col0="processes",col1="Listing of all processes",
32215 col2="Processes"@},
32216 item=@{col0="procgroups",col1="Listing of all process groups",
32217 col2="Process groups"@},
32218 item=@{col0="semaphores",col1="Listing of all semaphores",
32219 col2="Semaphores"@},
32220 item=@{col0="shm",col1="Listing of all shared-memory regions",
32221 col2="Shared-memory regions"@},
32222 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
32224 item=@{col0="threads",col1="Listing of all threads",
32228 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
32229 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
32230 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
32231 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
32232 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
32233 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
32234 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
32235 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
32237 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
32238 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
32242 (Note that the MI output here includes a @code{"Title"} column that
32243 does not appear in command-line @code{info os}; this column is useful
32244 for MI clients that want to enumerate the types of data, such as in a
32245 popup menu, but is needless clutter on the command line, and
32246 @code{info os} omits it.)
32248 @subheading The @code{-add-inferior} Command
32249 @findex -add-inferior
32251 @subheading Synopsis
32257 Creates a new inferior (@pxref{Inferiors and Programs}). The created
32258 inferior is not associated with any executable. Such association may
32259 be established with the @samp{-file-exec-and-symbols} command
32260 (@pxref{GDB/MI File Commands}). The command response has a single
32261 field, @samp{inferior}, whose value is the identifier of the
32262 thread group corresponding to the new inferior.
32264 @subheading Example
32269 ^done,inferior="i3"
32272 @subheading The @code{-interpreter-exec} Command
32273 @findex -interpreter-exec
32275 @subheading Synopsis
32278 -interpreter-exec @var{interpreter} @var{command}
32280 @anchor{-interpreter-exec}
32282 Execute the specified @var{command} in the given @var{interpreter}.
32284 @subheading @value{GDBN} Command
32286 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
32288 @subheading Example
32292 -interpreter-exec console "break main"
32293 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
32294 &"During symbol reading, bad structure-type format.\n"
32295 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
32300 @subheading The @code{-inferior-tty-set} Command
32301 @findex -inferior-tty-set
32303 @subheading Synopsis
32306 -inferior-tty-set /dev/pts/1
32309 Set terminal for future runs of the program being debugged.
32311 @subheading @value{GDBN} Command
32313 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
32315 @subheading Example
32319 -inferior-tty-set /dev/pts/1
32324 @subheading The @code{-inferior-tty-show} Command
32325 @findex -inferior-tty-show
32327 @subheading Synopsis
32333 Show terminal for future runs of program being debugged.
32335 @subheading @value{GDBN} Command
32337 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
32339 @subheading Example
32343 -inferior-tty-set /dev/pts/1
32347 ^done,inferior_tty_terminal="/dev/pts/1"
32351 @subheading The @code{-enable-timings} Command
32352 @findex -enable-timings
32354 @subheading Synopsis
32357 -enable-timings [yes | no]
32360 Toggle the printing of the wallclock, user and system times for an MI
32361 command as a field in its output. This command is to help frontend
32362 developers optimize the performance of their code. No argument is
32363 equivalent to @samp{yes}.
32365 @subheading @value{GDBN} Command
32369 @subheading Example
32377 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
32378 addr="0x080484ed",func="main",file="myprog.c",
32379 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
32381 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
32389 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
32390 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
32391 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
32392 fullname="/home/nickrob/myprog.c",line="73"@}
32397 @chapter @value{GDBN} Annotations
32399 This chapter describes annotations in @value{GDBN}. Annotations were
32400 designed to interface @value{GDBN} to graphical user interfaces or other
32401 similar programs which want to interact with @value{GDBN} at a
32402 relatively high level.
32404 The annotation mechanism has largely been superseded by @sc{gdb/mi}
32408 This is Edition @value{EDITION}, @value{DATE}.
32412 * Annotations Overview:: What annotations are; the general syntax.
32413 * Server Prefix:: Issuing a command without affecting user state.
32414 * Prompting:: Annotations marking @value{GDBN}'s need for input.
32415 * Errors:: Annotations for error messages.
32416 * Invalidation:: Some annotations describe things now invalid.
32417 * Annotations for Running::
32418 Whether the program is running, how it stopped, etc.
32419 * Source Annotations:: Annotations describing source code.
32422 @node Annotations Overview
32423 @section What is an Annotation?
32424 @cindex annotations
32426 Annotations start with a newline character, two @samp{control-z}
32427 characters, and the name of the annotation. If there is no additional
32428 information associated with this annotation, the name of the annotation
32429 is followed immediately by a newline. If there is additional
32430 information, the name of the annotation is followed by a space, the
32431 additional information, and a newline. The additional information
32432 cannot contain newline characters.
32434 Any output not beginning with a newline and two @samp{control-z}
32435 characters denotes literal output from @value{GDBN}. Currently there is
32436 no need for @value{GDBN} to output a newline followed by two
32437 @samp{control-z} characters, but if there was such a need, the
32438 annotations could be extended with an @samp{escape} annotation which
32439 means those three characters as output.
32441 The annotation @var{level}, which is specified using the
32442 @option{--annotate} command line option (@pxref{Mode Options}), controls
32443 how much information @value{GDBN} prints together with its prompt,
32444 values of expressions, source lines, and other types of output. Level 0
32445 is for no annotations, level 1 is for use when @value{GDBN} is run as a
32446 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
32447 for programs that control @value{GDBN}, and level 2 annotations have
32448 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
32449 Interface, annotate, GDB's Obsolete Annotations}).
32452 @kindex set annotate
32453 @item set annotate @var{level}
32454 The @value{GDBN} command @code{set annotate} sets the level of
32455 annotations to the specified @var{level}.
32457 @item show annotate
32458 @kindex show annotate
32459 Show the current annotation level.
32462 This chapter describes level 3 annotations.
32464 A simple example of starting up @value{GDBN} with annotations is:
32467 $ @kbd{gdb --annotate=3}
32469 Copyright 2003 Free Software Foundation, Inc.
32470 GDB is free software, covered by the GNU General Public License,
32471 and you are welcome to change it and/or distribute copies of it
32472 under certain conditions.
32473 Type "show copying" to see the conditions.
32474 There is absolutely no warranty for GDB. Type "show warranty"
32476 This GDB was configured as "i386-pc-linux-gnu"
32487 Here @samp{quit} is input to @value{GDBN}; the rest is output from
32488 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
32489 denotes a @samp{control-z} character) are annotations; the rest is
32490 output from @value{GDBN}.
32492 @node Server Prefix
32493 @section The Server Prefix
32494 @cindex server prefix
32496 If you prefix a command with @samp{server } then it will not affect
32497 the command history, nor will it affect @value{GDBN}'s notion of which
32498 command to repeat if @key{RET} is pressed on a line by itself. This
32499 means that commands can be run behind a user's back by a front-end in
32500 a transparent manner.
32502 The @code{server } prefix does not affect the recording of values into
32503 the value history; to print a value without recording it into the
32504 value history, use the @code{output} command instead of the
32505 @code{print} command.
32507 Using this prefix also disables confirmation requests
32508 (@pxref{confirmation requests}).
32511 @section Annotation for @value{GDBN} Input
32513 @cindex annotations for prompts
32514 When @value{GDBN} prompts for input, it annotates this fact so it is possible
32515 to know when to send output, when the output from a given command is
32518 Different kinds of input each have a different @dfn{input type}. Each
32519 input type has three annotations: a @code{pre-} annotation, which
32520 denotes the beginning of any prompt which is being output, a plain
32521 annotation, which denotes the end of the prompt, and then a @code{post-}
32522 annotation which denotes the end of any echo which may (or may not) be
32523 associated with the input. For example, the @code{prompt} input type
32524 features the following annotations:
32532 The input types are
32535 @findex pre-prompt annotation
32536 @findex prompt annotation
32537 @findex post-prompt annotation
32539 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
32541 @findex pre-commands annotation
32542 @findex commands annotation
32543 @findex post-commands annotation
32545 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
32546 command. The annotations are repeated for each command which is input.
32548 @findex pre-overload-choice annotation
32549 @findex overload-choice annotation
32550 @findex post-overload-choice annotation
32551 @item overload-choice
32552 When @value{GDBN} wants the user to select between various overloaded functions.
32554 @findex pre-query annotation
32555 @findex query annotation
32556 @findex post-query annotation
32558 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
32560 @findex pre-prompt-for-continue annotation
32561 @findex prompt-for-continue annotation
32562 @findex post-prompt-for-continue annotation
32563 @item prompt-for-continue
32564 When @value{GDBN} is asking the user to press return to continue. Note: Don't
32565 expect this to work well; instead use @code{set height 0} to disable
32566 prompting. This is because the counting of lines is buggy in the
32567 presence of annotations.
32572 @cindex annotations for errors, warnings and interrupts
32574 @findex quit annotation
32579 This annotation occurs right before @value{GDBN} responds to an interrupt.
32581 @findex error annotation
32586 This annotation occurs right before @value{GDBN} responds to an error.
32588 Quit and error annotations indicate that any annotations which @value{GDBN} was
32589 in the middle of may end abruptly. For example, if a
32590 @code{value-history-begin} annotation is followed by a @code{error}, one
32591 cannot expect to receive the matching @code{value-history-end}. One
32592 cannot expect not to receive it either, however; an error annotation
32593 does not necessarily mean that @value{GDBN} is immediately returning all the way
32596 @findex error-begin annotation
32597 A quit or error annotation may be preceded by
32603 Any output between that and the quit or error annotation is the error
32606 Warning messages are not yet annotated.
32607 @c If we want to change that, need to fix warning(), type_error(),
32608 @c range_error(), and possibly other places.
32611 @section Invalidation Notices
32613 @cindex annotations for invalidation messages
32614 The following annotations say that certain pieces of state may have
32618 @findex frames-invalid annotation
32619 @item ^Z^Zframes-invalid
32621 The frames (for example, output from the @code{backtrace} command) may
32624 @findex breakpoints-invalid annotation
32625 @item ^Z^Zbreakpoints-invalid
32627 The breakpoints may have changed. For example, the user just added or
32628 deleted a breakpoint.
32631 @node Annotations for Running
32632 @section Running the Program
32633 @cindex annotations for running programs
32635 @findex starting annotation
32636 @findex stopping annotation
32637 When the program starts executing due to a @value{GDBN} command such as
32638 @code{step} or @code{continue},
32644 is output. When the program stops,
32650 is output. Before the @code{stopped} annotation, a variety of
32651 annotations describe how the program stopped.
32654 @findex exited annotation
32655 @item ^Z^Zexited @var{exit-status}
32656 The program exited, and @var{exit-status} is the exit status (zero for
32657 successful exit, otherwise nonzero).
32659 @findex signalled annotation
32660 @findex signal-name annotation
32661 @findex signal-name-end annotation
32662 @findex signal-string annotation
32663 @findex signal-string-end annotation
32664 @item ^Z^Zsignalled
32665 The program exited with a signal. After the @code{^Z^Zsignalled}, the
32666 annotation continues:
32672 ^Z^Zsignal-name-end
32676 ^Z^Zsignal-string-end
32681 where @var{name} is the name of the signal, such as @code{SIGILL} or
32682 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
32683 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
32684 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
32685 user's benefit and have no particular format.
32687 @findex signal annotation
32689 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
32690 just saying that the program received the signal, not that it was
32691 terminated with it.
32693 @findex breakpoint annotation
32694 @item ^Z^Zbreakpoint @var{number}
32695 The program hit breakpoint number @var{number}.
32697 @findex watchpoint annotation
32698 @item ^Z^Zwatchpoint @var{number}
32699 The program hit watchpoint number @var{number}.
32702 @node Source Annotations
32703 @section Displaying Source
32704 @cindex annotations for source display
32706 @findex source annotation
32707 The following annotation is used instead of displaying source code:
32710 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
32713 where @var{filename} is an absolute file name indicating which source
32714 file, @var{line} is the line number within that file (where 1 is the
32715 first line in the file), @var{character} is the character position
32716 within the file (where 0 is the first character in the file) (for most
32717 debug formats this will necessarily point to the beginning of a line),
32718 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
32719 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
32720 @var{addr} is the address in the target program associated with the
32721 source which is being displayed. The @var{addr} is in the form @samp{0x}
32722 followed by one or more lowercase hex digits (note that this does not
32723 depend on the language).
32725 @node JIT Interface
32726 @chapter JIT Compilation Interface
32727 @cindex just-in-time compilation
32728 @cindex JIT compilation interface
32730 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
32731 interface. A JIT compiler is a program or library that generates native
32732 executable code at runtime and executes it, usually in order to achieve good
32733 performance while maintaining platform independence.
32735 Programs that use JIT compilation are normally difficult to debug because
32736 portions of their code are generated at runtime, instead of being loaded from
32737 object files, which is where @value{GDBN} normally finds the program's symbols
32738 and debug information. In order to debug programs that use JIT compilation,
32739 @value{GDBN} has an interface that allows the program to register in-memory
32740 symbol files with @value{GDBN} at runtime.
32742 If you are using @value{GDBN} to debug a program that uses this interface, then
32743 it should work transparently so long as you have not stripped the binary. If
32744 you are developing a JIT compiler, then the interface is documented in the rest
32745 of this chapter. At this time, the only known client of this interface is the
32748 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
32749 JIT compiler communicates with @value{GDBN} by writing data into a global
32750 variable and calling a fuction at a well-known symbol. When @value{GDBN}
32751 attaches, it reads a linked list of symbol files from the global variable to
32752 find existing code, and puts a breakpoint in the function so that it can find
32753 out about additional code.
32756 * Declarations:: Relevant C struct declarations
32757 * Registering Code:: Steps to register code
32758 * Unregistering Code:: Steps to unregister code
32759 * Custom Debug Info:: Emit debug information in a custom format
32763 @section JIT Declarations
32765 These are the relevant struct declarations that a C program should include to
32766 implement the interface:
32776 struct jit_code_entry
32778 struct jit_code_entry *next_entry;
32779 struct jit_code_entry *prev_entry;
32780 const char *symfile_addr;
32781 uint64_t symfile_size;
32784 struct jit_descriptor
32787 /* This type should be jit_actions_t, but we use uint32_t
32788 to be explicit about the bitwidth. */
32789 uint32_t action_flag;
32790 struct jit_code_entry *relevant_entry;
32791 struct jit_code_entry *first_entry;
32794 /* GDB puts a breakpoint in this function. */
32795 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
32797 /* Make sure to specify the version statically, because the
32798 debugger may check the version before we can set it. */
32799 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
32802 If the JIT is multi-threaded, then it is important that the JIT synchronize any
32803 modifications to this global data properly, which can easily be done by putting
32804 a global mutex around modifications to these structures.
32806 @node Registering Code
32807 @section Registering Code
32809 To register code with @value{GDBN}, the JIT should follow this protocol:
32813 Generate an object file in memory with symbols and other desired debug
32814 information. The file must include the virtual addresses of the sections.
32817 Create a code entry for the file, which gives the start and size of the symbol
32821 Add it to the linked list in the JIT descriptor.
32824 Point the relevant_entry field of the descriptor at the entry.
32827 Set @code{action_flag} to @code{JIT_REGISTER} and call
32828 @code{__jit_debug_register_code}.
32831 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
32832 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
32833 new code. However, the linked list must still be maintained in order to allow
32834 @value{GDBN} to attach to a running process and still find the symbol files.
32836 @node Unregistering Code
32837 @section Unregistering Code
32839 If code is freed, then the JIT should use the following protocol:
32843 Remove the code entry corresponding to the code from the linked list.
32846 Point the @code{relevant_entry} field of the descriptor at the code entry.
32849 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
32850 @code{__jit_debug_register_code}.
32853 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
32854 and the JIT will leak the memory used for the associated symbol files.
32856 @node Custom Debug Info
32857 @section Custom Debug Info
32858 @cindex custom JIT debug info
32859 @cindex JIT debug info reader
32861 Generating debug information in platform-native file formats (like ELF
32862 or COFF) may be an overkill for JIT compilers; especially if all the
32863 debug info is used for is displaying a meaningful backtrace. The
32864 issue can be resolved by having the JIT writers decide on a debug info
32865 format and also provide a reader that parses the debug info generated
32866 by the JIT compiler. This section gives a brief overview on writing
32867 such a parser. More specific details can be found in the source file
32868 @file{gdb/jit-reader.in}, which is also installed as a header at
32869 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
32871 The reader is implemented as a shared object (so this functionality is
32872 not available on platforms which don't allow loading shared objects at
32873 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
32874 @code{jit-reader-unload} are provided, to be used to load and unload
32875 the readers from a preconfigured directory. Once loaded, the shared
32876 object is used the parse the debug information emitted by the JIT
32880 * Using JIT Debug Info Readers:: How to use supplied readers correctly
32881 * Writing JIT Debug Info Readers:: Creating a debug-info reader
32884 @node Using JIT Debug Info Readers
32885 @subsection Using JIT Debug Info Readers
32886 @kindex jit-reader-load
32887 @kindex jit-reader-unload
32889 Readers can be loaded and unloaded using the @code{jit-reader-load}
32890 and @code{jit-reader-unload} commands.
32893 @item jit-reader-load @var{reader}
32894 Load the JIT reader named @var{reader}, which is a shared
32895 object specified as either an absolute or a relative file name. In
32896 the latter case, @value{GDBN} will try to load the reader from a
32897 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
32898 system (here @var{libdir} is the system library directory, often
32899 @file{/usr/local/lib}).
32901 Only one reader can be active at a time; trying to load a second
32902 reader when one is already loaded will result in @value{GDBN}
32903 reporting an error. A new JIT reader can be loaded by first unloading
32904 the current one using @code{jit-reader-unload} and then invoking
32905 @code{jit-reader-load}.
32907 @item jit-reader-unload
32908 Unload the currently loaded JIT reader.
32912 @node Writing JIT Debug Info Readers
32913 @subsection Writing JIT Debug Info Readers
32914 @cindex writing JIT debug info readers
32916 As mentioned, a reader is essentially a shared object conforming to a
32917 certain ABI. This ABI is described in @file{jit-reader.h}.
32919 @file{jit-reader.h} defines the structures, macros and functions
32920 required to write a reader. It is installed (along with
32921 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
32922 the system include directory.
32924 Readers need to be released under a GPL compatible license. A reader
32925 can be declared as released under such a license by placing the macro
32926 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
32928 The entry point for readers is the symbol @code{gdb_init_reader},
32929 which is expected to be a function with the prototype
32931 @findex gdb_init_reader
32933 extern struct gdb_reader_funcs *gdb_init_reader (void);
32936 @cindex @code{struct gdb_reader_funcs}
32938 @code{struct gdb_reader_funcs} contains a set of pointers to callback
32939 functions. These functions are executed to read the debug info
32940 generated by the JIT compiler (@code{read}), to unwind stack frames
32941 (@code{unwind}) and to create canonical frame IDs
32942 (@code{get_Frame_id}). It also has a callback that is called when the
32943 reader is being unloaded (@code{destroy}). The struct looks like this
32946 struct gdb_reader_funcs
32948 /* Must be set to GDB_READER_INTERFACE_VERSION. */
32949 int reader_version;
32951 /* For use by the reader. */
32954 gdb_read_debug_info *read;
32955 gdb_unwind_frame *unwind;
32956 gdb_get_frame_id *get_frame_id;
32957 gdb_destroy_reader *destroy;
32961 @cindex @code{struct gdb_symbol_callbacks}
32962 @cindex @code{struct gdb_unwind_callbacks}
32964 The callbacks are provided with another set of callbacks by
32965 @value{GDBN} to do their job. For @code{read}, these callbacks are
32966 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
32967 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
32968 @code{struct gdb_symbol_callbacks} has callbacks to create new object
32969 files and new symbol tables inside those object files. @code{struct
32970 gdb_unwind_callbacks} has callbacks to read registers off the current
32971 frame and to write out the values of the registers in the previous
32972 frame. Both have a callback (@code{target_read}) to read bytes off the
32973 target's address space.
32975 @node In-Process Agent
32976 @chapter In-Process Agent
32977 @cindex debugging agent
32978 The traditional debugging model is conceptually low-speed, but works fine,
32979 because most bugs can be reproduced in debugging-mode execution. However,
32980 as multi-core or many-core processors are becoming mainstream, and
32981 multi-threaded programs become more and more popular, there should be more
32982 and more bugs that only manifest themselves at normal-mode execution, for
32983 example, thread races, because debugger's interference with the program's
32984 timing may conceal the bugs. On the other hand, in some applications,
32985 it is not feasible for the debugger to interrupt the program's execution
32986 long enough for the developer to learn anything helpful about its behavior.
32987 If the program's correctness depends on its real-time behavior, delays
32988 introduced by a debugger might cause the program to fail, even when the
32989 code itself is correct. It is useful to be able to observe the program's
32990 behavior without interrupting it.
32992 Therefore, traditional debugging model is too intrusive to reproduce
32993 some bugs. In order to reduce the interference with the program, we can
32994 reduce the number of operations performed by debugger. The
32995 @dfn{In-Process Agent}, a shared library, is running within the same
32996 process with inferior, and is able to perform some debugging operations
32997 itself. As a result, debugger is only involved when necessary, and
32998 performance of debugging can be improved accordingly. Note that
32999 interference with program can be reduced but can't be removed completely,
33000 because the in-process agent will still stop or slow down the program.
33002 The in-process agent can interpret and execute Agent Expressions
33003 (@pxref{Agent Expressions}) during performing debugging operations. The
33004 agent expressions can be used for different purposes, such as collecting
33005 data in tracepoints, and condition evaluation in breakpoints.
33007 @anchor{Control Agent}
33008 You can control whether the in-process agent is used as an aid for
33009 debugging with the following commands:
33012 @kindex set agent on
33014 Causes the in-process agent to perform some operations on behalf of the
33015 debugger. Just which operations requested by the user will be done
33016 by the in-process agent depends on the its capabilities. For example,
33017 if you request to evaluate breakpoint conditions in the in-process agent,
33018 and the in-process agent has such capability as well, then breakpoint
33019 conditions will be evaluated in the in-process agent.
33021 @kindex set agent off
33022 @item set agent off
33023 Disables execution of debugging operations by the in-process agent. All
33024 of the operations will be performed by @value{GDBN}.
33028 Display the current setting of execution of debugging operations by
33029 the in-process agent.
33033 * In-Process Agent Protocol::
33036 @node In-Process Agent Protocol
33037 @section In-Process Agent Protocol
33038 @cindex in-process agent protocol
33040 The in-process agent is able to communicate with both @value{GDBN} and
33041 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
33042 used for communications between @value{GDBN} or GDBserver and the IPA.
33043 In general, @value{GDBN} or GDBserver sends commands
33044 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
33045 in-process agent replies back with the return result of the command, or
33046 some other information. The data sent to in-process agent is composed
33047 of primitive data types, such as 4-byte or 8-byte type, and composite
33048 types, which are called objects (@pxref{IPA Protocol Objects}).
33051 * IPA Protocol Objects::
33052 * IPA Protocol Commands::
33055 @node IPA Protocol Objects
33056 @subsection IPA Protocol Objects
33057 @cindex ipa protocol objects
33059 The commands sent to and results received from agent may contain some
33060 complex data types called @dfn{objects}.
33062 The in-process agent is running on the same machine with @value{GDBN}
33063 or GDBserver, so it doesn't have to handle as much differences between
33064 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
33065 However, there are still some differences of two ends in two processes:
33069 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
33070 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
33072 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
33073 GDBserver is compiled with one, and in-process agent is compiled with
33077 Here are the IPA Protocol Objects:
33081 agent expression object. It represents an agent expression
33082 (@pxref{Agent Expressions}).
33083 @anchor{agent expression object}
33085 tracepoint action object. It represents a tracepoint action
33086 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
33087 memory, static trace data and to evaluate expression.
33088 @anchor{tracepoint action object}
33090 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
33091 @anchor{tracepoint object}
33095 The following table describes important attributes of each IPA protocol
33098 @multitable @columnfractions .30 .20 .50
33099 @headitem Name @tab Size @tab Description
33100 @item @emph{agent expression object} @tab @tab
33101 @item length @tab 4 @tab length of bytes code
33102 @item byte code @tab @var{length} @tab contents of byte code
33103 @item @emph{tracepoint action for collecting memory} @tab @tab
33104 @item 'M' @tab 1 @tab type of tracepoint action
33105 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
33106 address of the lowest byte to collect, otherwise @var{addr} is the offset
33107 of @var{basereg} for memory collecting.
33108 @item len @tab 8 @tab length of memory for collecting
33109 @item basereg @tab 4 @tab the register number containing the starting
33110 memory address for collecting.
33111 @item @emph{tracepoint action for collecting registers} @tab @tab
33112 @item 'R' @tab 1 @tab type of tracepoint action
33113 @item @emph{tracepoint action for collecting static trace data} @tab @tab
33114 @item 'L' @tab 1 @tab type of tracepoint action
33115 @item @emph{tracepoint action for expression evaluation} @tab @tab
33116 @item 'X' @tab 1 @tab type of tracepoint action
33117 @item agent expression @tab length of @tab @ref{agent expression object}
33118 @item @emph{tracepoint object} @tab @tab
33119 @item number @tab 4 @tab number of tracepoint
33120 @item address @tab 8 @tab address of tracepoint inserted on
33121 @item type @tab 4 @tab type of tracepoint
33122 @item enabled @tab 1 @tab enable or disable of tracepoint
33123 @item step_count @tab 8 @tab step
33124 @item pass_count @tab 8 @tab pass
33125 @item numactions @tab 4 @tab number of tracepoint actions
33126 @item hit count @tab 8 @tab hit count
33127 @item trace frame usage @tab 8 @tab trace frame usage
33128 @item compiled_cond @tab 8 @tab compiled condition
33129 @item orig_size @tab 8 @tab orig size
33130 @item condition @tab 4 if condition is NULL otherwise length of
33131 @ref{agent expression object}
33132 @tab zero if condition is NULL, otherwise is
33133 @ref{agent expression object}
33134 @item actions @tab variable
33135 @tab numactions number of @ref{tracepoint action object}
33138 @node IPA Protocol Commands
33139 @subsection IPA Protocol Commands
33140 @cindex ipa protocol commands
33142 The spaces in each command are delimiters to ease reading this commands
33143 specification. They don't exist in real commands.
33147 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
33148 Installs a new fast tracepoint described by @var{tracepoint_object}
33149 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
33150 head of @dfn{jumppad}, which is used to jump to data collection routine
33155 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
33156 @var{target_address} is address of tracepoint in the inferior.
33157 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
33158 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
33159 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
33160 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
33167 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
33168 is about to kill inferiors.
33176 @item probe_marker_at:@var{address}
33177 Asks in-process agent to probe the marker at @var{address}.
33184 @item unprobe_marker_at:@var{address}
33185 Asks in-process agent to unprobe the marker at @var{address}.
33189 @chapter Reporting Bugs in @value{GDBN}
33190 @cindex bugs in @value{GDBN}
33191 @cindex reporting bugs in @value{GDBN}
33193 Your bug reports play an essential role in making @value{GDBN} reliable.
33195 Reporting a bug may help you by bringing a solution to your problem, or it
33196 may not. But in any case the principal function of a bug report is to help
33197 the entire community by making the next version of @value{GDBN} work better. Bug
33198 reports are your contribution to the maintenance of @value{GDBN}.
33200 In order for a bug report to serve its purpose, you must include the
33201 information that enables us to fix the bug.
33204 * Bug Criteria:: Have you found a bug?
33205 * Bug Reporting:: How to report bugs
33209 @section Have You Found a Bug?
33210 @cindex bug criteria
33212 If you are not sure whether you have found a bug, here are some guidelines:
33215 @cindex fatal signal
33216 @cindex debugger crash
33217 @cindex crash of debugger
33219 If the debugger gets a fatal signal, for any input whatever, that is a
33220 @value{GDBN} bug. Reliable debuggers never crash.
33222 @cindex error on valid input
33224 If @value{GDBN} produces an error message for valid input, that is a
33225 bug. (Note that if you're cross debugging, the problem may also be
33226 somewhere in the connection to the target.)
33228 @cindex invalid input
33230 If @value{GDBN} does not produce an error message for invalid input,
33231 that is a bug. However, you should note that your idea of
33232 ``invalid input'' might be our idea of ``an extension'' or ``support
33233 for traditional practice''.
33236 If you are an experienced user of debugging tools, your suggestions
33237 for improvement of @value{GDBN} are welcome in any case.
33240 @node Bug Reporting
33241 @section How to Report Bugs
33242 @cindex bug reports
33243 @cindex @value{GDBN} bugs, reporting
33245 A number of companies and individuals offer support for @sc{gnu} products.
33246 If you obtained @value{GDBN} from a support organization, we recommend you
33247 contact that organization first.
33249 You can find contact information for many support companies and
33250 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
33252 @c should add a web page ref...
33255 @ifset BUGURL_DEFAULT
33256 In any event, we also recommend that you submit bug reports for
33257 @value{GDBN}. The preferred method is to submit them directly using
33258 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
33259 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
33262 @strong{Do not send bug reports to @samp{info-gdb}, or to
33263 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
33264 not want to receive bug reports. Those that do have arranged to receive
33267 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
33268 serves as a repeater. The mailing list and the newsgroup carry exactly
33269 the same messages. Often people think of posting bug reports to the
33270 newsgroup instead of mailing them. This appears to work, but it has one
33271 problem which can be crucial: a newsgroup posting often lacks a mail
33272 path back to the sender. Thus, if we need to ask for more information,
33273 we may be unable to reach you. For this reason, it is better to send
33274 bug reports to the mailing list.
33276 @ifclear BUGURL_DEFAULT
33277 In any event, we also recommend that you submit bug reports for
33278 @value{GDBN} to @value{BUGURL}.
33282 The fundamental principle of reporting bugs usefully is this:
33283 @strong{report all the facts}. If you are not sure whether to state a
33284 fact or leave it out, state it!
33286 Often people omit facts because they think they know what causes the
33287 problem and assume that some details do not matter. Thus, you might
33288 assume that the name of the variable you use in an example does not matter.
33289 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
33290 stray memory reference which happens to fetch from the location where that
33291 name is stored in memory; perhaps, if the name were different, the contents
33292 of that location would fool the debugger into doing the right thing despite
33293 the bug. Play it safe and give a specific, complete example. That is the
33294 easiest thing for you to do, and the most helpful.
33296 Keep in mind that the purpose of a bug report is to enable us to fix the
33297 bug. It may be that the bug has been reported previously, but neither
33298 you nor we can know that unless your bug report is complete and
33301 Sometimes people give a few sketchy facts and ask, ``Does this ring a
33302 bell?'' Those bug reports are useless, and we urge everyone to
33303 @emph{refuse to respond to them} except to chide the sender to report
33306 To enable us to fix the bug, you should include all these things:
33310 The version of @value{GDBN}. @value{GDBN} announces it if you start
33311 with no arguments; you can also print it at any time using @code{show
33314 Without this, we will not know whether there is any point in looking for
33315 the bug in the current version of @value{GDBN}.
33318 The type of machine you are using, and the operating system name and
33322 The details of the @value{GDBN} build-time configuration.
33323 @value{GDBN} shows these details if you invoke it with the
33324 @option{--configuration} command-line option, or if you type
33325 @code{show configuration} at @value{GDBN}'s prompt.
33328 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
33329 ``@value{GCC}--2.8.1''.
33332 What compiler (and its version) was used to compile the program you are
33333 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
33334 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
33335 to get this information; for other compilers, see the documentation for
33339 The command arguments you gave the compiler to compile your example and
33340 observe the bug. For example, did you use @samp{-O}? To guarantee
33341 you will not omit something important, list them all. A copy of the
33342 Makefile (or the output from make) is sufficient.
33344 If we were to try to guess the arguments, we would probably guess wrong
33345 and then we might not encounter the bug.
33348 A complete input script, and all necessary source files, that will
33352 A description of what behavior you observe that you believe is
33353 incorrect. For example, ``It gets a fatal signal.''
33355 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
33356 will certainly notice it. But if the bug is incorrect output, we might
33357 not notice unless it is glaringly wrong. You might as well not give us
33358 a chance to make a mistake.
33360 Even if the problem you experience is a fatal signal, you should still
33361 say so explicitly. Suppose something strange is going on, such as, your
33362 copy of @value{GDBN} is out of synch, or you have encountered a bug in
33363 the C library on your system. (This has happened!) Your copy might
33364 crash and ours would not. If you told us to expect a crash, then when
33365 ours fails to crash, we would know that the bug was not happening for
33366 us. If you had not told us to expect a crash, then we would not be able
33367 to draw any conclusion from our observations.
33370 @cindex recording a session script
33371 To collect all this information, you can use a session recording program
33372 such as @command{script}, which is available on many Unix systems.
33373 Just run your @value{GDBN} session inside @command{script} and then
33374 include the @file{typescript} file with your bug report.
33376 Another way to record a @value{GDBN} session is to run @value{GDBN}
33377 inside Emacs and then save the entire buffer to a file.
33380 If you wish to suggest changes to the @value{GDBN} source, send us context
33381 diffs. If you even discuss something in the @value{GDBN} source, refer to
33382 it by context, not by line number.
33384 The line numbers in our development sources will not match those in your
33385 sources. Your line numbers would convey no useful information to us.
33389 Here are some things that are not necessary:
33393 A description of the envelope of the bug.
33395 Often people who encounter a bug spend a lot of time investigating
33396 which changes to the input file will make the bug go away and which
33397 changes will not affect it.
33399 This is often time consuming and not very useful, because the way we
33400 will find the bug is by running a single example under the debugger
33401 with breakpoints, not by pure deduction from a series of examples.
33402 We recommend that you save your time for something else.
33404 Of course, if you can find a simpler example to report @emph{instead}
33405 of the original one, that is a convenience for us. Errors in the
33406 output will be easier to spot, running under the debugger will take
33407 less time, and so on.
33409 However, simplification is not vital; if you do not want to do this,
33410 report the bug anyway and send us the entire test case you used.
33413 A patch for the bug.
33415 A patch for the bug does help us if it is a good one. But do not omit
33416 the necessary information, such as the test case, on the assumption that
33417 a patch is all we need. We might see problems with your patch and decide
33418 to fix the problem another way, or we might not understand it at all.
33420 Sometimes with a program as complicated as @value{GDBN} it is very hard to
33421 construct an example that will make the program follow a certain path
33422 through the code. If you do not send us the example, we will not be able
33423 to construct one, so we will not be able to verify that the bug is fixed.
33425 And if we cannot understand what bug you are trying to fix, or why your
33426 patch should be an improvement, we will not install it. A test case will
33427 help us to understand.
33430 A guess about what the bug is or what it depends on.
33432 Such guesses are usually wrong. Even we cannot guess right about such
33433 things without first using the debugger to find the facts.
33436 @c The readline documentation is distributed with the readline code
33437 @c and consists of the two following files:
33440 @c Use -I with makeinfo to point to the appropriate directory,
33441 @c environment var TEXINPUTS with TeX.
33442 @ifclear SYSTEM_READLINE
33443 @include rluser.texi
33444 @include hsuser.texi
33448 @appendix In Memoriam
33450 The @value{GDBN} project mourns the loss of the following long-time
33455 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
33456 to Free Software in general. Outside of @value{GDBN}, he was known in
33457 the Amiga world for his series of Fish Disks, and the GeekGadget project.
33459 @item Michael Snyder
33460 Michael was one of the Global Maintainers of the @value{GDBN} project,
33461 with contributions recorded as early as 1996, until 2011. In addition
33462 to his day to day participation, he was a large driving force behind
33463 adding Reverse Debugging to @value{GDBN}.
33466 Beyond their technical contributions to the project, they were also
33467 enjoyable members of the Free Software Community. We will miss them.
33469 @node Formatting Documentation
33470 @appendix Formatting Documentation
33472 @cindex @value{GDBN} reference card
33473 @cindex reference card
33474 The @value{GDBN} 4 release includes an already-formatted reference card, ready
33475 for printing with PostScript or Ghostscript, in the @file{gdb}
33476 subdirectory of the main source directory@footnote{In
33477 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
33478 release.}. If you can use PostScript or Ghostscript with your printer,
33479 you can print the reference card immediately with @file{refcard.ps}.
33481 The release also includes the source for the reference card. You
33482 can format it, using @TeX{}, by typing:
33488 The @value{GDBN} reference card is designed to print in @dfn{landscape}
33489 mode on US ``letter'' size paper;
33490 that is, on a sheet 11 inches wide by 8.5 inches
33491 high. You will need to specify this form of printing as an option to
33492 your @sc{dvi} output program.
33494 @cindex documentation
33496 All the documentation for @value{GDBN} comes as part of the machine-readable
33497 distribution. The documentation is written in Texinfo format, which is
33498 a documentation system that uses a single source file to produce both
33499 on-line information and a printed manual. You can use one of the Info
33500 formatting commands to create the on-line version of the documentation
33501 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
33503 @value{GDBN} includes an already formatted copy of the on-line Info
33504 version of this manual in the @file{gdb} subdirectory. The main Info
33505 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
33506 subordinate files matching @samp{gdb.info*} in the same directory. If
33507 necessary, you can print out these files, or read them with any editor;
33508 but they are easier to read using the @code{info} subsystem in @sc{gnu}
33509 Emacs or the standalone @code{info} program, available as part of the
33510 @sc{gnu} Texinfo distribution.
33512 If you want to format these Info files yourself, you need one of the
33513 Info formatting programs, such as @code{texinfo-format-buffer} or
33516 If you have @code{makeinfo} installed, and are in the top level
33517 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
33518 version @value{GDBVN}), you can make the Info file by typing:
33525 If you want to typeset and print copies of this manual, you need @TeX{},
33526 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
33527 Texinfo definitions file.
33529 @TeX{} is a typesetting program; it does not print files directly, but
33530 produces output files called @sc{dvi} files. To print a typeset
33531 document, you need a program to print @sc{dvi} files. If your system
33532 has @TeX{} installed, chances are it has such a program. The precise
33533 command to use depends on your system; @kbd{lpr -d} is common; another
33534 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
33535 require a file name without any extension or a @samp{.dvi} extension.
33537 @TeX{} also requires a macro definitions file called
33538 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
33539 written in Texinfo format. On its own, @TeX{} cannot either read or
33540 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
33541 and is located in the @file{gdb-@var{version-number}/texinfo}
33544 If you have @TeX{} and a @sc{dvi} printer program installed, you can
33545 typeset and print this manual. First switch to the @file{gdb}
33546 subdirectory of the main source directory (for example, to
33547 @file{gdb-@value{GDBVN}/gdb}) and type:
33553 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
33555 @node Installing GDB
33556 @appendix Installing @value{GDBN}
33557 @cindex installation
33560 * Requirements:: Requirements for building @value{GDBN}
33561 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
33562 * Separate Objdir:: Compiling @value{GDBN} in another directory
33563 * Config Names:: Specifying names for hosts and targets
33564 * Configure Options:: Summary of options for configure
33565 * System-wide configuration:: Having a system-wide init file
33569 @section Requirements for Building @value{GDBN}
33570 @cindex building @value{GDBN}, requirements for
33572 Building @value{GDBN} requires various tools and packages to be available.
33573 Other packages will be used only if they are found.
33575 @heading Tools/Packages Necessary for Building @value{GDBN}
33577 @item ISO C90 compiler
33578 @value{GDBN} is written in ISO C90. It should be buildable with any
33579 working C90 compiler, e.g.@: GCC.
33583 @heading Tools/Packages Optional for Building @value{GDBN}
33587 @value{GDBN} can use the Expat XML parsing library. This library may be
33588 included with your operating system distribution; if it is not, you
33589 can get the latest version from @url{http://expat.sourceforge.net}.
33590 The @file{configure} script will search for this library in several
33591 standard locations; if it is installed in an unusual path, you can
33592 use the @option{--with-libexpat-prefix} option to specify its location.
33598 Remote protocol memory maps (@pxref{Memory Map Format})
33600 Target descriptions (@pxref{Target Descriptions})
33602 Remote shared library lists (@xref{Library List Format},
33603 or alternatively @pxref{Library List Format for SVR4 Targets})
33605 MS-Windows shared libraries (@pxref{Shared Libraries})
33607 Traceframe info (@pxref{Traceframe Info Format})
33609 Branch trace (@pxref{Branch Trace Format},
33610 @pxref{Branch Trace Configuration Format})
33614 @cindex compressed debug sections
33615 @value{GDBN} will use the @samp{zlib} library, if available, to read
33616 compressed debug sections. Some linkers, such as GNU gold, are capable
33617 of producing binaries with compressed debug sections. If @value{GDBN}
33618 is compiled with @samp{zlib}, it will be able to read the debug
33619 information in such binaries.
33621 The @samp{zlib} library is likely included with your operating system
33622 distribution; if it is not, you can get the latest version from
33623 @url{http://zlib.net}.
33626 @value{GDBN}'s features related to character sets (@pxref{Character
33627 Sets}) require a functioning @code{iconv} implementation. If you are
33628 on a GNU system, then this is provided by the GNU C Library. Some
33629 other systems also provide a working @code{iconv}.
33631 If @value{GDBN} is using the @code{iconv} program which is installed
33632 in a non-standard place, you will need to tell @value{GDBN} where to find it.
33633 This is done with @option{--with-iconv-bin} which specifies the
33634 directory that contains the @code{iconv} program.
33636 On systems without @code{iconv}, you can install GNU Libiconv. If you
33637 have previously installed Libiconv, you can use the
33638 @option{--with-libiconv-prefix} option to configure.
33640 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
33641 arrange to build Libiconv if a directory named @file{libiconv} appears
33642 in the top-most source directory. If Libiconv is built this way, and
33643 if the operating system does not provide a suitable @code{iconv}
33644 implementation, then the just-built library will automatically be used
33645 by @value{GDBN}. One easy way to set this up is to download GNU
33646 Libiconv, unpack it, and then rename the directory holding the
33647 Libiconv source code to @samp{libiconv}.
33650 @node Running Configure
33651 @section Invoking the @value{GDBN} @file{configure} Script
33652 @cindex configuring @value{GDBN}
33653 @value{GDBN} comes with a @file{configure} script that automates the process
33654 of preparing @value{GDBN} for installation; you can then use @code{make} to
33655 build the @code{gdb} program.
33657 @c irrelevant in info file; it's as current as the code it lives with.
33658 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
33659 look at the @file{README} file in the sources; we may have improved the
33660 installation procedures since publishing this manual.}
33663 The @value{GDBN} distribution includes all the source code you need for
33664 @value{GDBN} in a single directory, whose name is usually composed by
33665 appending the version number to @samp{gdb}.
33667 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
33668 @file{gdb-@value{GDBVN}} directory. That directory contains:
33671 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
33672 script for configuring @value{GDBN} and all its supporting libraries
33674 @item gdb-@value{GDBVN}/gdb
33675 the source specific to @value{GDBN} itself
33677 @item gdb-@value{GDBVN}/bfd
33678 source for the Binary File Descriptor library
33680 @item gdb-@value{GDBVN}/include
33681 @sc{gnu} include files
33683 @item gdb-@value{GDBVN}/libiberty
33684 source for the @samp{-liberty} free software library
33686 @item gdb-@value{GDBVN}/opcodes
33687 source for the library of opcode tables and disassemblers
33689 @item gdb-@value{GDBVN}/readline
33690 source for the @sc{gnu} command-line interface
33692 @item gdb-@value{GDBVN}/glob
33693 source for the @sc{gnu} filename pattern-matching subroutine
33695 @item gdb-@value{GDBVN}/mmalloc
33696 source for the @sc{gnu} memory-mapped malloc package
33699 The simplest way to configure and build @value{GDBN} is to run @file{configure}
33700 from the @file{gdb-@var{version-number}} source directory, which in
33701 this example is the @file{gdb-@value{GDBVN}} directory.
33703 First switch to the @file{gdb-@var{version-number}} source directory
33704 if you are not already in it; then run @file{configure}. Pass the
33705 identifier for the platform on which @value{GDBN} will run as an
33711 cd gdb-@value{GDBVN}
33712 ./configure @var{host}
33717 where @var{host} is an identifier such as @samp{sun4} or
33718 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
33719 (You can often leave off @var{host}; @file{configure} tries to guess the
33720 correct value by examining your system.)
33722 Running @samp{configure @var{host}} and then running @code{make} builds the
33723 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
33724 libraries, then @code{gdb} itself. The configured source files, and the
33725 binaries, are left in the corresponding source directories.
33728 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
33729 system does not recognize this automatically when you run a different
33730 shell, you may need to run @code{sh} on it explicitly:
33733 sh configure @var{host}
33736 If you run @file{configure} from a directory that contains source
33737 directories for multiple libraries or programs, such as the
33738 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
33740 creates configuration files for every directory level underneath (unless
33741 you tell it not to, with the @samp{--norecursion} option).
33743 You should run the @file{configure} script from the top directory in the
33744 source tree, the @file{gdb-@var{version-number}} directory. If you run
33745 @file{configure} from one of the subdirectories, you will configure only
33746 that subdirectory. That is usually not what you want. In particular,
33747 if you run the first @file{configure} from the @file{gdb} subdirectory
33748 of the @file{gdb-@var{version-number}} directory, you will omit the
33749 configuration of @file{bfd}, @file{readline}, and other sibling
33750 directories of the @file{gdb} subdirectory. This leads to build errors
33751 about missing include files such as @file{bfd/bfd.h}.
33753 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
33754 However, you should make sure that the shell on your path (named by
33755 the @samp{SHELL} environment variable) is publicly readable. Remember
33756 that @value{GDBN} uses the shell to start your program---some systems refuse to
33757 let @value{GDBN} debug child processes whose programs are not readable.
33759 @node Separate Objdir
33760 @section Compiling @value{GDBN} in Another Directory
33762 If you want to run @value{GDBN} versions for several host or target machines,
33763 you need a different @code{gdb} compiled for each combination of
33764 host and target. @file{configure} is designed to make this easy by
33765 allowing you to generate each configuration in a separate subdirectory,
33766 rather than in the source directory. If your @code{make} program
33767 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
33768 @code{make} in each of these directories builds the @code{gdb}
33769 program specified there.
33771 To build @code{gdb} in a separate directory, run @file{configure}
33772 with the @samp{--srcdir} option to specify where to find the source.
33773 (You also need to specify a path to find @file{configure}
33774 itself from your working directory. If the path to @file{configure}
33775 would be the same as the argument to @samp{--srcdir}, you can leave out
33776 the @samp{--srcdir} option; it is assumed.)
33778 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
33779 separate directory for a Sun 4 like this:
33783 cd gdb-@value{GDBVN}
33786 ../gdb-@value{GDBVN}/configure sun4
33791 When @file{configure} builds a configuration using a remote source
33792 directory, it creates a tree for the binaries with the same structure
33793 (and using the same names) as the tree under the source directory. In
33794 the example, you'd find the Sun 4 library @file{libiberty.a} in the
33795 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
33796 @file{gdb-sun4/gdb}.
33798 Make sure that your path to the @file{configure} script has just one
33799 instance of @file{gdb} in it. If your path to @file{configure} looks
33800 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
33801 one subdirectory of @value{GDBN}, not the whole package. This leads to
33802 build errors about missing include files such as @file{bfd/bfd.h}.
33804 One popular reason to build several @value{GDBN} configurations in separate
33805 directories is to configure @value{GDBN} for cross-compiling (where
33806 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
33807 programs that run on another machine---the @dfn{target}).
33808 You specify a cross-debugging target by
33809 giving the @samp{--target=@var{target}} option to @file{configure}.
33811 When you run @code{make} to build a program or library, you must run
33812 it in a configured directory---whatever directory you were in when you
33813 called @file{configure} (or one of its subdirectories).
33815 The @code{Makefile} that @file{configure} generates in each source
33816 directory also runs recursively. If you type @code{make} in a source
33817 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
33818 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
33819 will build all the required libraries, and then build GDB.
33821 When you have multiple hosts or targets configured in separate
33822 directories, you can run @code{make} on them in parallel (for example,
33823 if they are NFS-mounted on each of the hosts); they will not interfere
33827 @section Specifying Names for Hosts and Targets
33829 The specifications used for hosts and targets in the @file{configure}
33830 script are based on a three-part naming scheme, but some short predefined
33831 aliases are also supported. The full naming scheme encodes three pieces
33832 of information in the following pattern:
33835 @var{architecture}-@var{vendor}-@var{os}
33838 For example, you can use the alias @code{sun4} as a @var{host} argument,
33839 or as the value for @var{target} in a @code{--target=@var{target}}
33840 option. The equivalent full name is @samp{sparc-sun-sunos4}.
33842 The @file{configure} script accompanying @value{GDBN} does not provide
33843 any query facility to list all supported host and target names or
33844 aliases. @file{configure} calls the Bourne shell script
33845 @code{config.sub} to map abbreviations to full names; you can read the
33846 script, if you wish, or you can use it to test your guesses on
33847 abbreviations---for example:
33850 % sh config.sub i386-linux
33852 % sh config.sub alpha-linux
33853 alpha-unknown-linux-gnu
33854 % sh config.sub hp9k700
33856 % sh config.sub sun4
33857 sparc-sun-sunos4.1.1
33858 % sh config.sub sun3
33859 m68k-sun-sunos4.1.1
33860 % sh config.sub i986v
33861 Invalid configuration `i986v': machine `i986v' not recognized
33865 @code{config.sub} is also distributed in the @value{GDBN} source
33866 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
33868 @node Configure Options
33869 @section @file{configure} Options
33871 Here is a summary of the @file{configure} options and arguments that
33872 are most often useful for building @value{GDBN}. @file{configure} also has
33873 several other options not listed here. @inforef{What Configure
33874 Does,,configure.info}, for a full explanation of @file{configure}.
33877 configure @r{[}--help@r{]}
33878 @r{[}--prefix=@var{dir}@r{]}
33879 @r{[}--exec-prefix=@var{dir}@r{]}
33880 @r{[}--srcdir=@var{dirname}@r{]}
33881 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
33882 @r{[}--target=@var{target}@r{]}
33887 You may introduce options with a single @samp{-} rather than
33888 @samp{--} if you prefer; but you may abbreviate option names if you use
33893 Display a quick summary of how to invoke @file{configure}.
33895 @item --prefix=@var{dir}
33896 Configure the source to install programs and files under directory
33899 @item --exec-prefix=@var{dir}
33900 Configure the source to install programs under directory
33903 @c avoid splitting the warning from the explanation:
33905 @item --srcdir=@var{dirname}
33906 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
33907 @code{make} that implements the @code{VPATH} feature.}@*
33908 Use this option to make configurations in directories separate from the
33909 @value{GDBN} source directories. Among other things, you can use this to
33910 build (or maintain) several configurations simultaneously, in separate
33911 directories. @file{configure} writes configuration-specific files in
33912 the current directory, but arranges for them to use the source in the
33913 directory @var{dirname}. @file{configure} creates directories under
33914 the working directory in parallel to the source directories below
33917 @item --norecursion
33918 Configure only the directory level where @file{configure} is executed; do not
33919 propagate configuration to subdirectories.
33921 @item --target=@var{target}
33922 Configure @value{GDBN} for cross-debugging programs running on the specified
33923 @var{target}. Without this option, @value{GDBN} is configured to debug
33924 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
33926 There is no convenient way to generate a list of all available targets.
33928 @item @var{host} @dots{}
33929 Configure @value{GDBN} to run on the specified @var{host}.
33931 There is no convenient way to generate a list of all available hosts.
33934 There are many other options available as well, but they are generally
33935 needed for special purposes only.
33937 @node System-wide configuration
33938 @section System-wide configuration and settings
33939 @cindex system-wide init file
33941 @value{GDBN} can be configured to have a system-wide init file;
33942 this file will be read and executed at startup (@pxref{Startup, , What
33943 @value{GDBN} does during startup}).
33945 Here is the corresponding configure option:
33948 @item --with-system-gdbinit=@var{file}
33949 Specify that the default location of the system-wide init file is
33953 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
33954 it may be subject to relocation. Two possible cases:
33958 If the default location of this init file contains @file{$prefix},
33959 it will be subject to relocation. Suppose that the configure options
33960 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
33961 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
33962 init file is looked for as @file{$install/etc/gdbinit} instead of
33963 @file{$prefix/etc/gdbinit}.
33966 By contrast, if the default location does not contain the prefix,
33967 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
33968 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
33969 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
33970 wherever @value{GDBN} is installed.
33973 If the configured location of the system-wide init file (as given by the
33974 @option{--with-system-gdbinit} option at configure time) is in the
33975 data-directory (as specified by @option{--with-gdb-datadir} at configure
33976 time) or in one of its subdirectories, then @value{GDBN} will look for the
33977 system-wide init file in the directory specified by the
33978 @option{--data-directory} command-line option.
33979 Note that the system-wide init file is only read once, during @value{GDBN}
33980 initialization. If the data-directory is changed after @value{GDBN} has
33981 started with the @code{set data-directory} command, the file will not be
33985 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
33988 @node System-wide Configuration Scripts
33989 @subsection Installed System-wide Configuration Scripts
33990 @cindex system-wide configuration scripts
33992 The @file{system-gdbinit} directory, located inside the data-directory
33993 (as specified by @option{--with-gdb-datadir} at configure time) contains
33994 a number of scripts which can be used as system-wide init files. To
33995 automatically source those scripts at startup, @value{GDBN} should be
33996 configured with @option{--with-system-gdbinit}. Otherwise, any user
33997 should be able to source them by hand as needed.
33999 The following scripts are currently available:
34002 @item @file{elinos.py}
34004 @cindex ELinOS system-wide configuration script
34005 This script is useful when debugging a program on an ELinOS target.
34006 It takes advantage of the environment variables defined in a standard
34007 ELinOS environment in order to determine the location of the system
34008 shared libraries, and then sets the @samp{solib-absolute-prefix}
34009 and @samp{solib-search-path} variables appropriately.
34011 @item @file{wrs-linux.py}
34012 @pindex wrs-linux.py
34013 @cindex Wind River Linux system-wide configuration script
34014 This script is useful when debugging a program on a target running
34015 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
34016 the host-side sysroot used by the target system.
34020 @node Maintenance Commands
34021 @appendix Maintenance Commands
34022 @cindex maintenance commands
34023 @cindex internal commands
34025 In addition to commands intended for @value{GDBN} users, @value{GDBN}
34026 includes a number of commands intended for @value{GDBN} developers,
34027 that are not documented elsewhere in this manual. These commands are
34028 provided here for reference. (For commands that turn on debugging
34029 messages, see @ref{Debugging Output}.)
34032 @kindex maint agent
34033 @kindex maint agent-eval
34034 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
34035 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
34036 Translate the given @var{expression} into remote agent bytecodes.
34037 This command is useful for debugging the Agent Expression mechanism
34038 (@pxref{Agent Expressions}). The @samp{agent} version produces an
34039 expression useful for data collection, such as by tracepoints, while
34040 @samp{maint agent-eval} produces an expression that evaluates directly
34041 to a result. For instance, a collection expression for @code{globa +
34042 globb} will include bytecodes to record four bytes of memory at each
34043 of the addresses of @code{globa} and @code{globb}, while discarding
34044 the result of the addition, while an evaluation expression will do the
34045 addition and return the sum.
34046 If @code{-at} is given, generate remote agent bytecode for @var{location}.
34047 If not, generate remote agent bytecode for current frame PC address.
34049 @kindex maint agent-printf
34050 @item maint agent-printf @var{format},@var{expr},...
34051 Translate the given format string and list of argument expressions
34052 into remote agent bytecodes and display them as a disassembled list.
34053 This command is useful for debugging the agent version of dynamic
34054 printf (@pxref{Dynamic Printf}).
34056 @kindex maint info breakpoints
34057 @item @anchor{maint info breakpoints}maint info breakpoints
34058 Using the same format as @samp{info breakpoints}, display both the
34059 breakpoints you've set explicitly, and those @value{GDBN} is using for
34060 internal purposes. Internal breakpoints are shown with negative
34061 breakpoint numbers. The type column identifies what kind of breakpoint
34066 Normal, explicitly set breakpoint.
34069 Normal, explicitly set watchpoint.
34072 Internal breakpoint, used to handle correctly stepping through
34073 @code{longjmp} calls.
34075 @item longjmp resume
34076 Internal breakpoint at the target of a @code{longjmp}.
34079 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
34082 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
34085 Shared library events.
34089 @kindex maint info btrace
34090 @item maint info btrace
34091 Pint information about raw branch tracing data.
34093 @kindex maint btrace packet-history
34094 @item maint btrace packet-history
34095 Print the raw branch trace packets that are used to compute the
34096 execution history for the @samp{record btrace} command. Both the
34097 information and the format in which it is printed depend on the btrace
34102 For the BTS recording format, print a list of blocks of sequential
34103 code. For each block, the following information is printed:
34107 Newer blocks have higher numbers. The oldest block has number zero.
34108 @item Lowest @samp{PC}
34109 @item Highest @samp{PC}
34113 For the Intel Processor Trace recording format, print a list of
34114 Intel Processor Trace packets. For each packet, the following
34115 information is printed:
34118 @item Packet number
34119 Newer packets have higher numbers. The oldest packet has number zero.
34121 The packet's offset in the trace stream.
34122 @item Packet opcode and payload
34126 @kindex maint btrace clear-packet-history
34127 @item maint btrace clear-packet-history
34128 Discards the cached packet history printed by the @samp{maint btrace
34129 packet-history} command. The history will be computed again when
34132 @kindex maint btrace clear
34133 @item maint btrace clear
34134 Discard the branch trace data. The data will be fetched anew and the
34135 branch trace will be recomputed when needed.
34137 This implicitly truncates the branch trace to a single branch trace
34138 buffer. When updating branch trace incrementally, the branch trace
34139 available to @value{GDBN} may be bigger than a single branch trace
34142 @kindex maint set btrace pt skip-pad
34143 @item maint set btrace pt skip-pad
34144 @kindex maint show btrace pt skip-pad
34145 @item maint show btrace pt skip-pad
34146 Control whether @value{GDBN} will skip PAD packets when computing the
34149 @kindex set displaced-stepping
34150 @kindex show displaced-stepping
34151 @cindex displaced stepping support
34152 @cindex out-of-line single-stepping
34153 @item set displaced-stepping
34154 @itemx show displaced-stepping
34155 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
34156 if the target supports it. Displaced stepping is a way to single-step
34157 over breakpoints without removing them from the inferior, by executing
34158 an out-of-line copy of the instruction that was originally at the
34159 breakpoint location. It is also known as out-of-line single-stepping.
34162 @item set displaced-stepping on
34163 If the target architecture supports it, @value{GDBN} will use
34164 displaced stepping to step over breakpoints.
34166 @item set displaced-stepping off
34167 @value{GDBN} will not use displaced stepping to step over breakpoints,
34168 even if such is supported by the target architecture.
34170 @cindex non-stop mode, and @samp{set displaced-stepping}
34171 @item set displaced-stepping auto
34172 This is the default mode. @value{GDBN} will use displaced stepping
34173 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
34174 architecture supports displaced stepping.
34177 @kindex maint check-psymtabs
34178 @item maint check-psymtabs
34179 Check the consistency of currently expanded psymtabs versus symtabs.
34180 Use this to check, for example, whether a symbol is in one but not the other.
34182 @kindex maint check-symtabs
34183 @item maint check-symtabs
34184 Check the consistency of currently expanded symtabs.
34186 @kindex maint expand-symtabs
34187 @item maint expand-symtabs [@var{regexp}]
34188 Expand symbol tables.
34189 If @var{regexp} is specified, only expand symbol tables for file
34190 names matching @var{regexp}.
34192 @kindex maint set catch-demangler-crashes
34193 @kindex maint show catch-demangler-crashes
34194 @cindex demangler crashes
34195 @item maint set catch-demangler-crashes [on|off]
34196 @itemx maint show catch-demangler-crashes
34197 Control whether @value{GDBN} should attempt to catch crashes in the
34198 symbol name demangler. The default is to attempt to catch crashes.
34199 If enabled, the first time a crash is caught, a core file is created,
34200 the offending symbol is displayed and the user is presented with the
34201 option to terminate the current session.
34203 @kindex maint cplus first_component
34204 @item maint cplus first_component @var{name}
34205 Print the first C@t{++} class/namespace component of @var{name}.
34207 @kindex maint cplus namespace
34208 @item maint cplus namespace
34209 Print the list of possible C@t{++} namespaces.
34211 @kindex maint deprecate
34212 @kindex maint undeprecate
34213 @cindex deprecated commands
34214 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
34215 @itemx maint undeprecate @var{command}
34216 Deprecate or undeprecate the named @var{command}. Deprecated commands
34217 cause @value{GDBN} to issue a warning when you use them. The optional
34218 argument @var{replacement} says which newer command should be used in
34219 favor of the deprecated one; if it is given, @value{GDBN} will mention
34220 the replacement as part of the warning.
34222 @kindex maint dump-me
34223 @item maint dump-me
34224 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
34225 Cause a fatal signal in the debugger and force it to dump its core.
34226 This is supported only on systems which support aborting a program
34227 with the @code{SIGQUIT} signal.
34229 @kindex maint internal-error
34230 @kindex maint internal-warning
34231 @kindex maint demangler-warning
34232 @cindex demangler crashes
34233 @item maint internal-error @r{[}@var{message-text}@r{]}
34234 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
34235 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
34237 Cause @value{GDBN} to call the internal function @code{internal_error},
34238 @code{internal_warning} or @code{demangler_warning} and hence behave
34239 as though an internal problem has been detected. In addition to
34240 reporting the internal problem, these functions give the user the
34241 opportunity to either quit @value{GDBN} or (for @code{internal_error}
34242 and @code{internal_warning}) create a core file of the current
34243 @value{GDBN} session.
34245 These commands take an optional parameter @var{message-text} that is
34246 used as the text of the error or warning message.
34248 Here's an example of using @code{internal-error}:
34251 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
34252 @dots{}/maint.c:121: internal-error: testing, 1, 2
34253 A problem internal to GDB has been detected. Further
34254 debugging may prove unreliable.
34255 Quit this debugging session? (y or n) @kbd{n}
34256 Create a core file? (y or n) @kbd{n}
34260 @cindex @value{GDBN} internal error
34261 @cindex internal errors, control of @value{GDBN} behavior
34262 @cindex demangler crashes
34264 @kindex maint set internal-error
34265 @kindex maint show internal-error
34266 @kindex maint set internal-warning
34267 @kindex maint show internal-warning
34268 @kindex maint set demangler-warning
34269 @kindex maint show demangler-warning
34270 @item maint set internal-error @var{action} [ask|yes|no]
34271 @itemx maint show internal-error @var{action}
34272 @itemx maint set internal-warning @var{action} [ask|yes|no]
34273 @itemx maint show internal-warning @var{action}
34274 @itemx maint set demangler-warning @var{action} [ask|yes|no]
34275 @itemx maint show demangler-warning @var{action}
34276 When @value{GDBN} reports an internal problem (error or warning) it
34277 gives the user the opportunity to both quit @value{GDBN} and create a
34278 core file of the current @value{GDBN} session. These commands let you
34279 override the default behaviour for each particular @var{action},
34280 described in the table below.
34284 You can specify that @value{GDBN} should always (yes) or never (no)
34285 quit. The default is to ask the user what to do.
34288 You can specify that @value{GDBN} should always (yes) or never (no)
34289 create a core file. The default is to ask the user what to do. Note
34290 that there is no @code{corefile} option for @code{demangler-warning}:
34291 demangler warnings always create a core file and this cannot be
34295 @kindex maint packet
34296 @item maint packet @var{text}
34297 If @value{GDBN} is talking to an inferior via the serial protocol,
34298 then this command sends the string @var{text} to the inferior, and
34299 displays the response packet. @value{GDBN} supplies the initial
34300 @samp{$} character, the terminating @samp{#} character, and the
34303 @kindex maint print architecture
34304 @item maint print architecture @r{[}@var{file}@r{]}
34305 Print the entire architecture configuration. The optional argument
34306 @var{file} names the file where the output goes.
34308 @kindex maint print c-tdesc
34309 @item maint print c-tdesc
34310 Print the current target description (@pxref{Target Descriptions}) as
34311 a C source file. The created source file can be used in @value{GDBN}
34312 when an XML parser is not available to parse the description.
34314 @kindex maint print dummy-frames
34315 @item maint print dummy-frames
34316 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
34319 (@value{GDBP}) @kbd{b add}
34321 (@value{GDBP}) @kbd{print add(2,3)}
34322 Breakpoint 2, add (a=2, b=3) at @dots{}
34324 The program being debugged stopped while in a function called from GDB.
34326 (@value{GDBP}) @kbd{maint print dummy-frames}
34327 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
34331 Takes an optional file parameter.
34333 @kindex maint print registers
34334 @kindex maint print raw-registers
34335 @kindex maint print cooked-registers
34336 @kindex maint print register-groups
34337 @kindex maint print remote-registers
34338 @item maint print registers @r{[}@var{file}@r{]}
34339 @itemx maint print raw-registers @r{[}@var{file}@r{]}
34340 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
34341 @itemx maint print register-groups @r{[}@var{file}@r{]}
34342 @itemx maint print remote-registers @r{[}@var{file}@r{]}
34343 Print @value{GDBN}'s internal register data structures.
34345 The command @code{maint print raw-registers} includes the contents of
34346 the raw register cache; the command @code{maint print
34347 cooked-registers} includes the (cooked) value of all registers,
34348 including registers which aren't available on the target nor visible
34349 to user; the command @code{maint print register-groups} includes the
34350 groups that each register is a member of; and the command @code{maint
34351 print remote-registers} includes the remote target's register numbers
34352 and offsets in the `G' packets.
34354 These commands take an optional parameter, a file name to which to
34355 write the information.
34357 @kindex maint print reggroups
34358 @item maint print reggroups @r{[}@var{file}@r{]}
34359 Print @value{GDBN}'s internal register group data structures. The
34360 optional argument @var{file} tells to what file to write the
34363 The register groups info looks like this:
34366 (@value{GDBP}) @kbd{maint print reggroups}
34379 This command forces @value{GDBN} to flush its internal register cache.
34381 @kindex maint print objfiles
34382 @cindex info for known object files
34383 @item maint print objfiles @r{[}@var{regexp}@r{]}
34384 Print a dump of all known object files.
34385 If @var{regexp} is specified, only print object files whose names
34386 match @var{regexp}. For each object file, this command prints its name,
34387 address in memory, and all of its psymtabs and symtabs.
34389 @kindex maint print user-registers
34390 @cindex user registers
34391 @item maint print user-registers
34392 List all currently available @dfn{user registers}. User registers
34393 typically provide alternate names for actual hardware registers. They
34394 include the four ``standard'' registers @code{$fp}, @code{$pc},
34395 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
34396 registers can be used in expressions in the same way as the canonical
34397 register names, but only the latter are listed by the @code{info
34398 registers} and @code{maint print registers} commands.
34400 @kindex maint print section-scripts
34401 @cindex info for known .debug_gdb_scripts-loaded scripts
34402 @item maint print section-scripts [@var{regexp}]
34403 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
34404 If @var{regexp} is specified, only print scripts loaded by object files
34405 matching @var{regexp}.
34406 For each script, this command prints its name as specified in the objfile,
34407 and the full path if known.
34408 @xref{dotdebug_gdb_scripts section}.
34410 @kindex maint print statistics
34411 @cindex bcache statistics
34412 @item maint print statistics
34413 This command prints, for each object file in the program, various data
34414 about that object file followed by the byte cache (@dfn{bcache})
34415 statistics for the object file. The objfile data includes the number
34416 of minimal, partial, full, and stabs symbols, the number of types
34417 defined by the objfile, the number of as yet unexpanded psym tables,
34418 the number of line tables and string tables, and the amount of memory
34419 used by the various tables. The bcache statistics include the counts,
34420 sizes, and counts of duplicates of all and unique objects, max,
34421 average, and median entry size, total memory used and its overhead and
34422 savings, and various measures of the hash table size and chain
34425 @kindex maint print target-stack
34426 @cindex target stack description
34427 @item maint print target-stack
34428 A @dfn{target} is an interface between the debugger and a particular
34429 kind of file or process. Targets can be stacked in @dfn{strata},
34430 so that more than one target can potentially respond to a request.
34431 In particular, memory accesses will walk down the stack of targets
34432 until they find a target that is interested in handling that particular
34435 This command prints a short description of each layer that was pushed on
34436 the @dfn{target stack}, starting from the top layer down to the bottom one.
34438 @kindex maint print type
34439 @cindex type chain of a data type
34440 @item maint print type @var{expr}
34441 Print the type chain for a type specified by @var{expr}. The argument
34442 can be either a type name or a symbol. If it is a symbol, the type of
34443 that symbol is described. The type chain produced by this command is
34444 a recursive definition of the data type as stored in @value{GDBN}'s
34445 data structures, including its flags and contained types.
34447 @kindex maint set dwarf always-disassemble
34448 @kindex maint show dwarf always-disassemble
34449 @item maint set dwarf always-disassemble
34450 @item maint show dwarf always-disassemble
34451 Control the behavior of @code{info address} when using DWARF debugging
34454 The default is @code{off}, which means that @value{GDBN} should try to
34455 describe a variable's location in an easily readable format. When
34456 @code{on}, @value{GDBN} will instead display the DWARF location
34457 expression in an assembly-like format. Note that some locations are
34458 too complex for @value{GDBN} to describe simply; in this case you will
34459 always see the disassembly form.
34461 Here is an example of the resulting disassembly:
34464 (gdb) info addr argc
34465 Symbol "argc" is a complex DWARF expression:
34469 For more information on these expressions, see
34470 @uref{http://www.dwarfstd.org/, the DWARF standard}.
34472 @kindex maint set dwarf max-cache-age
34473 @kindex maint show dwarf max-cache-age
34474 @item maint set dwarf max-cache-age
34475 @itemx maint show dwarf max-cache-age
34476 Control the DWARF compilation unit cache.
34478 @cindex DWARF compilation units cache
34479 In object files with inter-compilation-unit references, such as those
34480 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
34481 reader needs to frequently refer to previously read compilation units.
34482 This setting controls how long a compilation unit will remain in the
34483 cache if it is not referenced. A higher limit means that cached
34484 compilation units will be stored in memory longer, and more total
34485 memory will be used. Setting it to zero disables caching, which will
34486 slow down @value{GDBN} startup, but reduce memory consumption.
34488 @kindex maint set profile
34489 @kindex maint show profile
34490 @cindex profiling GDB
34491 @item maint set profile
34492 @itemx maint show profile
34493 Control profiling of @value{GDBN}.
34495 Profiling will be disabled until you use the @samp{maint set profile}
34496 command to enable it. When you enable profiling, the system will begin
34497 collecting timing and execution count data; when you disable profiling or
34498 exit @value{GDBN}, the results will be written to a log file. Remember that
34499 if you use profiling, @value{GDBN} will overwrite the profiling log file
34500 (often called @file{gmon.out}). If you have a record of important profiling
34501 data in a @file{gmon.out} file, be sure to move it to a safe location.
34503 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
34504 compiled with the @samp{-pg} compiler option.
34506 @kindex maint set show-debug-regs
34507 @kindex maint show show-debug-regs
34508 @cindex hardware debug registers
34509 @item maint set show-debug-regs
34510 @itemx maint show show-debug-regs
34511 Control whether to show variables that mirror the hardware debug
34512 registers. Use @code{on} to enable, @code{off} to disable. If
34513 enabled, the debug registers values are shown when @value{GDBN} inserts or
34514 removes a hardware breakpoint or watchpoint, and when the inferior
34515 triggers a hardware-assisted breakpoint or watchpoint.
34517 @kindex maint set show-all-tib
34518 @kindex maint show show-all-tib
34519 @item maint set show-all-tib
34520 @itemx maint show show-all-tib
34521 Control whether to show all non zero areas within a 1k block starting
34522 at thread local base, when using the @samp{info w32 thread-information-block}
34525 @kindex maint set target-async
34526 @kindex maint show target-async
34527 @item maint set target-async
34528 @itemx maint show target-async
34529 This controls whether @value{GDBN} targets operate in synchronous or
34530 asynchronous mode (@pxref{Background Execution}). Normally the
34531 default is asynchronous, if it is available; but this can be changed
34532 to more easily debug problems occurring only in synchronous mode.
34534 @kindex maint set target-non-stop @var{mode} [on|off|auto]
34535 @kindex maint show target-non-stop
34536 @item maint set target-non-stop
34537 @itemx maint show target-non-stop
34539 This controls whether @value{GDBN} targets always operate in non-stop
34540 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
34541 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
34542 if supported by the target.
34545 @item maint set target-non-stop auto
34546 This is the default mode. @value{GDBN} controls the target in
34547 non-stop mode if the target supports it.
34549 @item maint set target-non-stop on
34550 @value{GDBN} controls the target in non-stop mode even if the target
34551 does not indicate support.
34553 @item maint set target-non-stop off
34554 @value{GDBN} does not control the target in non-stop mode even if the
34555 target supports it.
34558 @kindex maint set per-command
34559 @kindex maint show per-command
34560 @item maint set per-command
34561 @itemx maint show per-command
34562 @cindex resources used by commands
34564 @value{GDBN} can display the resources used by each command.
34565 This is useful in debugging performance problems.
34568 @item maint set per-command space [on|off]
34569 @itemx maint show per-command space
34570 Enable or disable the printing of the memory used by GDB for each command.
34571 If enabled, @value{GDBN} will display how much memory each command
34572 took, following the command's own output.
34573 This can also be requested by invoking @value{GDBN} with the
34574 @option{--statistics} command-line switch (@pxref{Mode Options}).
34576 @item maint set per-command time [on|off]
34577 @itemx maint show per-command time
34578 Enable or disable the printing of the execution time of @value{GDBN}
34580 If enabled, @value{GDBN} will display how much time it
34581 took to execute each command, following the command's own output.
34582 Both CPU time and wallclock time are printed.
34583 Printing both is useful when trying to determine whether the cost is
34584 CPU or, e.g., disk/network latency.
34585 Note that the CPU time printed is for @value{GDBN} only, it does not include
34586 the execution time of the inferior because there's no mechanism currently
34587 to compute how much time was spent by @value{GDBN} and how much time was
34588 spent by the program been debugged.
34589 This can also be requested by invoking @value{GDBN} with the
34590 @option{--statistics} command-line switch (@pxref{Mode Options}).
34592 @item maint set per-command symtab [on|off]
34593 @itemx maint show per-command symtab
34594 Enable or disable the printing of basic symbol table statistics
34596 If enabled, @value{GDBN} will display the following information:
34600 number of symbol tables
34602 number of primary symbol tables
34604 number of blocks in the blockvector
34608 @kindex maint space
34609 @cindex memory used by commands
34610 @item maint space @var{value}
34611 An alias for @code{maint set per-command space}.
34612 A non-zero value enables it, zero disables it.
34615 @cindex time of command execution
34616 @item maint time @var{value}
34617 An alias for @code{maint set per-command time}.
34618 A non-zero value enables it, zero disables it.
34620 @kindex maint translate-address
34621 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
34622 Find the symbol stored at the location specified by the address
34623 @var{addr} and an optional section name @var{section}. If found,
34624 @value{GDBN} prints the name of the closest symbol and an offset from
34625 the symbol's location to the specified address. This is similar to
34626 the @code{info address} command (@pxref{Symbols}), except that this
34627 command also allows to find symbols in other sections.
34629 If section was not specified, the section in which the symbol was found
34630 is also printed. For dynamically linked executables, the name of
34631 executable or shared library containing the symbol is printed as well.
34635 The following command is useful for non-interactive invocations of
34636 @value{GDBN}, such as in the test suite.
34639 @item set watchdog @var{nsec}
34640 @kindex set watchdog
34641 @cindex watchdog timer
34642 @cindex timeout for commands
34643 Set the maximum number of seconds @value{GDBN} will wait for the
34644 target operation to finish. If this time expires, @value{GDBN}
34645 reports and error and the command is aborted.
34647 @item show watchdog
34648 Show the current setting of the target wait timeout.
34651 @node Remote Protocol
34652 @appendix @value{GDBN} Remote Serial Protocol
34657 * Stop Reply Packets::
34658 * General Query Packets::
34659 * Architecture-Specific Protocol Details::
34660 * Tracepoint Packets::
34661 * Host I/O Packets::
34663 * Notification Packets::
34664 * Remote Non-Stop::
34665 * Packet Acknowledgment::
34667 * File-I/O Remote Protocol Extension::
34668 * Library List Format::
34669 * Library List Format for SVR4 Targets::
34670 * Memory Map Format::
34671 * Thread List Format::
34672 * Traceframe Info Format::
34673 * Branch Trace Format::
34674 * Branch Trace Configuration Format::
34680 There may be occasions when you need to know something about the
34681 protocol---for example, if there is only one serial port to your target
34682 machine, you might want your program to do something special if it
34683 recognizes a packet meant for @value{GDBN}.
34685 In the examples below, @samp{->} and @samp{<-} are used to indicate
34686 transmitted and received data, respectively.
34688 @cindex protocol, @value{GDBN} remote serial
34689 @cindex serial protocol, @value{GDBN} remote
34690 @cindex remote serial protocol
34691 All @value{GDBN} commands and responses (other than acknowledgments
34692 and notifications, see @ref{Notification Packets}) are sent as a
34693 @var{packet}. A @var{packet} is introduced with the character
34694 @samp{$}, the actual @var{packet-data}, and the terminating character
34695 @samp{#} followed by a two-digit @var{checksum}:
34698 @code{$}@var{packet-data}@code{#}@var{checksum}
34702 @cindex checksum, for @value{GDBN} remote
34704 The two-digit @var{checksum} is computed as the modulo 256 sum of all
34705 characters between the leading @samp{$} and the trailing @samp{#} (an
34706 eight bit unsigned checksum).
34708 Implementors should note that prior to @value{GDBN} 5.0 the protocol
34709 specification also included an optional two-digit @var{sequence-id}:
34712 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
34715 @cindex sequence-id, for @value{GDBN} remote
34717 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
34718 has never output @var{sequence-id}s. Stubs that handle packets added
34719 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
34721 When either the host or the target machine receives a packet, the first
34722 response expected is an acknowledgment: either @samp{+} (to indicate
34723 the package was received correctly) or @samp{-} (to request
34727 -> @code{$}@var{packet-data}@code{#}@var{checksum}
34732 The @samp{+}/@samp{-} acknowledgments can be disabled
34733 once a connection is established.
34734 @xref{Packet Acknowledgment}, for details.
34736 The host (@value{GDBN}) sends @var{command}s, and the target (the
34737 debugging stub incorporated in your program) sends a @var{response}. In
34738 the case of step and continue @var{command}s, the response is only sent
34739 when the operation has completed, and the target has again stopped all
34740 threads in all attached processes. This is the default all-stop mode
34741 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
34742 execution mode; see @ref{Remote Non-Stop}, for details.
34744 @var{packet-data} consists of a sequence of characters with the
34745 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
34748 @cindex remote protocol, field separator
34749 Fields within the packet should be separated using @samp{,} @samp{;} or
34750 @samp{:}. Except where otherwise noted all numbers are represented in
34751 @sc{hex} with leading zeros suppressed.
34753 Implementors should note that prior to @value{GDBN} 5.0, the character
34754 @samp{:} could not appear as the third character in a packet (as it
34755 would potentially conflict with the @var{sequence-id}).
34757 @cindex remote protocol, binary data
34758 @anchor{Binary Data}
34759 Binary data in most packets is encoded either as two hexadecimal
34760 digits per byte of binary data. This allowed the traditional remote
34761 protocol to work over connections which were only seven-bit clean.
34762 Some packets designed more recently assume an eight-bit clean
34763 connection, and use a more efficient encoding to send and receive
34766 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
34767 as an escape character. Any escaped byte is transmitted as the escape
34768 character followed by the original character XORed with @code{0x20}.
34769 For example, the byte @code{0x7d} would be transmitted as the two
34770 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
34771 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
34772 @samp{@}}) must always be escaped. Responses sent by the stub
34773 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
34774 is not interpreted as the start of a run-length encoded sequence
34777 Response @var{data} can be run-length encoded to save space.
34778 Run-length encoding replaces runs of identical characters with one
34779 instance of the repeated character, followed by a @samp{*} and a
34780 repeat count. The repeat count is itself sent encoded, to avoid
34781 binary characters in @var{data}: a value of @var{n} is sent as
34782 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
34783 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
34784 code 32) for a repeat count of 3. (This is because run-length
34785 encoding starts to win for counts 3 or more.) Thus, for example,
34786 @samp{0* } is a run-length encoding of ``0000'': the space character
34787 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
34790 The printable characters @samp{#} and @samp{$} or with a numeric value
34791 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
34792 seven repeats (@samp{$}) can be expanded using a repeat count of only
34793 five (@samp{"}). For example, @samp{00000000} can be encoded as
34796 The error response returned for some packets includes a two character
34797 error number. That number is not well defined.
34799 @cindex empty response, for unsupported packets
34800 For any @var{command} not supported by the stub, an empty response
34801 (@samp{$#00}) should be returned. That way it is possible to extend the
34802 protocol. A newer @value{GDBN} can tell if a packet is supported based
34805 At a minimum, a stub is required to support the @samp{g} and @samp{G}
34806 commands for register access, and the @samp{m} and @samp{M} commands
34807 for memory access. Stubs that only control single-threaded targets
34808 can implement run control with the @samp{c} (continue), and @samp{s}
34809 (step) commands. Stubs that support multi-threading targets should
34810 support the @samp{vCont} command. All other commands are optional.
34815 The following table provides a complete list of all currently defined
34816 @var{command}s and their corresponding response @var{data}.
34817 @xref{File-I/O Remote Protocol Extension}, for details about the File
34818 I/O extension of the remote protocol.
34820 Each packet's description has a template showing the packet's overall
34821 syntax, followed by an explanation of the packet's meaning. We
34822 include spaces in some of the templates for clarity; these are not
34823 part of the packet's syntax. No @value{GDBN} packet uses spaces to
34824 separate its components. For example, a template like @samp{foo
34825 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
34826 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
34827 @var{baz}. @value{GDBN} does not transmit a space character between the
34828 @samp{foo} and the @var{bar}, or between the @var{bar} and the
34831 @cindex @var{thread-id}, in remote protocol
34832 @anchor{thread-id syntax}
34833 Several packets and replies include a @var{thread-id} field to identify
34834 a thread. Normally these are positive numbers with a target-specific
34835 interpretation, formatted as big-endian hex strings. A @var{thread-id}
34836 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
34839 In addition, the remote protocol supports a multiprocess feature in
34840 which the @var{thread-id} syntax is extended to optionally include both
34841 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
34842 The @var{pid} (process) and @var{tid} (thread) components each have the
34843 format described above: a positive number with target-specific
34844 interpretation formatted as a big-endian hex string, literal @samp{-1}
34845 to indicate all processes or threads (respectively), or @samp{0} to
34846 indicate an arbitrary process or thread. Specifying just a process, as
34847 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
34848 error to specify all processes but a specific thread, such as
34849 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
34850 for those packets and replies explicitly documented to include a process
34851 ID, rather than a @var{thread-id}.
34853 The multiprocess @var{thread-id} syntax extensions are only used if both
34854 @value{GDBN} and the stub report support for the @samp{multiprocess}
34855 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
34858 Note that all packet forms beginning with an upper- or lower-case
34859 letter, other than those described here, are reserved for future use.
34861 Here are the packet descriptions.
34866 @cindex @samp{!} packet
34867 @anchor{extended mode}
34868 Enable extended mode. In extended mode, the remote server is made
34869 persistent. The @samp{R} packet is used to restart the program being
34875 The remote target both supports and has enabled extended mode.
34879 @cindex @samp{?} packet
34881 Indicate the reason the target halted. The reply is the same as for
34882 step and continue. This packet has a special interpretation when the
34883 target is in non-stop mode; see @ref{Remote Non-Stop}.
34886 @xref{Stop Reply Packets}, for the reply specifications.
34888 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
34889 @cindex @samp{A} packet
34890 Initialized @code{argv[]} array passed into program. @var{arglen}
34891 specifies the number of bytes in the hex encoded byte stream
34892 @var{arg}. See @code{gdbserver} for more details.
34897 The arguments were set.
34903 @cindex @samp{b} packet
34904 (Don't use this packet; its behavior is not well-defined.)
34905 Change the serial line speed to @var{baud}.
34907 JTC: @emph{When does the transport layer state change? When it's
34908 received, or after the ACK is transmitted. In either case, there are
34909 problems if the command or the acknowledgment packet is dropped.}
34911 Stan: @emph{If people really wanted to add something like this, and get
34912 it working for the first time, they ought to modify ser-unix.c to send
34913 some kind of out-of-band message to a specially-setup stub and have the
34914 switch happen "in between" packets, so that from remote protocol's point
34915 of view, nothing actually happened.}
34917 @item B @var{addr},@var{mode}
34918 @cindex @samp{B} packet
34919 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
34920 breakpoint at @var{addr}.
34922 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
34923 (@pxref{insert breakpoint or watchpoint packet}).
34925 @cindex @samp{bc} packet
34928 Backward continue. Execute the target system in reverse. No parameter.
34929 @xref{Reverse Execution}, for more information.
34932 @xref{Stop Reply Packets}, for the reply specifications.
34934 @cindex @samp{bs} packet
34937 Backward single step. Execute one instruction in reverse. No parameter.
34938 @xref{Reverse Execution}, for more information.
34941 @xref{Stop Reply Packets}, for the reply specifications.
34943 @item c @r{[}@var{addr}@r{]}
34944 @cindex @samp{c} packet
34945 Continue at @var{addr}, which is the address to resume. If @var{addr}
34946 is omitted, resume at current address.
34948 This packet is deprecated for multi-threading support. @xref{vCont
34952 @xref{Stop Reply Packets}, for the reply specifications.
34954 @item C @var{sig}@r{[};@var{addr}@r{]}
34955 @cindex @samp{C} packet
34956 Continue with signal @var{sig} (hex signal number). If
34957 @samp{;@var{addr}} is omitted, resume at same address.
34959 This packet is deprecated for multi-threading support. @xref{vCont
34963 @xref{Stop Reply Packets}, for the reply specifications.
34966 @cindex @samp{d} packet
34969 Don't use this packet; instead, define a general set packet
34970 (@pxref{General Query Packets}).
34974 @cindex @samp{D} packet
34975 The first form of the packet is used to detach @value{GDBN} from the
34976 remote system. It is sent to the remote target
34977 before @value{GDBN} disconnects via the @code{detach} command.
34979 The second form, including a process ID, is used when multiprocess
34980 protocol extensions are enabled (@pxref{multiprocess extensions}), to
34981 detach only a specific process. The @var{pid} is specified as a
34982 big-endian hex string.
34992 @item F @var{RC},@var{EE},@var{CF};@var{XX}
34993 @cindex @samp{F} packet
34994 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
34995 This is part of the File-I/O protocol extension. @xref{File-I/O
34996 Remote Protocol Extension}, for the specification.
34999 @anchor{read registers packet}
35000 @cindex @samp{g} packet
35001 Read general registers.
35005 @item @var{XX@dots{}}
35006 Each byte of register data is described by two hex digits. The bytes
35007 with the register are transmitted in target byte order. The size of
35008 each register and their position within the @samp{g} packet are
35009 determined by the @value{GDBN} internal gdbarch functions
35010 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
35011 specification of several standard @samp{g} packets is specified below.
35013 When reading registers from a trace frame (@pxref{Analyze Collected
35014 Data,,Using the Collected Data}), the stub may also return a string of
35015 literal @samp{x}'s in place of the register data digits, to indicate
35016 that the corresponding register has not been collected, thus its value
35017 is unavailable. For example, for an architecture with 4 registers of
35018 4 bytes each, the following reply indicates to @value{GDBN} that
35019 registers 0 and 2 have not been collected, while registers 1 and 3
35020 have been collected, and both have zero value:
35024 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
35031 @item G @var{XX@dots{}}
35032 @cindex @samp{G} packet
35033 Write general registers. @xref{read registers packet}, for a
35034 description of the @var{XX@dots{}} data.
35044 @item H @var{op} @var{thread-id}
35045 @cindex @samp{H} packet
35046 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
35047 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
35048 should be @samp{c} for step and continue operations (note that this
35049 is deprecated, supporting the @samp{vCont} command is a better
35050 option), and @samp{g} for other operations. The thread designator
35051 @var{thread-id} has the format and interpretation described in
35052 @ref{thread-id syntax}.
35063 @c 'H': How restrictive (or permissive) is the thread model. If a
35064 @c thread is selected and stopped, are other threads allowed
35065 @c to continue to execute? As I mentioned above, I think the
35066 @c semantics of each command when a thread is selected must be
35067 @c described. For example:
35069 @c 'g': If the stub supports threads and a specific thread is
35070 @c selected, returns the register block from that thread;
35071 @c otherwise returns current registers.
35073 @c 'G' If the stub supports threads and a specific thread is
35074 @c selected, sets the registers of the register block of
35075 @c that thread; otherwise sets current registers.
35077 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
35078 @anchor{cycle step packet}
35079 @cindex @samp{i} packet
35080 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
35081 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
35082 step starting at that address.
35085 @cindex @samp{I} packet
35086 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
35090 @cindex @samp{k} packet
35093 The exact effect of this packet is not specified.
35095 For a bare-metal target, it may power cycle or reset the target
35096 system. For that reason, the @samp{k} packet has no reply.
35098 For a single-process target, it may kill that process if possible.
35100 A multiple-process target may choose to kill just one process, or all
35101 that are under @value{GDBN}'s control. For more precise control, use
35102 the vKill packet (@pxref{vKill packet}).
35104 If the target system immediately closes the connection in response to
35105 @samp{k}, @value{GDBN} does not consider the lack of packet
35106 acknowledgment to be an error, and assumes the kill was successful.
35108 If connected using @kbd{target extended-remote}, and the target does
35109 not close the connection in response to a kill request, @value{GDBN}
35110 probes the target state as if a new connection was opened
35111 (@pxref{? packet}).
35113 @item m @var{addr},@var{length}
35114 @cindex @samp{m} packet
35115 Read @var{length} addressable memory units starting at address @var{addr}
35116 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
35117 any particular boundary.
35119 The stub need not use any particular size or alignment when gathering
35120 data from memory for the response; even if @var{addr} is word-aligned
35121 and @var{length} is a multiple of the word size, the stub is free to
35122 use byte accesses, or not. For this reason, this packet may not be
35123 suitable for accessing memory-mapped I/O devices.
35124 @cindex alignment of remote memory accesses
35125 @cindex size of remote memory accesses
35126 @cindex memory, alignment and size of remote accesses
35130 @item @var{XX@dots{}}
35131 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
35132 The reply may contain fewer addressable memory units than requested if the
35133 server was able to read only part of the region of memory.
35138 @item M @var{addr},@var{length}:@var{XX@dots{}}
35139 @cindex @samp{M} packet
35140 Write @var{length} addressable memory units starting at address @var{addr}
35141 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
35142 byte is transmitted as a two-digit hexadecimal number.
35149 for an error (this includes the case where only part of the data was
35154 @cindex @samp{p} packet
35155 Read the value of register @var{n}; @var{n} is in hex.
35156 @xref{read registers packet}, for a description of how the returned
35157 register value is encoded.
35161 @item @var{XX@dots{}}
35162 the register's value
35166 Indicating an unrecognized @var{query}.
35169 @item P @var{n@dots{}}=@var{r@dots{}}
35170 @anchor{write register packet}
35171 @cindex @samp{P} packet
35172 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
35173 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
35174 digits for each byte in the register (target byte order).
35184 @item q @var{name} @var{params}@dots{}
35185 @itemx Q @var{name} @var{params}@dots{}
35186 @cindex @samp{q} packet
35187 @cindex @samp{Q} packet
35188 General query (@samp{q}) and set (@samp{Q}). These packets are
35189 described fully in @ref{General Query Packets}.
35192 @cindex @samp{r} packet
35193 Reset the entire system.
35195 Don't use this packet; use the @samp{R} packet instead.
35198 @cindex @samp{R} packet
35199 Restart the program being debugged. The @var{XX}, while needed, is ignored.
35200 This packet is only available in extended mode (@pxref{extended mode}).
35202 The @samp{R} packet has no reply.
35204 @item s @r{[}@var{addr}@r{]}
35205 @cindex @samp{s} packet
35206 Single step, resuming at @var{addr}. If
35207 @var{addr} is omitted, resume at same address.
35209 This packet is deprecated for multi-threading support. @xref{vCont
35213 @xref{Stop Reply Packets}, for the reply specifications.
35215 @item S @var{sig}@r{[};@var{addr}@r{]}
35216 @anchor{step with signal packet}
35217 @cindex @samp{S} packet
35218 Step with signal. This is analogous to the @samp{C} packet, but
35219 requests a single-step, rather than a normal resumption of execution.
35221 This packet is deprecated for multi-threading support. @xref{vCont
35225 @xref{Stop Reply Packets}, for the reply specifications.
35227 @item t @var{addr}:@var{PP},@var{MM}
35228 @cindex @samp{t} packet
35229 Search backwards starting at address @var{addr} for a match with pattern
35230 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
35231 There must be at least 3 digits in @var{addr}.
35233 @item T @var{thread-id}
35234 @cindex @samp{T} packet
35235 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
35240 thread is still alive
35246 Packets starting with @samp{v} are identified by a multi-letter name,
35247 up to the first @samp{;} or @samp{?} (or the end of the packet).
35249 @item vAttach;@var{pid}
35250 @cindex @samp{vAttach} packet
35251 Attach to a new process with the specified process ID @var{pid}.
35252 The process ID is a
35253 hexadecimal integer identifying the process. In all-stop mode, all
35254 threads in the attached process are stopped; in non-stop mode, it may be
35255 attached without being stopped if that is supported by the target.
35257 @c In non-stop mode, on a successful vAttach, the stub should set the
35258 @c current thread to a thread of the newly-attached process. After
35259 @c attaching, GDB queries for the attached process's thread ID with qC.
35260 @c Also note that, from a user perspective, whether or not the
35261 @c target is stopped on attach in non-stop mode depends on whether you
35262 @c use the foreground or background version of the attach command, not
35263 @c on what vAttach does; GDB does the right thing with respect to either
35264 @c stopping or restarting threads.
35266 This packet is only available in extended mode (@pxref{extended mode}).
35272 @item @r{Any stop packet}
35273 for success in all-stop mode (@pxref{Stop Reply Packets})
35275 for success in non-stop mode (@pxref{Remote Non-Stop})
35278 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
35279 @cindex @samp{vCont} packet
35280 @anchor{vCont packet}
35281 Resume the inferior, specifying different actions for each thread.
35282 If an action is specified with no @var{thread-id}, then it is applied to any
35283 threads that don't have a specific action specified; if no default action is
35284 specified then other threads should remain stopped in all-stop mode and
35285 in their current state in non-stop mode.
35286 Specifying multiple
35287 default actions is an error; specifying no actions is also an error.
35288 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
35290 Currently supported actions are:
35296 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
35300 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
35303 @item r @var{start},@var{end}
35304 Step once, and then keep stepping as long as the thread stops at
35305 addresses between @var{start} (inclusive) and @var{end} (exclusive).
35306 The remote stub reports a stop reply when either the thread goes out
35307 of the range or is stopped due to an unrelated reason, such as hitting
35308 a breakpoint. @xref{range stepping}.
35310 If the range is empty (@var{start} == @var{end}), then the action
35311 becomes equivalent to the @samp{s} action. In other words,
35312 single-step once, and report the stop (even if the stepped instruction
35313 jumps to @var{start}).
35315 (A stop reply may be sent at any point even if the PC is still within
35316 the stepping range; for example, it is valid to implement this packet
35317 in a degenerate way as a single instruction step operation.)
35321 The optional argument @var{addr} normally associated with the
35322 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
35323 not supported in @samp{vCont}.
35325 The @samp{t} action is only relevant in non-stop mode
35326 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
35327 A stop reply should be generated for any affected thread not already stopped.
35328 When a thread is stopped by means of a @samp{t} action,
35329 the corresponding stop reply should indicate that the thread has stopped with
35330 signal @samp{0}, regardless of whether the target uses some other signal
35331 as an implementation detail.
35333 The stub must support @samp{vCont} if it reports support for
35334 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
35335 this case @samp{vCont} actions can be specified to apply to all threads
35336 in a process by using the @samp{p@var{pid}.-1} form of the
35340 @xref{Stop Reply Packets}, for the reply specifications.
35343 @cindex @samp{vCont?} packet
35344 Request a list of actions supported by the @samp{vCont} packet.
35348 @item vCont@r{[};@var{action}@dots{}@r{]}
35349 The @samp{vCont} packet is supported. Each @var{action} is a supported
35350 command in the @samp{vCont} packet.
35352 The @samp{vCont} packet is not supported.
35355 @anchor{vCtrlC packet}
35357 @cindex @samp{vCtrlC} packet
35358 Interrupt remote target as if a control-C was pressed on the remote
35359 terminal. This is the equivalent to reacting to the @code{^C}
35360 (@samp{\003}, the control-C character) character in all-stop mode
35361 while the target is running, except this works in non-stop mode.
35362 @xref{interrupting remote targets}, for more info on the all-stop
35373 @item vFile:@var{operation}:@var{parameter}@dots{}
35374 @cindex @samp{vFile} packet
35375 Perform a file operation on the target system. For details,
35376 see @ref{Host I/O Packets}.
35378 @item vFlashErase:@var{addr},@var{length}
35379 @cindex @samp{vFlashErase} packet
35380 Direct the stub to erase @var{length} bytes of flash starting at
35381 @var{addr}. The region may enclose any number of flash blocks, but
35382 its start and end must fall on block boundaries, as indicated by the
35383 flash block size appearing in the memory map (@pxref{Memory Map
35384 Format}). @value{GDBN} groups flash memory programming operations
35385 together, and sends a @samp{vFlashDone} request after each group; the
35386 stub is allowed to delay erase operation until the @samp{vFlashDone}
35387 packet is received.
35397 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
35398 @cindex @samp{vFlashWrite} packet
35399 Direct the stub to write data to flash address @var{addr}. The data
35400 is passed in binary form using the same encoding as for the @samp{X}
35401 packet (@pxref{Binary Data}). The memory ranges specified by
35402 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
35403 not overlap, and must appear in order of increasing addresses
35404 (although @samp{vFlashErase} packets for higher addresses may already
35405 have been received; the ordering is guaranteed only between
35406 @samp{vFlashWrite} packets). If a packet writes to an address that was
35407 neither erased by a preceding @samp{vFlashErase} packet nor by some other
35408 target-specific method, the results are unpredictable.
35416 for vFlashWrite addressing non-flash memory
35422 @cindex @samp{vFlashDone} packet
35423 Indicate to the stub that flash programming operation is finished.
35424 The stub is permitted to delay or batch the effects of a group of
35425 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
35426 @samp{vFlashDone} packet is received. The contents of the affected
35427 regions of flash memory are unpredictable until the @samp{vFlashDone}
35428 request is completed.
35430 @item vKill;@var{pid}
35431 @cindex @samp{vKill} packet
35432 @anchor{vKill packet}
35433 Kill the process with the specified process ID @var{pid}, which is a
35434 hexadecimal integer identifying the process. This packet is used in
35435 preference to @samp{k} when multiprocess protocol extensions are
35436 supported; see @ref{multiprocess extensions}.
35446 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
35447 @cindex @samp{vRun} packet
35448 Run the program @var{filename}, passing it each @var{argument} on its
35449 command line. The file and arguments are hex-encoded strings. If
35450 @var{filename} is an empty string, the stub may use a default program
35451 (e.g.@: the last program run). The program is created in the stopped
35454 @c FIXME: What about non-stop mode?
35456 This packet is only available in extended mode (@pxref{extended mode}).
35462 @item @r{Any stop packet}
35463 for success (@pxref{Stop Reply Packets})
35467 @cindex @samp{vStopped} packet
35468 @xref{Notification Packets}.
35470 @item X @var{addr},@var{length}:@var{XX@dots{}}
35472 @cindex @samp{X} packet
35473 Write data to memory, where the data is transmitted in binary.
35474 Memory is specified by its address @var{addr} and number of addressable memory
35475 units @var{length} (@pxref{addressable memory unit});
35476 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
35486 @item z @var{type},@var{addr},@var{kind}
35487 @itemx Z @var{type},@var{addr},@var{kind}
35488 @anchor{insert breakpoint or watchpoint packet}
35489 @cindex @samp{z} packet
35490 @cindex @samp{Z} packets
35491 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
35492 watchpoint starting at address @var{address} of kind @var{kind}.
35494 Each breakpoint and watchpoint packet @var{type} is documented
35497 @emph{Implementation notes: A remote target shall return an empty string
35498 for an unrecognized breakpoint or watchpoint packet @var{type}. A
35499 remote target shall support either both or neither of a given
35500 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
35501 avoid potential problems with duplicate packets, the operations should
35502 be implemented in an idempotent way.}
35504 @item z0,@var{addr},@var{kind}
35505 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
35506 @cindex @samp{z0} packet
35507 @cindex @samp{Z0} packet
35508 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
35509 @var{addr} of type @var{kind}.
35511 A memory breakpoint is implemented by replacing the instruction at
35512 @var{addr} with a software breakpoint or trap instruction. The
35513 @var{kind} is target-specific and typically indicates the size of
35514 the breakpoint in bytes that should be inserted. E.g., the @sc{arm}
35515 and @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
35516 architectures have additional meanings for @var{kind};
35517 @var{cond_list} is an optional list of conditional expressions in bytecode
35518 form that should be evaluated on the target's side. These are the
35519 conditions that should be taken into consideration when deciding if
35520 the breakpoint trigger should be reported back to @var{GDBN}.
35522 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
35523 for how to best report a memory breakpoint event to @value{GDBN}.
35525 The @var{cond_list} parameter is comprised of a series of expressions,
35526 concatenated without separators. Each expression has the following form:
35530 @item X @var{len},@var{expr}
35531 @var{len} is the length of the bytecode expression and @var{expr} is the
35532 actual conditional expression in bytecode form.
35536 The optional @var{cmd_list} parameter introduces commands that may be
35537 run on the target, rather than being reported back to @value{GDBN}.
35538 The parameter starts with a numeric flag @var{persist}; if the flag is
35539 nonzero, then the breakpoint may remain active and the commands
35540 continue to be run even when @value{GDBN} disconnects from the target.
35541 Following this flag is a series of expressions concatenated with no
35542 separators. Each expression has the following form:
35546 @item X @var{len},@var{expr}
35547 @var{len} is the length of the bytecode expression and @var{expr} is the
35548 actual conditional expression in bytecode form.
35552 see @ref{Architecture-Specific Protocol Details}.
35554 @emph{Implementation note: It is possible for a target to copy or move
35555 code that contains memory breakpoints (e.g., when implementing
35556 overlays). The behavior of this packet, in the presence of such a
35557 target, is not defined.}
35569 @item z1,@var{addr},@var{kind}
35570 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}
35571 @cindex @samp{z1} packet
35572 @cindex @samp{Z1} packet
35573 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
35574 address @var{addr}.
35576 A hardware breakpoint is implemented using a mechanism that is not
35577 dependant on being able to modify the target's memory. The @var{kind}
35578 and @var{cond_list} have the same meaning as in @samp{Z0} packets.
35580 @emph{Implementation note: A hardware breakpoint is not affected by code
35593 @item z2,@var{addr},@var{kind}
35594 @itemx Z2,@var{addr},@var{kind}
35595 @cindex @samp{z2} packet
35596 @cindex @samp{Z2} packet
35597 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
35598 The number of bytes to watch is specified by @var{kind}.
35610 @item z3,@var{addr},@var{kind}
35611 @itemx Z3,@var{addr},@var{kind}
35612 @cindex @samp{z3} packet
35613 @cindex @samp{Z3} packet
35614 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
35615 The number of bytes to watch is specified by @var{kind}.
35627 @item z4,@var{addr},@var{kind}
35628 @itemx Z4,@var{addr},@var{kind}
35629 @cindex @samp{z4} packet
35630 @cindex @samp{Z4} packet
35631 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
35632 The number of bytes to watch is specified by @var{kind}.
35646 @node Stop Reply Packets
35647 @section Stop Reply Packets
35648 @cindex stop reply packets
35650 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
35651 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
35652 receive any of the below as a reply. Except for @samp{?}
35653 and @samp{vStopped}, that reply is only returned
35654 when the target halts. In the below the exact meaning of @dfn{signal
35655 number} is defined by the header @file{include/gdb/signals.h} in the
35656 @value{GDBN} source code.
35658 As in the description of request packets, we include spaces in the
35659 reply templates for clarity; these are not part of the reply packet's
35660 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
35666 The program received signal number @var{AA} (a two-digit hexadecimal
35667 number). This is equivalent to a @samp{T} response with no
35668 @var{n}:@var{r} pairs.
35670 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
35671 @cindex @samp{T} packet reply
35672 The program received signal number @var{AA} (a two-digit hexadecimal
35673 number). This is equivalent to an @samp{S} response, except that the
35674 @samp{@var{n}:@var{r}} pairs can carry values of important registers
35675 and other information directly in the stop reply packet, reducing
35676 round-trip latency. Single-step and breakpoint traps are reported
35677 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
35681 If @var{n} is a hexadecimal number, it is a register number, and the
35682 corresponding @var{r} gives that register's value. The data @var{r} is a
35683 series of bytes in target byte order, with each byte given by a
35684 two-digit hex number.
35687 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
35688 the stopped thread, as specified in @ref{thread-id syntax}.
35691 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
35692 the core on which the stop event was detected.
35695 If @var{n} is a recognized @dfn{stop reason}, it describes a more
35696 specific event that stopped the target. The currently defined stop
35697 reasons are listed below. The @var{aa} should be @samp{05}, the trap
35698 signal. At most one stop reason should be present.
35701 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
35702 and go on to the next; this allows us to extend the protocol in the
35706 The currently defined stop reasons are:
35712 The packet indicates a watchpoint hit, and @var{r} is the data address, in
35715 @item syscall_entry
35716 @itemx syscall_return
35717 The packet indicates a syscall entry or return, and @var{r} is the
35718 syscall number, in hex.
35720 @cindex shared library events, remote reply
35722 The packet indicates that the loaded libraries have changed.
35723 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
35724 list of loaded libraries. The @var{r} part is ignored.
35726 @cindex replay log events, remote reply
35728 The packet indicates that the target cannot continue replaying
35729 logged execution events, because it has reached the end (or the
35730 beginning when executing backward) of the log. The value of @var{r}
35731 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
35732 for more information.
35735 @anchor{swbreak stop reason}
35736 The packet indicates a memory breakpoint instruction was executed,
35737 irrespective of whether it was @value{GDBN} that planted the
35738 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
35739 part must be left empty.
35741 On some architectures, such as x86, at the architecture level, when a
35742 breakpoint instruction executes the program counter points at the
35743 breakpoint address plus an offset. On such targets, the stub is
35744 responsible for adjusting the PC to point back at the breakpoint
35747 This packet should not be sent by default; older @value{GDBN} versions
35748 did not support it. @value{GDBN} requests it, by supplying an
35749 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
35750 remote stub must also supply the appropriate @samp{qSupported} feature
35751 indicating support.
35753 This packet is required for correct non-stop mode operation.
35756 The packet indicates the target stopped for a hardware breakpoint.
35757 The @var{r} part must be left empty.
35759 The same remarks about @samp{qSupported} and non-stop mode above
35762 @cindex fork events, remote reply
35764 The packet indicates that @code{fork} was called, and @var{r}
35765 is the thread ID of the new child process. Refer to
35766 @ref{thread-id syntax} for the format of the @var{thread-id}
35767 field. This packet is only applicable to targets that support
35770 This packet should not be sent by default; older @value{GDBN} versions
35771 did not support it. @value{GDBN} requests it, by supplying an
35772 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
35773 remote stub must also supply the appropriate @samp{qSupported} feature
35774 indicating support.
35776 @cindex vfork events, remote reply
35778 The packet indicates that @code{vfork} was called, and @var{r}
35779 is the thread ID of the new child process. Refer to
35780 @ref{thread-id syntax} for the format of the @var{thread-id}
35781 field. This packet is only applicable to targets that support
35784 This packet should not be sent by default; older @value{GDBN} versions
35785 did not support it. @value{GDBN} requests it, by supplying an
35786 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
35787 remote stub must also supply the appropriate @samp{qSupported} feature
35788 indicating support.
35790 @cindex vforkdone events, remote reply
35792 The packet indicates that a child process created by a vfork
35793 has either called @code{exec} or terminated, so that the
35794 address spaces of the parent and child process are no longer
35795 shared. The @var{r} part is ignored. This packet is only
35796 applicable to targets that support vforkdone events.
35798 This packet should not be sent by default; older @value{GDBN} versions
35799 did not support it. @value{GDBN} requests it, by supplying an
35800 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
35801 remote stub must also supply the appropriate @samp{qSupported} feature
35802 indicating support.
35804 @cindex exec events, remote reply
35806 The packet indicates that @code{execve} was called, and @var{r}
35807 is the absolute pathname of the file that was executed, in hex.
35808 This packet is only applicable to targets that support exec events.
35810 This packet should not be sent by default; older @value{GDBN} versions
35811 did not support it. @value{GDBN} requests it, by supplying an
35812 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
35813 remote stub must also supply the appropriate @samp{qSupported} feature
35814 indicating support.
35816 @cindex thread create event, remote reply
35817 @anchor{thread create event}
35819 The packet indicates that the thread was just created. The new thread
35820 is stopped until @value{GDBN} sets it running with a resumption packet
35821 (@pxref{vCont packet}). This packet should not be sent by default;
35822 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
35823 also the @samp{w} (@ref{thread exit event}) remote reply below.
35828 @itemx W @var{AA} ; process:@var{pid}
35829 The process exited, and @var{AA} is the exit status. This is only
35830 applicable to certain targets.
35832 The second form of the response, including the process ID of the exited
35833 process, can be used only when @value{GDBN} has reported support for
35834 multiprocess protocol extensions; see @ref{multiprocess extensions}.
35835 The @var{pid} is formatted as a big-endian hex string.
35838 @itemx X @var{AA} ; process:@var{pid}
35839 The process terminated with signal @var{AA}.
35841 The second form of the response, including the process ID of the
35842 terminated process, can be used only when @value{GDBN} has reported
35843 support for multiprocess protocol extensions; see @ref{multiprocess
35844 extensions}. The @var{pid} is formatted as a big-endian hex string.
35846 @anchor{thread exit event}
35847 @cindex thread exit event, remote reply
35848 @item w @var{AA} ; @var{tid}
35850 The thread exited, and @var{AA} is the exit status. This response
35851 should not be sent by default; @value{GDBN} requests it with the
35852 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
35855 There are no resumed threads left in the target. In other words, even
35856 though the process is alive, the last resumed thread has exited. For
35857 example, say the target process has two threads: thread 1 and thread
35858 2. The client leaves thread 1 stopped, and resumes thread 2, which
35859 subsequently exits. At this point, even though the process is still
35860 alive, and thus no @samp{W} stop reply is sent, no thread is actually
35861 executing either. The @samp{N} stop reply thus informs the client
35862 that it can stop waiting for stop replies. This packet should not be
35863 sent by default; older @value{GDBN} versions did not support it.
35864 @value{GDBN} requests it, by supplying an appropriate
35865 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
35866 also supply the appropriate @samp{qSupported} feature indicating
35869 @item O @var{XX}@dots{}
35870 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
35871 written as the program's console output. This can happen at any time
35872 while the program is running and the debugger should continue to wait
35873 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
35875 @item F @var{call-id},@var{parameter}@dots{}
35876 @var{call-id} is the identifier which says which host system call should
35877 be called. This is just the name of the function. Translation into the
35878 correct system call is only applicable as it's defined in @value{GDBN}.
35879 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
35882 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
35883 this very system call.
35885 The target replies with this packet when it expects @value{GDBN} to
35886 call a host system call on behalf of the target. @value{GDBN} replies
35887 with an appropriate @samp{F} packet and keeps up waiting for the next
35888 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
35889 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
35890 Protocol Extension}, for more details.
35894 @node General Query Packets
35895 @section General Query Packets
35896 @cindex remote query requests
35898 Packets starting with @samp{q} are @dfn{general query packets};
35899 packets starting with @samp{Q} are @dfn{general set packets}. General
35900 query and set packets are a semi-unified form for retrieving and
35901 sending information to and from the stub.
35903 The initial letter of a query or set packet is followed by a name
35904 indicating what sort of thing the packet applies to. For example,
35905 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
35906 definitions with the stub. These packet names follow some
35911 The name must not contain commas, colons or semicolons.
35913 Most @value{GDBN} query and set packets have a leading upper case
35916 The names of custom vendor packets should use a company prefix, in
35917 lower case, followed by a period. For example, packets designed at
35918 the Acme Corporation might begin with @samp{qacme.foo} (for querying
35919 foos) or @samp{Qacme.bar} (for setting bars).
35922 The name of a query or set packet should be separated from any
35923 parameters by a @samp{:}; the parameters themselves should be
35924 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
35925 full packet name, and check for a separator or the end of the packet,
35926 in case two packet names share a common prefix. New packets should not begin
35927 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
35928 packets predate these conventions, and have arguments without any terminator
35929 for the packet name; we suspect they are in widespread use in places that
35930 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
35931 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
35934 Like the descriptions of the other packets, each description here
35935 has a template showing the packet's overall syntax, followed by an
35936 explanation of the packet's meaning. We include spaces in some of the
35937 templates for clarity; these are not part of the packet's syntax. No
35938 @value{GDBN} packet uses spaces to separate its components.
35940 Here are the currently defined query and set packets:
35946 Turn on or off the agent as a helper to perform some debugging operations
35947 delegated from @value{GDBN} (@pxref{Control Agent}).
35949 @item QAllow:@var{op}:@var{val}@dots{}
35950 @cindex @samp{QAllow} packet
35951 Specify which operations @value{GDBN} expects to request of the
35952 target, as a semicolon-separated list of operation name and value
35953 pairs. Possible values for @var{op} include @samp{WriteReg},
35954 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
35955 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
35956 indicating that @value{GDBN} will not request the operation, or 1,
35957 indicating that it may. (The target can then use this to set up its
35958 own internals optimally, for instance if the debugger never expects to
35959 insert breakpoints, it may not need to install its own trap handler.)
35962 @cindex current thread, remote request
35963 @cindex @samp{qC} packet
35964 Return the current thread ID.
35968 @item QC @var{thread-id}
35969 Where @var{thread-id} is a thread ID as documented in
35970 @ref{thread-id syntax}.
35971 @item @r{(anything else)}
35972 Any other reply implies the old thread ID.
35975 @item qCRC:@var{addr},@var{length}
35976 @cindex CRC of memory block, remote request
35977 @cindex @samp{qCRC} packet
35978 @anchor{qCRC packet}
35979 Compute the CRC checksum of a block of memory using CRC-32 defined in
35980 IEEE 802.3. The CRC is computed byte at a time, taking the most
35981 significant bit of each byte first. The initial pattern code
35982 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
35984 @emph{Note:} This is the same CRC used in validating separate debug
35985 files (@pxref{Separate Debug Files, , Debugging Information in Separate
35986 Files}). However the algorithm is slightly different. When validating
35987 separate debug files, the CRC is computed taking the @emph{least}
35988 significant bit of each byte first, and the final result is inverted to
35989 detect trailing zeros.
35994 An error (such as memory fault)
35995 @item C @var{crc32}
35996 The specified memory region's checksum is @var{crc32}.
35999 @item QDisableRandomization:@var{value}
36000 @cindex disable address space randomization, remote request
36001 @cindex @samp{QDisableRandomization} packet
36002 Some target operating systems will randomize the virtual address space
36003 of the inferior process as a security feature, but provide a feature
36004 to disable such randomization, e.g.@: to allow for a more deterministic
36005 debugging experience. On such systems, this packet with a @var{value}
36006 of 1 directs the target to disable address space randomization for
36007 processes subsequently started via @samp{vRun} packets, while a packet
36008 with a @var{value} of 0 tells the target to enable address space
36011 This packet is only available in extended mode (@pxref{extended mode}).
36016 The request succeeded.
36019 An error occurred. The error number @var{nn} is given as hex digits.
36022 An empty reply indicates that @samp{QDisableRandomization} is not supported
36026 This packet is not probed by default; the remote stub must request it,
36027 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36028 This should only be done on targets that actually support disabling
36029 address space randomization.
36032 @itemx qsThreadInfo
36033 @cindex list active threads, remote request
36034 @cindex @samp{qfThreadInfo} packet
36035 @cindex @samp{qsThreadInfo} packet
36036 Obtain a list of all active thread IDs from the target (OS). Since there
36037 may be too many active threads to fit into one reply packet, this query
36038 works iteratively: it may require more than one query/reply sequence to
36039 obtain the entire list of threads. The first query of the sequence will
36040 be the @samp{qfThreadInfo} query; subsequent queries in the
36041 sequence will be the @samp{qsThreadInfo} query.
36043 NOTE: This packet replaces the @samp{qL} query (see below).
36047 @item m @var{thread-id}
36049 @item m @var{thread-id},@var{thread-id}@dots{}
36050 a comma-separated list of thread IDs
36052 (lower case letter @samp{L}) denotes end of list.
36055 In response to each query, the target will reply with a list of one or
36056 more thread IDs, separated by commas.
36057 @value{GDBN} will respond to each reply with a request for more thread
36058 ids (using the @samp{qs} form of the query), until the target responds
36059 with @samp{l} (lower-case ell, for @dfn{last}).
36060 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
36063 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
36064 initial connection with the remote target, and the very first thread ID
36065 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
36066 message. Therefore, the stub should ensure that the first thread ID in
36067 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
36069 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
36070 @cindex get thread-local storage address, remote request
36071 @cindex @samp{qGetTLSAddr} packet
36072 Fetch the address associated with thread local storage specified
36073 by @var{thread-id}, @var{offset}, and @var{lm}.
36075 @var{thread-id} is the thread ID associated with the
36076 thread for which to fetch the TLS address. @xref{thread-id syntax}.
36078 @var{offset} is the (big endian, hex encoded) offset associated with the
36079 thread local variable. (This offset is obtained from the debug
36080 information associated with the variable.)
36082 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
36083 load module associated with the thread local storage. For example,
36084 a @sc{gnu}/Linux system will pass the link map address of the shared
36085 object associated with the thread local storage under consideration.
36086 Other operating environments may choose to represent the load module
36087 differently, so the precise meaning of this parameter will vary.
36091 @item @var{XX}@dots{}
36092 Hex encoded (big endian) bytes representing the address of the thread
36093 local storage requested.
36096 An error occurred. The error number @var{nn} is given as hex digits.
36099 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
36102 @item qGetTIBAddr:@var{thread-id}
36103 @cindex get thread information block address
36104 @cindex @samp{qGetTIBAddr} packet
36105 Fetch address of the Windows OS specific Thread Information Block.
36107 @var{thread-id} is the thread ID associated with the thread.
36111 @item @var{XX}@dots{}
36112 Hex encoded (big endian) bytes representing the linear address of the
36113 thread information block.
36116 An error occured. This means that either the thread was not found, or the
36117 address could not be retrieved.
36120 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
36123 @item qL @var{startflag} @var{threadcount} @var{nextthread}
36124 Obtain thread information from RTOS. Where: @var{startflag} (one hex
36125 digit) is one to indicate the first query and zero to indicate a
36126 subsequent query; @var{threadcount} (two hex digits) is the maximum
36127 number of threads the response packet can contain; and @var{nextthread}
36128 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
36129 returned in the response as @var{argthread}.
36131 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
36135 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
36136 Where: @var{count} (two hex digits) is the number of threads being
36137 returned; @var{done} (one hex digit) is zero to indicate more threads
36138 and one indicates no further threads; @var{argthreadid} (eight hex
36139 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
36140 is a sequence of thread IDs, @var{threadid} (eight hex
36141 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
36145 @cindex section offsets, remote request
36146 @cindex @samp{qOffsets} packet
36147 Get section offsets that the target used when relocating the downloaded
36152 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
36153 Relocate the @code{Text} section by @var{xxx} from its original address.
36154 Relocate the @code{Data} section by @var{yyy} from its original address.
36155 If the object file format provides segment information (e.g.@: @sc{elf}
36156 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
36157 segments by the supplied offsets.
36159 @emph{Note: while a @code{Bss} offset may be included in the response,
36160 @value{GDBN} ignores this and instead applies the @code{Data} offset
36161 to the @code{Bss} section.}
36163 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
36164 Relocate the first segment of the object file, which conventionally
36165 contains program code, to a starting address of @var{xxx}. If
36166 @samp{DataSeg} is specified, relocate the second segment, which
36167 conventionally contains modifiable data, to a starting address of
36168 @var{yyy}. @value{GDBN} will report an error if the object file
36169 does not contain segment information, or does not contain at least
36170 as many segments as mentioned in the reply. Extra segments are
36171 kept at fixed offsets relative to the last relocated segment.
36174 @item qP @var{mode} @var{thread-id}
36175 @cindex thread information, remote request
36176 @cindex @samp{qP} packet
36177 Returns information on @var{thread-id}. Where: @var{mode} is a hex
36178 encoded 32 bit mode; @var{thread-id} is a thread ID
36179 (@pxref{thread-id syntax}).
36181 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
36184 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
36188 @cindex non-stop mode, remote request
36189 @cindex @samp{QNonStop} packet
36191 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
36192 @xref{Remote Non-Stop}, for more information.
36197 The request succeeded.
36200 An error occurred. The error number @var{nn} is given as hex digits.
36203 An empty reply indicates that @samp{QNonStop} is not supported by
36207 This packet is not probed by default; the remote stub must request it,
36208 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36209 Use of this packet is controlled by the @code{set non-stop} command;
36210 @pxref{Non-Stop Mode}.
36212 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
36213 @itemx QCatchSyscalls:0
36214 @cindex catch syscalls from inferior, remote request
36215 @cindex @samp{QCatchSyscalls} packet
36216 @anchor{QCatchSyscalls}
36217 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
36218 catching syscalls from the inferior process.
36220 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
36221 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
36222 is listed, every system call should be reported.
36224 Note that if a syscall not in the list is reported, @value{GDBN} will
36225 still filter the event according to its own list from all corresponding
36226 @code{catch syscall} commands. However, it is more efficient to only
36227 report the requested syscalls.
36229 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
36230 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
36232 If the inferior process execs, the state of @samp{QCatchSyscalls} is
36233 kept for the new process too. On targets where exec may affect syscall
36234 numbers, for example with exec between 32 and 64-bit processes, the
36235 client should send a new packet with the new syscall list.
36240 The request succeeded.
36243 An error occurred. @var{nn} are hex digits.
36246 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
36250 Use of this packet is controlled by the @code{set remote catch-syscalls}
36251 command (@pxref{Remote Configuration, set remote catch-syscalls}).
36252 This packet is not probed by default; the remote stub must request it,
36253 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36255 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
36256 @cindex pass signals to inferior, remote request
36257 @cindex @samp{QPassSignals} packet
36258 @anchor{QPassSignals}
36259 Each listed @var{signal} should be passed directly to the inferior process.
36260 Signals are numbered identically to continue packets and stop replies
36261 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
36262 strictly greater than the previous item. These signals do not need to stop
36263 the inferior, or be reported to @value{GDBN}. All other signals should be
36264 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
36265 combine; any earlier @samp{QPassSignals} list is completely replaced by the
36266 new list. This packet improves performance when using @samp{handle
36267 @var{signal} nostop noprint pass}.
36272 The request succeeded.
36275 An error occurred. The error number @var{nn} is given as hex digits.
36278 An empty reply indicates that @samp{QPassSignals} is not supported by
36282 Use of this packet is controlled by the @code{set remote pass-signals}
36283 command (@pxref{Remote Configuration, set remote pass-signals}).
36284 This packet is not probed by default; the remote stub must request it,
36285 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36287 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
36288 @cindex signals the inferior may see, remote request
36289 @cindex @samp{QProgramSignals} packet
36290 @anchor{QProgramSignals}
36291 Each listed @var{signal} may be delivered to the inferior process.
36292 Others should be silently discarded.
36294 In some cases, the remote stub may need to decide whether to deliver a
36295 signal to the program or not without @value{GDBN} involvement. One
36296 example of that is while detaching --- the program's threads may have
36297 stopped for signals that haven't yet had a chance of being reported to
36298 @value{GDBN}, and so the remote stub can use the signal list specified
36299 by this packet to know whether to deliver or ignore those pending
36302 This does not influence whether to deliver a signal as requested by a
36303 resumption packet (@pxref{vCont packet}).
36305 Signals are numbered identically to continue packets and stop replies
36306 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
36307 strictly greater than the previous item. Multiple
36308 @samp{QProgramSignals} packets do not combine; any earlier
36309 @samp{QProgramSignals} list is completely replaced by the new list.
36314 The request succeeded.
36317 An error occurred. The error number @var{nn} is given as hex digits.
36320 An empty reply indicates that @samp{QProgramSignals} is not supported
36324 Use of this packet is controlled by the @code{set remote program-signals}
36325 command (@pxref{Remote Configuration, set remote program-signals}).
36326 This packet is not probed by default; the remote stub must request it,
36327 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36329 @anchor{QThreadEvents}
36330 @item QThreadEvents:1
36331 @itemx QThreadEvents:0
36332 @cindex thread create/exit events, remote request
36333 @cindex @samp{QThreadEvents} packet
36335 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
36336 reporting of thread create and exit events. @xref{thread create
36337 event}, for the reply specifications. For example, this is used in
36338 non-stop mode when @value{GDBN} stops a set of threads and
36339 synchronously waits for the their corresponding stop replies. Without
36340 exit events, if one of the threads exits, @value{GDBN} would hang
36341 forever not knowing that it should no longer expect a stop for that
36342 same thread. @value{GDBN} does not enable this feature unless the
36343 stub reports that it supports it by including @samp{QThreadEvents+} in
36344 its @samp{qSupported} reply.
36349 The request succeeded.
36352 An error occurred. The error number @var{nn} is given as hex digits.
36355 An empty reply indicates that @samp{QThreadEvents} is not supported by
36359 Use of this packet is controlled by the @code{set remote thread-events}
36360 command (@pxref{Remote Configuration, set remote thread-events}).
36362 @item qRcmd,@var{command}
36363 @cindex execute remote command, remote request
36364 @cindex @samp{qRcmd} packet
36365 @var{command} (hex encoded) is passed to the local interpreter for
36366 execution. Invalid commands should be reported using the output
36367 string. Before the final result packet, the target may also respond
36368 with a number of intermediate @samp{O@var{output}} console output
36369 packets. @emph{Implementors should note that providing access to a
36370 stubs's interpreter may have security implications}.
36375 A command response with no output.
36377 A command response with the hex encoded output string @var{OUTPUT}.
36379 Indicate a badly formed request.
36381 An empty reply indicates that @samp{qRcmd} is not recognized.
36384 (Note that the @code{qRcmd} packet's name is separated from the
36385 command by a @samp{,}, not a @samp{:}, contrary to the naming
36386 conventions above. Please don't use this packet as a model for new
36389 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
36390 @cindex searching memory, in remote debugging
36392 @cindex @samp{qSearch:memory} packet
36394 @cindex @samp{qSearch memory} packet
36395 @anchor{qSearch memory}
36396 Search @var{length} bytes at @var{address} for @var{search-pattern}.
36397 Both @var{address} and @var{length} are encoded in hex;
36398 @var{search-pattern} is a sequence of bytes, also hex encoded.
36403 The pattern was not found.
36405 The pattern was found at @var{address}.
36407 A badly formed request or an error was encountered while searching memory.
36409 An empty reply indicates that @samp{qSearch:memory} is not recognized.
36412 @item QStartNoAckMode
36413 @cindex @samp{QStartNoAckMode} packet
36414 @anchor{QStartNoAckMode}
36415 Request that the remote stub disable the normal @samp{+}/@samp{-}
36416 protocol acknowledgments (@pxref{Packet Acknowledgment}).
36421 The stub has switched to no-acknowledgment mode.
36422 @value{GDBN} acknowledges this reponse,
36423 but neither the stub nor @value{GDBN} shall send or expect further
36424 @samp{+}/@samp{-} acknowledgments in the current connection.
36426 An empty reply indicates that the stub does not support no-acknowledgment mode.
36429 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
36430 @cindex supported packets, remote query
36431 @cindex features of the remote protocol
36432 @cindex @samp{qSupported} packet
36433 @anchor{qSupported}
36434 Tell the remote stub about features supported by @value{GDBN}, and
36435 query the stub for features it supports. This packet allows
36436 @value{GDBN} and the remote stub to take advantage of each others'
36437 features. @samp{qSupported} also consolidates multiple feature probes
36438 at startup, to improve @value{GDBN} performance---a single larger
36439 packet performs better than multiple smaller probe packets on
36440 high-latency links. Some features may enable behavior which must not
36441 be on by default, e.g.@: because it would confuse older clients or
36442 stubs. Other features may describe packets which could be
36443 automatically probed for, but are not. These features must be
36444 reported before @value{GDBN} will use them. This ``default
36445 unsupported'' behavior is not appropriate for all packets, but it
36446 helps to keep the initial connection time under control with new
36447 versions of @value{GDBN} which support increasing numbers of packets.
36451 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
36452 The stub supports or does not support each returned @var{stubfeature},
36453 depending on the form of each @var{stubfeature} (see below for the
36456 An empty reply indicates that @samp{qSupported} is not recognized,
36457 or that no features needed to be reported to @value{GDBN}.
36460 The allowed forms for each feature (either a @var{gdbfeature} in the
36461 @samp{qSupported} packet, or a @var{stubfeature} in the response)
36465 @item @var{name}=@var{value}
36466 The remote protocol feature @var{name} is supported, and associated
36467 with the specified @var{value}. The format of @var{value} depends
36468 on the feature, but it must not include a semicolon.
36470 The remote protocol feature @var{name} is supported, and does not
36471 need an associated value.
36473 The remote protocol feature @var{name} is not supported.
36475 The remote protocol feature @var{name} may be supported, and
36476 @value{GDBN} should auto-detect support in some other way when it is
36477 needed. This form will not be used for @var{gdbfeature} notifications,
36478 but may be used for @var{stubfeature} responses.
36481 Whenever the stub receives a @samp{qSupported} request, the
36482 supplied set of @value{GDBN} features should override any previous
36483 request. This allows @value{GDBN} to put the stub in a known
36484 state, even if the stub had previously been communicating with
36485 a different version of @value{GDBN}.
36487 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
36492 This feature indicates whether @value{GDBN} supports multiprocess
36493 extensions to the remote protocol. @value{GDBN} does not use such
36494 extensions unless the stub also reports that it supports them by
36495 including @samp{multiprocess+} in its @samp{qSupported} reply.
36496 @xref{multiprocess extensions}, for details.
36499 This feature indicates that @value{GDBN} supports the XML target
36500 description. If the stub sees @samp{xmlRegisters=} with target
36501 specific strings separated by a comma, it will report register
36505 This feature indicates whether @value{GDBN} supports the
36506 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
36507 instruction reply packet}).
36510 This feature indicates whether @value{GDBN} supports the swbreak stop
36511 reason in stop replies. @xref{swbreak stop reason}, for details.
36514 This feature indicates whether @value{GDBN} supports the hwbreak stop
36515 reason in stop replies. @xref{swbreak stop reason}, for details.
36518 This feature indicates whether @value{GDBN} supports fork event
36519 extensions to the remote protocol. @value{GDBN} does not use such
36520 extensions unless the stub also reports that it supports them by
36521 including @samp{fork-events+} in its @samp{qSupported} reply.
36524 This feature indicates whether @value{GDBN} supports vfork event
36525 extensions to the remote protocol. @value{GDBN} does not use such
36526 extensions unless the stub also reports that it supports them by
36527 including @samp{vfork-events+} in its @samp{qSupported} reply.
36530 This feature indicates whether @value{GDBN} supports exec event
36531 extensions to the remote protocol. @value{GDBN} does not use such
36532 extensions unless the stub also reports that it supports them by
36533 including @samp{exec-events+} in its @samp{qSupported} reply.
36535 @item vContSupported
36536 This feature indicates whether @value{GDBN} wants to know the
36537 supported actions in the reply to @samp{vCont?} packet.
36540 Stubs should ignore any unknown values for
36541 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
36542 packet supports receiving packets of unlimited length (earlier
36543 versions of @value{GDBN} may reject overly long responses). Additional values
36544 for @var{gdbfeature} may be defined in the future to let the stub take
36545 advantage of new features in @value{GDBN}, e.g.@: incompatible
36546 improvements in the remote protocol---the @samp{multiprocess} feature is
36547 an example of such a feature. The stub's reply should be independent
36548 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
36549 describes all the features it supports, and then the stub replies with
36550 all the features it supports.
36552 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
36553 responses, as long as each response uses one of the standard forms.
36555 Some features are flags. A stub which supports a flag feature
36556 should respond with a @samp{+} form response. Other features
36557 require values, and the stub should respond with an @samp{=}
36560 Each feature has a default value, which @value{GDBN} will use if
36561 @samp{qSupported} is not available or if the feature is not mentioned
36562 in the @samp{qSupported} response. The default values are fixed; a
36563 stub is free to omit any feature responses that match the defaults.
36565 Not all features can be probed, but for those which can, the probing
36566 mechanism is useful: in some cases, a stub's internal
36567 architecture may not allow the protocol layer to know some information
36568 about the underlying target in advance. This is especially common in
36569 stubs which may be configured for multiple targets.
36571 These are the currently defined stub features and their properties:
36573 @multitable @columnfractions 0.35 0.2 0.12 0.2
36574 @c NOTE: The first row should be @headitem, but we do not yet require
36575 @c a new enough version of Texinfo (4.7) to use @headitem.
36577 @tab Value Required
36581 @item @samp{PacketSize}
36586 @item @samp{qXfer:auxv:read}
36591 @item @samp{qXfer:btrace:read}
36596 @item @samp{qXfer:btrace-conf:read}
36601 @item @samp{qXfer:exec-file:read}
36606 @item @samp{qXfer:features:read}
36611 @item @samp{qXfer:libraries:read}
36616 @item @samp{qXfer:libraries-svr4:read}
36621 @item @samp{augmented-libraries-svr4-read}
36626 @item @samp{qXfer:memory-map:read}
36631 @item @samp{qXfer:sdata:read}
36636 @item @samp{qXfer:spu:read}
36641 @item @samp{qXfer:spu:write}
36646 @item @samp{qXfer:siginfo:read}
36651 @item @samp{qXfer:siginfo:write}
36656 @item @samp{qXfer:threads:read}
36661 @item @samp{qXfer:traceframe-info:read}
36666 @item @samp{qXfer:uib:read}
36671 @item @samp{qXfer:fdpic:read}
36676 @item @samp{Qbtrace:off}
36681 @item @samp{Qbtrace:bts}
36686 @item @samp{Qbtrace:pt}
36691 @item @samp{Qbtrace-conf:bts:size}
36696 @item @samp{Qbtrace-conf:pt:size}
36701 @item @samp{QNonStop}
36706 @item @samp{QCatchSyscalls}
36711 @item @samp{QPassSignals}
36716 @item @samp{QStartNoAckMode}
36721 @item @samp{multiprocess}
36726 @item @samp{ConditionalBreakpoints}
36731 @item @samp{ConditionalTracepoints}
36736 @item @samp{ReverseContinue}
36741 @item @samp{ReverseStep}
36746 @item @samp{TracepointSource}
36751 @item @samp{QAgent}
36756 @item @samp{QAllow}
36761 @item @samp{QDisableRandomization}
36766 @item @samp{EnableDisableTracepoints}
36771 @item @samp{QTBuffer:size}
36776 @item @samp{tracenz}
36781 @item @samp{BreakpointCommands}
36786 @item @samp{swbreak}
36791 @item @samp{hwbreak}
36796 @item @samp{fork-events}
36801 @item @samp{vfork-events}
36806 @item @samp{exec-events}
36811 @item @samp{QThreadEvents}
36816 @item @samp{no-resumed}
36823 These are the currently defined stub features, in more detail:
36826 @cindex packet size, remote protocol
36827 @item PacketSize=@var{bytes}
36828 The remote stub can accept packets up to at least @var{bytes} in
36829 length. @value{GDBN} will send packets up to this size for bulk
36830 transfers, and will never send larger packets. This is a limit on the
36831 data characters in the packet, including the frame and checksum.
36832 There is no trailing NUL byte in a remote protocol packet; if the stub
36833 stores packets in a NUL-terminated format, it should allow an extra
36834 byte in its buffer for the NUL. If this stub feature is not supported,
36835 @value{GDBN} guesses based on the size of the @samp{g} packet response.
36837 @item qXfer:auxv:read
36838 The remote stub understands the @samp{qXfer:auxv:read} packet
36839 (@pxref{qXfer auxiliary vector read}).
36841 @item qXfer:btrace:read
36842 The remote stub understands the @samp{qXfer:btrace:read}
36843 packet (@pxref{qXfer btrace read}).
36845 @item qXfer:btrace-conf:read
36846 The remote stub understands the @samp{qXfer:btrace-conf:read}
36847 packet (@pxref{qXfer btrace-conf read}).
36849 @item qXfer:exec-file:read
36850 The remote stub understands the @samp{qXfer:exec-file:read} packet
36851 (@pxref{qXfer executable filename read}).
36853 @item qXfer:features:read
36854 The remote stub understands the @samp{qXfer:features:read} packet
36855 (@pxref{qXfer target description read}).
36857 @item qXfer:libraries:read
36858 The remote stub understands the @samp{qXfer:libraries:read} packet
36859 (@pxref{qXfer library list read}).
36861 @item qXfer:libraries-svr4:read
36862 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
36863 (@pxref{qXfer svr4 library list read}).
36865 @item augmented-libraries-svr4-read
36866 The remote stub understands the augmented form of the
36867 @samp{qXfer:libraries-svr4:read} packet
36868 (@pxref{qXfer svr4 library list read}).
36870 @item qXfer:memory-map:read
36871 The remote stub understands the @samp{qXfer:memory-map:read} packet
36872 (@pxref{qXfer memory map read}).
36874 @item qXfer:sdata:read
36875 The remote stub understands the @samp{qXfer:sdata:read} packet
36876 (@pxref{qXfer sdata read}).
36878 @item qXfer:spu:read
36879 The remote stub understands the @samp{qXfer:spu:read} packet
36880 (@pxref{qXfer spu read}).
36882 @item qXfer:spu:write
36883 The remote stub understands the @samp{qXfer:spu:write} packet
36884 (@pxref{qXfer spu write}).
36886 @item qXfer:siginfo:read
36887 The remote stub understands the @samp{qXfer:siginfo:read} packet
36888 (@pxref{qXfer siginfo read}).
36890 @item qXfer:siginfo:write
36891 The remote stub understands the @samp{qXfer:siginfo:write} packet
36892 (@pxref{qXfer siginfo write}).
36894 @item qXfer:threads:read
36895 The remote stub understands the @samp{qXfer:threads:read} packet
36896 (@pxref{qXfer threads read}).
36898 @item qXfer:traceframe-info:read
36899 The remote stub understands the @samp{qXfer:traceframe-info:read}
36900 packet (@pxref{qXfer traceframe info read}).
36902 @item qXfer:uib:read
36903 The remote stub understands the @samp{qXfer:uib:read}
36904 packet (@pxref{qXfer unwind info block}).
36906 @item qXfer:fdpic:read
36907 The remote stub understands the @samp{qXfer:fdpic:read}
36908 packet (@pxref{qXfer fdpic loadmap read}).
36911 The remote stub understands the @samp{QNonStop} packet
36912 (@pxref{QNonStop}).
36914 @item QCatchSyscalls
36915 The remote stub understands the @samp{QCatchSyscalls} packet
36916 (@pxref{QCatchSyscalls}).
36919 The remote stub understands the @samp{QPassSignals} packet
36920 (@pxref{QPassSignals}).
36922 @item QStartNoAckMode
36923 The remote stub understands the @samp{QStartNoAckMode} packet and
36924 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
36927 @anchor{multiprocess extensions}
36928 @cindex multiprocess extensions, in remote protocol
36929 The remote stub understands the multiprocess extensions to the remote
36930 protocol syntax. The multiprocess extensions affect the syntax of
36931 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
36932 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
36933 replies. Note that reporting this feature indicates support for the
36934 syntactic extensions only, not that the stub necessarily supports
36935 debugging of more than one process at a time. The stub must not use
36936 multiprocess extensions in packet replies unless @value{GDBN} has also
36937 indicated it supports them in its @samp{qSupported} request.
36939 @item qXfer:osdata:read
36940 The remote stub understands the @samp{qXfer:osdata:read} packet
36941 ((@pxref{qXfer osdata read}).
36943 @item ConditionalBreakpoints
36944 The target accepts and implements evaluation of conditional expressions
36945 defined for breakpoints. The target will only report breakpoint triggers
36946 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
36948 @item ConditionalTracepoints
36949 The remote stub accepts and implements conditional expressions defined
36950 for tracepoints (@pxref{Tracepoint Conditions}).
36952 @item ReverseContinue
36953 The remote stub accepts and implements the reverse continue packet
36957 The remote stub accepts and implements the reverse step packet
36960 @item TracepointSource
36961 The remote stub understands the @samp{QTDPsrc} packet that supplies
36962 the source form of tracepoint definitions.
36965 The remote stub understands the @samp{QAgent} packet.
36968 The remote stub understands the @samp{QAllow} packet.
36970 @item QDisableRandomization
36971 The remote stub understands the @samp{QDisableRandomization} packet.
36973 @item StaticTracepoint
36974 @cindex static tracepoints, in remote protocol
36975 The remote stub supports static tracepoints.
36977 @item InstallInTrace
36978 @anchor{install tracepoint in tracing}
36979 The remote stub supports installing tracepoint in tracing.
36981 @item EnableDisableTracepoints
36982 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
36983 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
36984 to be enabled and disabled while a trace experiment is running.
36986 @item QTBuffer:size
36987 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
36988 packet that allows to change the size of the trace buffer.
36991 @cindex string tracing, in remote protocol
36992 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
36993 See @ref{Bytecode Descriptions} for details about the bytecode.
36995 @item BreakpointCommands
36996 @cindex breakpoint commands, in remote protocol
36997 The remote stub supports running a breakpoint's command list itself,
36998 rather than reporting the hit to @value{GDBN}.
37001 The remote stub understands the @samp{Qbtrace:off} packet.
37004 The remote stub understands the @samp{Qbtrace:bts} packet.
37007 The remote stub understands the @samp{Qbtrace:pt} packet.
37009 @item Qbtrace-conf:bts:size
37010 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
37012 @item Qbtrace-conf:pt:size
37013 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
37016 The remote stub reports the @samp{swbreak} stop reason for memory
37020 The remote stub reports the @samp{hwbreak} stop reason for hardware
37024 The remote stub reports the @samp{fork} stop reason for fork events.
37027 The remote stub reports the @samp{vfork} stop reason for vfork events
37028 and vforkdone events.
37031 The remote stub reports the @samp{exec} stop reason for exec events.
37033 @item vContSupported
37034 The remote stub reports the supported actions in the reply to
37035 @samp{vCont?} packet.
37037 @item QThreadEvents
37038 The remote stub understands the @samp{QThreadEvents} packet.
37041 The remote stub reports the @samp{N} stop reply.
37046 @cindex symbol lookup, remote request
37047 @cindex @samp{qSymbol} packet
37048 Notify the target that @value{GDBN} is prepared to serve symbol lookup
37049 requests. Accept requests from the target for the values of symbols.
37054 The target does not need to look up any (more) symbols.
37055 @item qSymbol:@var{sym_name}
37056 The target requests the value of symbol @var{sym_name} (hex encoded).
37057 @value{GDBN} may provide the value by using the
37058 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
37062 @item qSymbol:@var{sym_value}:@var{sym_name}
37063 Set the value of @var{sym_name} to @var{sym_value}.
37065 @var{sym_name} (hex encoded) is the name of a symbol whose value the
37066 target has previously requested.
37068 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
37069 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
37075 The target does not need to look up any (more) symbols.
37076 @item qSymbol:@var{sym_name}
37077 The target requests the value of a new symbol @var{sym_name} (hex
37078 encoded). @value{GDBN} will continue to supply the values of symbols
37079 (if available), until the target ceases to request them.
37084 @itemx QTDisconnected
37091 @itemx qTMinFTPILen
37093 @xref{Tracepoint Packets}.
37095 @item qThreadExtraInfo,@var{thread-id}
37096 @cindex thread attributes info, remote request
37097 @cindex @samp{qThreadExtraInfo} packet
37098 Obtain from the target OS a printable string description of thread
37099 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
37100 for the forms of @var{thread-id}. This
37101 string may contain anything that the target OS thinks is interesting
37102 for @value{GDBN} to tell the user about the thread. The string is
37103 displayed in @value{GDBN}'s @code{info threads} display. Some
37104 examples of possible thread extra info strings are @samp{Runnable}, or
37105 @samp{Blocked on Mutex}.
37109 @item @var{XX}@dots{}
37110 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
37111 comprising the printable string containing the extra information about
37112 the thread's attributes.
37115 (Note that the @code{qThreadExtraInfo} packet's name is separated from
37116 the command by a @samp{,}, not a @samp{:}, contrary to the naming
37117 conventions above. Please don't use this packet as a model for new
37136 @xref{Tracepoint Packets}.
37138 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
37139 @cindex read special object, remote request
37140 @cindex @samp{qXfer} packet
37141 @anchor{qXfer read}
37142 Read uninterpreted bytes from the target's special data area
37143 identified by the keyword @var{object}. Request @var{length} bytes
37144 starting at @var{offset} bytes into the data. The content and
37145 encoding of @var{annex} is specific to @var{object}; it can supply
37146 additional details about what data to access.
37148 Here are the specific requests of this form defined so far. All
37149 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
37150 formats, listed below.
37153 @item qXfer:auxv:read::@var{offset},@var{length}
37154 @anchor{qXfer auxiliary vector read}
37155 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
37156 auxiliary vector}. Note @var{annex} must be empty.
37158 This packet is not probed by default; the remote stub must request it,
37159 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37161 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
37162 @anchor{qXfer btrace read}
37164 Return a description of the current branch trace.
37165 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
37166 packet may have one of the following values:
37170 Returns all available branch trace.
37173 Returns all available branch trace if the branch trace changed since
37174 the last read request.
37177 Returns the new branch trace since the last read request. Adds a new
37178 block to the end of the trace that begins at zero and ends at the source
37179 location of the first branch in the trace buffer. This extra block is
37180 used to stitch traces together.
37182 If the trace buffer overflowed, returns an error indicating the overflow.
37185 This packet is not probed by default; the remote stub must request it
37186 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37188 @item qXfer:btrace-conf:read::@var{offset},@var{length}
37189 @anchor{qXfer btrace-conf read}
37191 Return a description of the current branch trace configuration.
37192 @xref{Branch Trace Configuration Format}.
37194 This packet is not probed by default; the remote stub must request it
37195 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37197 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
37198 @anchor{qXfer executable filename read}
37199 Return the full absolute name of the file that was executed to create
37200 a process running on the remote system. The annex specifies the
37201 numeric process ID of the process to query, encoded as a hexadecimal
37202 number. If the annex part is empty the remote stub should return the
37203 filename corresponding to the currently executing process.
37205 This packet is not probed by default; the remote stub must request it,
37206 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37208 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
37209 @anchor{qXfer target description read}
37210 Access the @dfn{target description}. @xref{Target Descriptions}. The
37211 annex specifies which XML document to access. The main description is
37212 always loaded from the @samp{target.xml} annex.
37214 This packet is not probed by default; the remote stub must request it,
37215 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37217 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
37218 @anchor{qXfer library list read}
37219 Access the target's list of loaded libraries. @xref{Library List Format}.
37220 The annex part of the generic @samp{qXfer} packet must be empty
37221 (@pxref{qXfer read}).
37223 Targets which maintain a list of libraries in the program's memory do
37224 not need to implement this packet; it is designed for platforms where
37225 the operating system manages the list of loaded libraries.
37227 This packet is not probed by default; the remote stub must request it,
37228 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37230 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
37231 @anchor{qXfer svr4 library list read}
37232 Access the target's list of loaded libraries when the target is an SVR4
37233 platform. @xref{Library List Format for SVR4 Targets}. The annex part
37234 of the generic @samp{qXfer} packet must be empty unless the remote
37235 stub indicated it supports the augmented form of this packet
37236 by supplying an appropriate @samp{qSupported} response
37237 (@pxref{qXfer read}, @ref{qSupported}).
37239 This packet is optional for better performance on SVR4 targets.
37240 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
37242 This packet is not probed by default; the remote stub must request it,
37243 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37245 If the remote stub indicates it supports the augmented form of this
37246 packet then the annex part of the generic @samp{qXfer} packet may
37247 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
37248 arguments. The currently supported arguments are:
37251 @item start=@var{address}
37252 A hexadecimal number specifying the address of the @samp{struct
37253 link_map} to start reading the library list from. If unset or zero
37254 then the first @samp{struct link_map} in the library list will be
37255 chosen as the starting point.
37257 @item prev=@var{address}
37258 A hexadecimal number specifying the address of the @samp{struct
37259 link_map} immediately preceding the @samp{struct link_map}
37260 specified by the @samp{start} argument. If unset or zero then
37261 the remote stub will expect that no @samp{struct link_map}
37262 exists prior to the starting point.
37266 Arguments that are not understood by the remote stub will be silently
37269 @item qXfer:memory-map:read::@var{offset},@var{length}
37270 @anchor{qXfer memory map read}
37271 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
37272 annex part of the generic @samp{qXfer} packet must be empty
37273 (@pxref{qXfer read}).
37275 This packet is not probed by default; the remote stub must request it,
37276 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37278 @item qXfer:sdata:read::@var{offset},@var{length}
37279 @anchor{qXfer sdata read}
37281 Read contents of the extra collected static tracepoint marker
37282 information. The annex part of the generic @samp{qXfer} packet must
37283 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
37286 This packet is not probed by default; the remote stub must request it,
37287 by supplying an appropriate @samp{qSupported} response
37288 (@pxref{qSupported}).
37290 @item qXfer:siginfo:read::@var{offset},@var{length}
37291 @anchor{qXfer siginfo read}
37292 Read contents of the extra signal information on the target
37293 system. The annex part of the generic @samp{qXfer} packet must be
37294 empty (@pxref{qXfer read}).
37296 This packet is not probed by default; the remote stub must request it,
37297 by supplying an appropriate @samp{qSupported} response
37298 (@pxref{qSupported}).
37300 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
37301 @anchor{qXfer spu read}
37302 Read contents of an @code{spufs} file on the target system. The
37303 annex specifies which file to read; it must be of the form
37304 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
37305 in the target process, and @var{name} identifes the @code{spufs} file
37306 in that context to be accessed.
37308 This packet is not probed by default; the remote stub must request it,
37309 by supplying an appropriate @samp{qSupported} response
37310 (@pxref{qSupported}).
37312 @item qXfer:threads:read::@var{offset},@var{length}
37313 @anchor{qXfer threads read}
37314 Access the list of threads on target. @xref{Thread List Format}. The
37315 annex part of the generic @samp{qXfer} packet must be empty
37316 (@pxref{qXfer read}).
37318 This packet is not probed by default; the remote stub must request it,
37319 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37321 @item qXfer:traceframe-info:read::@var{offset},@var{length}
37322 @anchor{qXfer traceframe info read}
37324 Return a description of the current traceframe's contents.
37325 @xref{Traceframe Info Format}. The annex part of the generic
37326 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
37328 This packet is not probed by default; the remote stub must request it,
37329 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37331 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
37332 @anchor{qXfer unwind info block}
37334 Return the unwind information block for @var{pc}. This packet is used
37335 on OpenVMS/ia64 to ask the kernel unwind information.
37337 This packet is not probed by default.
37339 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
37340 @anchor{qXfer fdpic loadmap read}
37341 Read contents of @code{loadmap}s on the target system. The
37342 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
37343 executable @code{loadmap} or interpreter @code{loadmap} to read.
37345 This packet is not probed by default; the remote stub must request it,
37346 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37348 @item qXfer:osdata:read::@var{offset},@var{length}
37349 @anchor{qXfer osdata read}
37350 Access the target's @dfn{operating system information}.
37351 @xref{Operating System Information}.
37358 Data @var{data} (@pxref{Binary Data}) has been read from the
37359 target. There may be more data at a higher address (although
37360 it is permitted to return @samp{m} even for the last valid
37361 block of data, as long as at least one byte of data was read).
37362 It is possible for @var{data} to have fewer bytes than the @var{length} in the
37366 Data @var{data} (@pxref{Binary Data}) has been read from the target.
37367 There is no more data to be read. It is possible for @var{data} to
37368 have fewer bytes than the @var{length} in the request.
37371 The @var{offset} in the request is at the end of the data.
37372 There is no more data to be read.
37375 The request was malformed, or @var{annex} was invalid.
37378 The offset was invalid, or there was an error encountered reading the data.
37379 The @var{nn} part is a hex-encoded @code{errno} value.
37382 An empty reply indicates the @var{object} string was not recognized by
37383 the stub, or that the object does not support reading.
37386 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
37387 @cindex write data into object, remote request
37388 @anchor{qXfer write}
37389 Write uninterpreted bytes into the target's special data area
37390 identified by the keyword @var{object}, starting at @var{offset} bytes
37391 into the data. The binary-encoded data (@pxref{Binary Data}) to be
37392 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
37393 is specific to @var{object}; it can supply additional details about what data
37396 Here are the specific requests of this form defined so far. All
37397 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
37398 formats, listed below.
37401 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
37402 @anchor{qXfer siginfo write}
37403 Write @var{data} to the extra signal information on the target system.
37404 The annex part of the generic @samp{qXfer} packet must be
37405 empty (@pxref{qXfer write}).
37407 This packet is not probed by default; the remote stub must request it,
37408 by supplying an appropriate @samp{qSupported} response
37409 (@pxref{qSupported}).
37411 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
37412 @anchor{qXfer spu write}
37413 Write @var{data} to an @code{spufs} file on the target system. The
37414 annex specifies which file to write; it must be of the form
37415 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
37416 in the target process, and @var{name} identifes the @code{spufs} file
37417 in that context to be accessed.
37419 This packet is not probed by default; the remote stub must request it,
37420 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37426 @var{nn} (hex encoded) is the number of bytes written.
37427 This may be fewer bytes than supplied in the request.
37430 The request was malformed, or @var{annex} was invalid.
37433 The offset was invalid, or there was an error encountered writing the data.
37434 The @var{nn} part is a hex-encoded @code{errno} value.
37437 An empty reply indicates the @var{object} string was not
37438 recognized by the stub, or that the object does not support writing.
37441 @item qXfer:@var{object}:@var{operation}:@dots{}
37442 Requests of this form may be added in the future. When a stub does
37443 not recognize the @var{object} keyword, or its support for
37444 @var{object} does not recognize the @var{operation} keyword, the stub
37445 must respond with an empty packet.
37447 @item qAttached:@var{pid}
37448 @cindex query attached, remote request
37449 @cindex @samp{qAttached} packet
37450 Return an indication of whether the remote server attached to an
37451 existing process or created a new process. When the multiprocess
37452 protocol extensions are supported (@pxref{multiprocess extensions}),
37453 @var{pid} is an integer in hexadecimal format identifying the target
37454 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
37455 the query packet will be simplified as @samp{qAttached}.
37457 This query is used, for example, to know whether the remote process
37458 should be detached or killed when a @value{GDBN} session is ended with
37459 the @code{quit} command.
37464 The remote server attached to an existing process.
37466 The remote server created a new process.
37468 A badly formed request or an error was encountered.
37472 Enable branch tracing for the current thread using Branch Trace Store.
37477 Branch tracing has been enabled.
37479 A badly formed request or an error was encountered.
37483 Enable branch tracing for the current thread using Intel Processor Trace.
37488 Branch tracing has been enabled.
37490 A badly formed request or an error was encountered.
37494 Disable branch tracing for the current thread.
37499 Branch tracing has been disabled.
37501 A badly formed request or an error was encountered.
37504 @item Qbtrace-conf:bts:size=@var{value}
37505 Set the requested ring buffer size for new threads that use the
37506 btrace recording method in bts format.
37511 The ring buffer size has been set.
37513 A badly formed request or an error was encountered.
37516 @item Qbtrace-conf:pt:size=@var{value}
37517 Set the requested ring buffer size for new threads that use the
37518 btrace recording method in pt format.
37523 The ring buffer size has been set.
37525 A badly formed request or an error was encountered.
37530 @node Architecture-Specific Protocol Details
37531 @section Architecture-Specific Protocol Details
37533 This section describes how the remote protocol is applied to specific
37534 target architectures. Also see @ref{Standard Target Features}, for
37535 details of XML target descriptions for each architecture.
37538 * ARM-Specific Protocol Details::
37539 * MIPS-Specific Protocol Details::
37542 @node ARM-Specific Protocol Details
37543 @subsection @acronym{ARM}-specific Protocol Details
37546 * ARM Breakpoint Kinds::
37549 @node ARM Breakpoint Kinds
37550 @subsubsection @acronym{ARM} Breakpoint Kinds
37551 @cindex breakpoint kinds, @acronym{ARM}
37553 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
37558 16-bit Thumb mode breakpoint.
37561 32-bit Thumb mode (Thumb-2) breakpoint.
37564 32-bit @acronym{ARM} mode breakpoint.
37568 @node MIPS-Specific Protocol Details
37569 @subsection @acronym{MIPS}-specific Protocol Details
37572 * MIPS Register packet Format::
37573 * MIPS Breakpoint Kinds::
37576 @node MIPS Register packet Format
37577 @subsubsection @acronym{MIPS} Register Packet Format
37578 @cindex register packet format, @acronym{MIPS}
37580 The following @code{g}/@code{G} packets have previously been defined.
37581 In the below, some thirty-two bit registers are transferred as
37582 sixty-four bits. Those registers should be zero/sign extended (which?)
37583 to fill the space allocated. Register bytes are transferred in target
37584 byte order. The two nibbles within a register byte are transferred
37585 most-significant -- least-significant.
37590 All registers are transferred as thirty-two bit quantities in the order:
37591 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
37592 registers; fsr; fir; fp.
37595 All registers are transferred as sixty-four bit quantities (including
37596 thirty-two bit registers such as @code{sr}). The ordering is the same
37601 @node MIPS Breakpoint Kinds
37602 @subsubsection @acronym{MIPS} Breakpoint Kinds
37603 @cindex breakpoint kinds, @acronym{MIPS}
37605 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
37610 16-bit @acronym{MIPS16} mode breakpoint.
37613 16-bit @acronym{microMIPS} mode breakpoint.
37616 32-bit standard @acronym{MIPS} mode breakpoint.
37619 32-bit @acronym{microMIPS} mode breakpoint.
37623 @node Tracepoint Packets
37624 @section Tracepoint Packets
37625 @cindex tracepoint packets
37626 @cindex packets, tracepoint
37628 Here we describe the packets @value{GDBN} uses to implement
37629 tracepoints (@pxref{Tracepoints}).
37633 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
37634 @cindex @samp{QTDP} packet
37635 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
37636 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
37637 the tracepoint is disabled. The @var{step} gives the tracepoint's step
37638 count, and @var{pass} gives its pass count. If an @samp{F} is present,
37639 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
37640 the number of bytes that the target should copy elsewhere to make room
37641 for the tracepoint. If an @samp{X} is present, it introduces a
37642 tracepoint condition, which consists of a hexadecimal length, followed
37643 by a comma and hex-encoded bytes, in a manner similar to action
37644 encodings as described below. If the trailing @samp{-} is present,
37645 further @samp{QTDP} packets will follow to specify this tracepoint's
37651 The packet was understood and carried out.
37653 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
37655 The packet was not recognized.
37658 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
37659 Define actions to be taken when a tracepoint is hit. The @var{n} and
37660 @var{addr} must be the same as in the initial @samp{QTDP} packet for
37661 this tracepoint. This packet may only be sent immediately after
37662 another @samp{QTDP} packet that ended with a @samp{-}. If the
37663 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
37664 specifying more actions for this tracepoint.
37666 In the series of action packets for a given tracepoint, at most one
37667 can have an @samp{S} before its first @var{action}. If such a packet
37668 is sent, it and the following packets define ``while-stepping''
37669 actions. Any prior packets define ordinary actions --- that is, those
37670 taken when the tracepoint is first hit. If no action packet has an
37671 @samp{S}, then all the packets in the series specify ordinary
37672 tracepoint actions.
37674 The @samp{@var{action}@dots{}} portion of the packet is a series of
37675 actions, concatenated without separators. Each action has one of the
37681 Collect the registers whose bits are set in @var{mask},
37682 a hexadecimal number whose @var{i}'th bit is set if register number
37683 @var{i} should be collected. (The least significant bit is numbered
37684 zero.) Note that @var{mask} may be any number of digits long; it may
37685 not fit in a 32-bit word.
37687 @item M @var{basereg},@var{offset},@var{len}
37688 Collect @var{len} bytes of memory starting at the address in register
37689 number @var{basereg}, plus @var{offset}. If @var{basereg} is
37690 @samp{-1}, then the range has a fixed address: @var{offset} is the
37691 address of the lowest byte to collect. The @var{basereg},
37692 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
37693 values (the @samp{-1} value for @var{basereg} is a special case).
37695 @item X @var{len},@var{expr}
37696 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
37697 it directs. The agent expression @var{expr} is as described in
37698 @ref{Agent Expressions}. Each byte of the expression is encoded as a
37699 two-digit hex number in the packet; @var{len} is the number of bytes
37700 in the expression (and thus one-half the number of hex digits in the
37705 Any number of actions may be packed together in a single @samp{QTDP}
37706 packet, as long as the packet does not exceed the maximum packet
37707 length (400 bytes, for many stubs). There may be only one @samp{R}
37708 action per tracepoint, and it must precede any @samp{M} or @samp{X}
37709 actions. Any registers referred to by @samp{M} and @samp{X} actions
37710 must be collected by a preceding @samp{R} action. (The
37711 ``while-stepping'' actions are treated as if they were attached to a
37712 separate tracepoint, as far as these restrictions are concerned.)
37717 The packet was understood and carried out.
37719 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
37721 The packet was not recognized.
37724 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
37725 @cindex @samp{QTDPsrc} packet
37726 Specify a source string of tracepoint @var{n} at address @var{addr}.
37727 This is useful to get accurate reproduction of the tracepoints
37728 originally downloaded at the beginning of the trace run. The @var{type}
37729 is the name of the tracepoint part, such as @samp{cond} for the
37730 tracepoint's conditional expression (see below for a list of types), while
37731 @var{bytes} is the string, encoded in hexadecimal.
37733 @var{start} is the offset of the @var{bytes} within the overall source
37734 string, while @var{slen} is the total length of the source string.
37735 This is intended for handling source strings that are longer than will
37736 fit in a single packet.
37737 @c Add detailed example when this info is moved into a dedicated
37738 @c tracepoint descriptions section.
37740 The available string types are @samp{at} for the location,
37741 @samp{cond} for the conditional, and @samp{cmd} for an action command.
37742 @value{GDBN} sends a separate packet for each command in the action
37743 list, in the same order in which the commands are stored in the list.
37745 The target does not need to do anything with source strings except
37746 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
37749 Although this packet is optional, and @value{GDBN} will only send it
37750 if the target replies with @samp{TracepointSource} @xref{General
37751 Query Packets}, it makes both disconnected tracing and trace files
37752 much easier to use. Otherwise the user must be careful that the
37753 tracepoints in effect while looking at trace frames are identical to
37754 the ones in effect during the trace run; even a small discrepancy
37755 could cause @samp{tdump} not to work, or a particular trace frame not
37758 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
37759 @cindex define trace state variable, remote request
37760 @cindex @samp{QTDV} packet
37761 Create a new trace state variable, number @var{n}, with an initial
37762 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
37763 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
37764 the option of not using this packet for initial values of zero; the
37765 target should simply create the trace state variables as they are
37766 mentioned in expressions. The value @var{builtin} should be 1 (one)
37767 if the trace state variable is builtin and 0 (zero) if it is not builtin.
37768 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
37769 @samp{qTsV} packet had it set. The contents of @var{name} is the
37770 hex-encoded name (without the leading @samp{$}) of the trace state
37773 @item QTFrame:@var{n}
37774 @cindex @samp{QTFrame} packet
37775 Select the @var{n}'th tracepoint frame from the buffer, and use the
37776 register and memory contents recorded there to answer subsequent
37777 request packets from @value{GDBN}.
37779 A successful reply from the stub indicates that the stub has found the
37780 requested frame. The response is a series of parts, concatenated
37781 without separators, describing the frame we selected. Each part has
37782 one of the following forms:
37786 The selected frame is number @var{n} in the trace frame buffer;
37787 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
37788 was no frame matching the criteria in the request packet.
37791 The selected trace frame records a hit of tracepoint number @var{t};
37792 @var{t} is a hexadecimal number.
37796 @item QTFrame:pc:@var{addr}
37797 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
37798 currently selected frame whose PC is @var{addr};
37799 @var{addr} is a hexadecimal number.
37801 @item QTFrame:tdp:@var{t}
37802 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
37803 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
37804 is a hexadecimal number.
37806 @item QTFrame:range:@var{start}:@var{end}
37807 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
37808 currently selected frame whose PC is between @var{start} (inclusive)
37809 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
37812 @item QTFrame:outside:@var{start}:@var{end}
37813 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
37814 frame @emph{outside} the given range of addresses (exclusive).
37817 @cindex @samp{qTMinFTPILen} packet
37818 This packet requests the minimum length of instruction at which a fast
37819 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
37820 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
37821 it depends on the target system being able to create trampolines in
37822 the first 64K of memory, which might or might not be possible for that
37823 system. So the reply to this packet will be 4 if it is able to
37830 The minimum instruction length is currently unknown.
37832 The minimum instruction length is @var{length}, where @var{length}
37833 is a hexadecimal number greater or equal to 1. A reply
37834 of 1 means that a fast tracepoint may be placed on any instruction
37835 regardless of size.
37837 An error has occurred.
37839 An empty reply indicates that the request is not supported by the stub.
37843 @cindex @samp{QTStart} packet
37844 Begin the tracepoint experiment. Begin collecting data from
37845 tracepoint hits in the trace frame buffer. This packet supports the
37846 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
37847 instruction reply packet}).
37850 @cindex @samp{QTStop} packet
37851 End the tracepoint experiment. Stop collecting trace frames.
37853 @item QTEnable:@var{n}:@var{addr}
37855 @cindex @samp{QTEnable} packet
37856 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
37857 experiment. If the tracepoint was previously disabled, then collection
37858 of data from it will resume.
37860 @item QTDisable:@var{n}:@var{addr}
37862 @cindex @samp{QTDisable} packet
37863 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
37864 experiment. No more data will be collected from the tracepoint unless
37865 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
37868 @cindex @samp{QTinit} packet
37869 Clear the table of tracepoints, and empty the trace frame buffer.
37871 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
37872 @cindex @samp{QTro} packet
37873 Establish the given ranges of memory as ``transparent''. The stub
37874 will answer requests for these ranges from memory's current contents,
37875 if they were not collected as part of the tracepoint hit.
37877 @value{GDBN} uses this to mark read-only regions of memory, like those
37878 containing program code. Since these areas never change, they should
37879 still have the same contents they did when the tracepoint was hit, so
37880 there's no reason for the stub to refuse to provide their contents.
37882 @item QTDisconnected:@var{value}
37883 @cindex @samp{QTDisconnected} packet
37884 Set the choice to what to do with the tracing run when @value{GDBN}
37885 disconnects from the target. A @var{value} of 1 directs the target to
37886 continue the tracing run, while 0 tells the target to stop tracing if
37887 @value{GDBN} is no longer in the picture.
37890 @cindex @samp{qTStatus} packet
37891 Ask the stub if there is a trace experiment running right now.
37893 The reply has the form:
37897 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
37898 @var{running} is a single digit @code{1} if the trace is presently
37899 running, or @code{0} if not. It is followed by semicolon-separated
37900 optional fields that an agent may use to report additional status.
37904 If the trace is not running, the agent may report any of several
37905 explanations as one of the optional fields:
37910 No trace has been run yet.
37912 @item tstop[:@var{text}]:0
37913 The trace was stopped by a user-originated stop command. The optional
37914 @var{text} field is a user-supplied string supplied as part of the
37915 stop command (for instance, an explanation of why the trace was
37916 stopped manually). It is hex-encoded.
37919 The trace stopped because the trace buffer filled up.
37921 @item tdisconnected:0
37922 The trace stopped because @value{GDBN} disconnected from the target.
37924 @item tpasscount:@var{tpnum}
37925 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
37927 @item terror:@var{text}:@var{tpnum}
37928 The trace stopped because tracepoint @var{tpnum} had an error. The
37929 string @var{text} is available to describe the nature of the error
37930 (for instance, a divide by zero in the condition expression); it
37934 The trace stopped for some other reason.
37938 Additional optional fields supply statistical and other information.
37939 Although not required, they are extremely useful for users monitoring
37940 the progress of a trace run. If a trace has stopped, and these
37941 numbers are reported, they must reflect the state of the just-stopped
37946 @item tframes:@var{n}
37947 The number of trace frames in the buffer.
37949 @item tcreated:@var{n}
37950 The total number of trace frames created during the run. This may
37951 be larger than the trace frame count, if the buffer is circular.
37953 @item tsize:@var{n}
37954 The total size of the trace buffer, in bytes.
37956 @item tfree:@var{n}
37957 The number of bytes still unused in the buffer.
37959 @item circular:@var{n}
37960 The value of the circular trace buffer flag. @code{1} means that the
37961 trace buffer is circular and old trace frames will be discarded if
37962 necessary to make room, @code{0} means that the trace buffer is linear
37965 @item disconn:@var{n}
37966 The value of the disconnected tracing flag. @code{1} means that
37967 tracing will continue after @value{GDBN} disconnects, @code{0} means
37968 that the trace run will stop.
37972 @item qTP:@var{tp}:@var{addr}
37973 @cindex tracepoint status, remote request
37974 @cindex @samp{qTP} packet
37975 Ask the stub for the current state of tracepoint number @var{tp} at
37976 address @var{addr}.
37980 @item V@var{hits}:@var{usage}
37981 The tracepoint has been hit @var{hits} times so far during the trace
37982 run, and accounts for @var{usage} in the trace buffer. Note that
37983 @code{while-stepping} steps are not counted as separate hits, but the
37984 steps' space consumption is added into the usage number.
37988 @item qTV:@var{var}
37989 @cindex trace state variable value, remote request
37990 @cindex @samp{qTV} packet
37991 Ask the stub for the value of the trace state variable number @var{var}.
37996 The value of the variable is @var{value}. This will be the current
37997 value of the variable if the user is examining a running target, or a
37998 saved value if the variable was collected in the trace frame that the
37999 user is looking at. Note that multiple requests may result in
38000 different reply values, such as when requesting values while the
38001 program is running.
38004 The value of the variable is unknown. This would occur, for example,
38005 if the user is examining a trace frame in which the requested variable
38010 @cindex @samp{qTfP} packet
38012 @cindex @samp{qTsP} packet
38013 These packets request data about tracepoints that are being used by
38014 the target. @value{GDBN} sends @code{qTfP} to get the first piece
38015 of data, and multiple @code{qTsP} to get additional pieces. Replies
38016 to these packets generally take the form of the @code{QTDP} packets
38017 that define tracepoints. (FIXME add detailed syntax)
38020 @cindex @samp{qTfV} packet
38022 @cindex @samp{qTsV} packet
38023 These packets request data about trace state variables that are on the
38024 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
38025 and multiple @code{qTsV} to get additional variables. Replies to
38026 these packets follow the syntax of the @code{QTDV} packets that define
38027 trace state variables.
38033 @cindex @samp{qTfSTM} packet
38034 @cindex @samp{qTsSTM} packet
38035 These packets request data about static tracepoint markers that exist
38036 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
38037 first piece of data, and multiple @code{qTsSTM} to get additional
38038 pieces. Replies to these packets take the following form:
38042 @item m @var{address}:@var{id}:@var{extra}
38044 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
38045 a comma-separated list of markers
38047 (lower case letter @samp{L}) denotes end of list.
38049 An error occurred. The error number @var{nn} is given as hex digits.
38051 An empty reply indicates that the request is not supported by the
38055 The @var{address} is encoded in hex;
38056 @var{id} and @var{extra} are strings encoded in hex.
38058 In response to each query, the target will reply with a list of one or
38059 more markers, separated by commas. @value{GDBN} will respond to each
38060 reply with a request for more markers (using the @samp{qs} form of the
38061 query), until the target responds with @samp{l} (lower-case ell, for
38064 @item qTSTMat:@var{address}
38066 @cindex @samp{qTSTMat} packet
38067 This packets requests data about static tracepoint markers in the
38068 target program at @var{address}. Replies to this packet follow the
38069 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
38070 tracepoint markers.
38072 @item QTSave:@var{filename}
38073 @cindex @samp{QTSave} packet
38074 This packet directs the target to save trace data to the file name
38075 @var{filename} in the target's filesystem. The @var{filename} is encoded
38076 as a hex string; the interpretation of the file name (relative vs
38077 absolute, wild cards, etc) is up to the target.
38079 @item qTBuffer:@var{offset},@var{len}
38080 @cindex @samp{qTBuffer} packet
38081 Return up to @var{len} bytes of the current contents of trace buffer,
38082 starting at @var{offset}. The trace buffer is treated as if it were
38083 a contiguous collection of traceframes, as per the trace file format.
38084 The reply consists as many hex-encoded bytes as the target can deliver
38085 in a packet; it is not an error to return fewer than were asked for.
38086 A reply consisting of just @code{l} indicates that no bytes are
38089 @item QTBuffer:circular:@var{value}
38090 This packet directs the target to use a circular trace buffer if
38091 @var{value} is 1, or a linear buffer if the value is 0.
38093 @item QTBuffer:size:@var{size}
38094 @anchor{QTBuffer-size}
38095 @cindex @samp{QTBuffer size} packet
38096 This packet directs the target to make the trace buffer be of size
38097 @var{size} if possible. A value of @code{-1} tells the target to
38098 use whatever size it prefers.
38100 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
38101 @cindex @samp{QTNotes} packet
38102 This packet adds optional textual notes to the trace run. Allowable
38103 types include @code{user}, @code{notes}, and @code{tstop}, the
38104 @var{text} fields are arbitrary strings, hex-encoded.
38108 @subsection Relocate instruction reply packet
38109 When installing fast tracepoints in memory, the target may need to
38110 relocate the instruction currently at the tracepoint address to a
38111 different address in memory. For most instructions, a simple copy is
38112 enough, but, for example, call instructions that implicitly push the
38113 return address on the stack, and relative branches or other
38114 PC-relative instructions require offset adjustment, so that the effect
38115 of executing the instruction at a different address is the same as if
38116 it had executed in the original location.
38118 In response to several of the tracepoint packets, the target may also
38119 respond with a number of intermediate @samp{qRelocInsn} request
38120 packets before the final result packet, to have @value{GDBN} handle
38121 this relocation operation. If a packet supports this mechanism, its
38122 documentation will explicitly say so. See for example the above
38123 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
38124 format of the request is:
38127 @item qRelocInsn:@var{from};@var{to}
38129 This requests @value{GDBN} to copy instruction at address @var{from}
38130 to address @var{to}, possibly adjusted so that executing the
38131 instruction at @var{to} has the same effect as executing it at
38132 @var{from}. @value{GDBN} writes the adjusted instruction to target
38133 memory starting at @var{to}.
38138 @item qRelocInsn:@var{adjusted_size}
38139 Informs the stub the relocation is complete. The @var{adjusted_size} is
38140 the length in bytes of resulting relocated instruction sequence.
38142 A badly formed request was detected, or an error was encountered while
38143 relocating the instruction.
38146 @node Host I/O Packets
38147 @section Host I/O Packets
38148 @cindex Host I/O, remote protocol
38149 @cindex file transfer, remote protocol
38151 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
38152 operations on the far side of a remote link. For example, Host I/O is
38153 used to upload and download files to a remote target with its own
38154 filesystem. Host I/O uses the same constant values and data structure
38155 layout as the target-initiated File-I/O protocol. However, the
38156 Host I/O packets are structured differently. The target-initiated
38157 protocol relies on target memory to store parameters and buffers.
38158 Host I/O requests are initiated by @value{GDBN}, and the
38159 target's memory is not involved. @xref{File-I/O Remote Protocol
38160 Extension}, for more details on the target-initiated protocol.
38162 The Host I/O request packets all encode a single operation along with
38163 its arguments. They have this format:
38167 @item vFile:@var{operation}: @var{parameter}@dots{}
38168 @var{operation} is the name of the particular request; the target
38169 should compare the entire packet name up to the second colon when checking
38170 for a supported operation. The format of @var{parameter} depends on
38171 the operation. Numbers are always passed in hexadecimal. Negative
38172 numbers have an explicit minus sign (i.e.@: two's complement is not
38173 used). Strings (e.g.@: filenames) are encoded as a series of
38174 hexadecimal bytes. The last argument to a system call may be a
38175 buffer of escaped binary data (@pxref{Binary Data}).
38179 The valid responses to Host I/O packets are:
38183 @item F @var{result} [, @var{errno}] [; @var{attachment}]
38184 @var{result} is the integer value returned by this operation, usually
38185 non-negative for success and -1 for errors. If an error has occured,
38186 @var{errno} will be included in the result specifying a
38187 value defined by the File-I/O protocol (@pxref{Errno Values}). For
38188 operations which return data, @var{attachment} supplies the data as a
38189 binary buffer. Binary buffers in response packets are escaped in the
38190 normal way (@pxref{Binary Data}). See the individual packet
38191 documentation for the interpretation of @var{result} and
38195 An empty response indicates that this operation is not recognized.
38199 These are the supported Host I/O operations:
38202 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
38203 Open a file at @var{filename} and return a file descriptor for it, or
38204 return -1 if an error occurs. The @var{filename} is a string,
38205 @var{flags} is an integer indicating a mask of open flags
38206 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
38207 of mode bits to use if the file is created (@pxref{mode_t Values}).
38208 @xref{open}, for details of the open flags and mode values.
38210 @item vFile:close: @var{fd}
38211 Close the open file corresponding to @var{fd} and return 0, or
38212 -1 if an error occurs.
38214 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
38215 Read data from the open file corresponding to @var{fd}. Up to
38216 @var{count} bytes will be read from the file, starting at @var{offset}
38217 relative to the start of the file. The target may read fewer bytes;
38218 common reasons include packet size limits and an end-of-file
38219 condition. The number of bytes read is returned. Zero should only be
38220 returned for a successful read at the end of the file, or if
38221 @var{count} was zero.
38223 The data read should be returned as a binary attachment on success.
38224 If zero bytes were read, the response should include an empty binary
38225 attachment (i.e.@: a trailing semicolon). The return value is the
38226 number of target bytes read; the binary attachment may be longer if
38227 some characters were escaped.
38229 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
38230 Write @var{data} (a binary buffer) to the open file corresponding
38231 to @var{fd}. Start the write at @var{offset} from the start of the
38232 file. Unlike many @code{write} system calls, there is no
38233 separate @var{count} argument; the length of @var{data} in the
38234 packet is used. @samp{vFile:write} returns the number of bytes written,
38235 which may be shorter than the length of @var{data}, or -1 if an
38238 @item vFile:fstat: @var{fd}
38239 Get information about the open file corresponding to @var{fd}.
38240 On success the information is returned as a binary attachment
38241 and the return value is the size of this attachment in bytes.
38242 If an error occurs the return value is -1. The format of the
38243 returned binary attachment is as described in @ref{struct stat}.
38245 @item vFile:unlink: @var{filename}
38246 Delete the file at @var{filename} on the target. Return 0,
38247 or -1 if an error occurs. The @var{filename} is a string.
38249 @item vFile:readlink: @var{filename}
38250 Read value of symbolic link @var{filename} on the target. Return
38251 the number of bytes read, or -1 if an error occurs.
38253 The data read should be returned as a binary attachment on success.
38254 If zero bytes were read, the response should include an empty binary
38255 attachment (i.e.@: a trailing semicolon). The return value is the
38256 number of target bytes read; the binary attachment may be longer if
38257 some characters were escaped.
38259 @item vFile:setfs: @var{pid}
38260 Select the filesystem on which @code{vFile} operations with
38261 @var{filename} arguments will operate. This is required for
38262 @value{GDBN} to be able to access files on remote targets where
38263 the remote stub does not share a common filesystem with the
38266 If @var{pid} is nonzero, select the filesystem as seen by process
38267 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
38268 the remote stub. Return 0 on success, or -1 if an error occurs.
38269 If @code{vFile:setfs:} indicates success, the selected filesystem
38270 remains selected until the next successful @code{vFile:setfs:}
38276 @section Interrupts
38277 @cindex interrupts (remote protocol)
38278 @anchor{interrupting remote targets}
38280 In all-stop mode, when a program on the remote target is running,
38281 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
38282 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
38283 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
38285 The precise meaning of @code{BREAK} is defined by the transport
38286 mechanism and may, in fact, be undefined. @value{GDBN} does not
38287 currently define a @code{BREAK} mechanism for any of the network
38288 interfaces except for TCP, in which case @value{GDBN} sends the
38289 @code{telnet} BREAK sequence.
38291 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
38292 transport mechanisms. It is represented by sending the single byte
38293 @code{0x03} without any of the usual packet overhead described in
38294 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
38295 transmitted as part of a packet, it is considered to be packet data
38296 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
38297 (@pxref{X packet}), used for binary downloads, may include an unescaped
38298 @code{0x03} as part of its packet.
38300 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
38301 When Linux kernel receives this sequence from serial port,
38302 it stops execution and connects to gdb.
38304 In non-stop mode, because packet resumptions are asynchronous
38305 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
38306 command to the remote stub, even when the target is running. For that
38307 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
38308 packet}) with the usual packet framing instead of the single byte
38311 Stubs are not required to recognize these interrupt mechanisms and the
38312 precise meaning associated with receipt of the interrupt is
38313 implementation defined. If the target supports debugging of multiple
38314 threads and/or processes, it should attempt to interrupt all
38315 currently-executing threads and processes.
38316 If the stub is successful at interrupting the
38317 running program, it should send one of the stop
38318 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
38319 of successfully stopping the program in all-stop mode, and a stop reply
38320 for each stopped thread in non-stop mode.
38321 Interrupts received while the
38322 program is stopped are queued and the program will be interrupted when
38323 it is resumed next time.
38325 @node Notification Packets
38326 @section Notification Packets
38327 @cindex notification packets
38328 @cindex packets, notification
38330 The @value{GDBN} remote serial protocol includes @dfn{notifications},
38331 packets that require no acknowledgment. Both the GDB and the stub
38332 may send notifications (although the only notifications defined at
38333 present are sent by the stub). Notifications carry information
38334 without incurring the round-trip latency of an acknowledgment, and so
38335 are useful for low-impact communications where occasional packet loss
38338 A notification packet has the form @samp{% @var{data} #
38339 @var{checksum}}, where @var{data} is the content of the notification,
38340 and @var{checksum} is a checksum of @var{data}, computed and formatted
38341 as for ordinary @value{GDBN} packets. A notification's @var{data}
38342 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
38343 receiving a notification, the recipient sends no @samp{+} or @samp{-}
38344 to acknowledge the notification's receipt or to report its corruption.
38346 Every notification's @var{data} begins with a name, which contains no
38347 colon characters, followed by a colon character.
38349 Recipients should silently ignore corrupted notifications and
38350 notifications they do not understand. Recipients should restart
38351 timeout periods on receipt of a well-formed notification, whether or
38352 not they understand it.
38354 Senders should only send the notifications described here when this
38355 protocol description specifies that they are permitted. In the
38356 future, we may extend the protocol to permit existing notifications in
38357 new contexts; this rule helps older senders avoid confusing newer
38360 (Older versions of @value{GDBN} ignore bytes received until they see
38361 the @samp{$} byte that begins an ordinary packet, so new stubs may
38362 transmit notifications without fear of confusing older clients. There
38363 are no notifications defined for @value{GDBN} to send at the moment, but we
38364 assume that most older stubs would ignore them, as well.)
38366 Each notification is comprised of three parts:
38368 @item @var{name}:@var{event}
38369 The notification packet is sent by the side that initiates the
38370 exchange (currently, only the stub does that), with @var{event}
38371 carrying the specific information about the notification, and
38372 @var{name} specifying the name of the notification.
38374 The acknowledge sent by the other side, usually @value{GDBN}, to
38375 acknowledge the exchange and request the event.
38378 The purpose of an asynchronous notification mechanism is to report to
38379 @value{GDBN} that something interesting happened in the remote stub.
38381 The remote stub may send notification @var{name}:@var{event}
38382 at any time, but @value{GDBN} acknowledges the notification when
38383 appropriate. The notification event is pending before @value{GDBN}
38384 acknowledges. Only one notification at a time may be pending; if
38385 additional events occur before @value{GDBN} has acknowledged the
38386 previous notification, they must be queued by the stub for later
38387 synchronous transmission in response to @var{ack} packets from
38388 @value{GDBN}. Because the notification mechanism is unreliable,
38389 the stub is permitted to resend a notification if it believes
38390 @value{GDBN} may not have received it.
38392 Specifically, notifications may appear when @value{GDBN} is not
38393 otherwise reading input from the stub, or when @value{GDBN} is
38394 expecting to read a normal synchronous response or a
38395 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
38396 Notification packets are distinct from any other communication from
38397 the stub so there is no ambiguity.
38399 After receiving a notification, @value{GDBN} shall acknowledge it by
38400 sending a @var{ack} packet as a regular, synchronous request to the
38401 stub. Such acknowledgment is not required to happen immediately, as
38402 @value{GDBN} is permitted to send other, unrelated packets to the
38403 stub first, which the stub should process normally.
38405 Upon receiving a @var{ack} packet, if the stub has other queued
38406 events to report to @value{GDBN}, it shall respond by sending a
38407 normal @var{event}. @value{GDBN} shall then send another @var{ack}
38408 packet to solicit further responses; again, it is permitted to send
38409 other, unrelated packets as well which the stub should process
38412 If the stub receives a @var{ack} packet and there are no additional
38413 @var{event} to report, the stub shall return an @samp{OK} response.
38414 At this point, @value{GDBN} has finished processing a notification
38415 and the stub has completed sending any queued events. @value{GDBN}
38416 won't accept any new notifications until the final @samp{OK} is
38417 received . If further notification events occur, the stub shall send
38418 a new notification, @value{GDBN} shall accept the notification, and
38419 the process shall be repeated.
38421 The process of asynchronous notification can be illustrated by the
38424 <- @code{%%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
38427 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
38429 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
38434 The following notifications are defined:
38435 @multitable @columnfractions 0.12 0.12 0.38 0.38
38444 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
38445 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
38446 for information on how these notifications are acknowledged by
38448 @tab Report an asynchronous stop event in non-stop mode.
38452 @node Remote Non-Stop
38453 @section Remote Protocol Support for Non-Stop Mode
38455 @value{GDBN}'s remote protocol supports non-stop debugging of
38456 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
38457 supports non-stop mode, it should report that to @value{GDBN} by including
38458 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
38460 @value{GDBN} typically sends a @samp{QNonStop} packet only when
38461 establishing a new connection with the stub. Entering non-stop mode
38462 does not alter the state of any currently-running threads, but targets
38463 must stop all threads in any already-attached processes when entering
38464 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
38465 probe the target state after a mode change.
38467 In non-stop mode, when an attached process encounters an event that
38468 would otherwise be reported with a stop reply, it uses the
38469 asynchronous notification mechanism (@pxref{Notification Packets}) to
38470 inform @value{GDBN}. In contrast to all-stop mode, where all threads
38471 in all processes are stopped when a stop reply is sent, in non-stop
38472 mode only the thread reporting the stop event is stopped. That is,
38473 when reporting a @samp{S} or @samp{T} response to indicate completion
38474 of a step operation, hitting a breakpoint, or a fault, only the
38475 affected thread is stopped; any other still-running threads continue
38476 to run. When reporting a @samp{W} or @samp{X} response, all running
38477 threads belonging to other attached processes continue to run.
38479 In non-stop mode, the target shall respond to the @samp{?} packet as
38480 follows. First, any incomplete stop reply notification/@samp{vStopped}
38481 sequence in progress is abandoned. The target must begin a new
38482 sequence reporting stop events for all stopped threads, whether or not
38483 it has previously reported those events to @value{GDBN}. The first
38484 stop reply is sent as a synchronous reply to the @samp{?} packet, and
38485 subsequent stop replies are sent as responses to @samp{vStopped} packets
38486 using the mechanism described above. The target must not send
38487 asynchronous stop reply notifications until the sequence is complete.
38488 If all threads are running when the target receives the @samp{?} packet,
38489 or if the target is not attached to any process, it shall respond
38492 If the stub supports non-stop mode, it should also support the
38493 @samp{swbreak} stop reason if software breakpoints are supported, and
38494 the @samp{hwbreak} stop reason if hardware breakpoints are supported
38495 (@pxref{swbreak stop reason}). This is because given the asynchronous
38496 nature of non-stop mode, between the time a thread hits a breakpoint
38497 and the time the event is finally processed by @value{GDBN}, the
38498 breakpoint may have already been removed from the target. Due to
38499 this, @value{GDBN} needs to be able to tell whether a trap stop was
38500 caused by a delayed breakpoint event, which should be ignored, as
38501 opposed to a random trap signal, which should be reported to the user.
38502 Note the @samp{swbreak} feature implies that the target is responsible
38503 for adjusting the PC when a software breakpoint triggers, if
38504 necessary, such as on the x86 architecture.
38506 @node Packet Acknowledgment
38507 @section Packet Acknowledgment
38509 @cindex acknowledgment, for @value{GDBN} remote
38510 @cindex packet acknowledgment, for @value{GDBN} remote
38511 By default, when either the host or the target machine receives a packet,
38512 the first response expected is an acknowledgment: either @samp{+} (to indicate
38513 the package was received correctly) or @samp{-} (to request retransmission).
38514 This mechanism allows the @value{GDBN} remote protocol to operate over
38515 unreliable transport mechanisms, such as a serial line.
38517 In cases where the transport mechanism is itself reliable (such as a pipe or
38518 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
38519 It may be desirable to disable them in that case to reduce communication
38520 overhead, or for other reasons. This can be accomplished by means of the
38521 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
38523 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
38524 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
38525 and response format still includes the normal checksum, as described in
38526 @ref{Overview}, but the checksum may be ignored by the receiver.
38528 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
38529 no-acknowledgment mode, it should report that to @value{GDBN}
38530 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
38531 @pxref{qSupported}.
38532 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
38533 disabled via the @code{set remote noack-packet off} command
38534 (@pxref{Remote Configuration}),
38535 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
38536 Only then may the stub actually turn off packet acknowledgments.
38537 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
38538 response, which can be safely ignored by the stub.
38540 Note that @code{set remote noack-packet} command only affects negotiation
38541 between @value{GDBN} and the stub when subsequent connections are made;
38542 it does not affect the protocol acknowledgment state for any current
38544 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
38545 new connection is established,
38546 there is also no protocol request to re-enable the acknowledgments
38547 for the current connection, once disabled.
38552 Example sequence of a target being re-started. Notice how the restart
38553 does not get any direct output:
38558 @emph{target restarts}
38561 <- @code{T001:1234123412341234}
38565 Example sequence of a target being stepped by a single instruction:
38568 -> @code{G1445@dots{}}
38573 <- @code{T001:1234123412341234}
38577 <- @code{1455@dots{}}
38581 @node File-I/O Remote Protocol Extension
38582 @section File-I/O Remote Protocol Extension
38583 @cindex File-I/O remote protocol extension
38586 * File-I/O Overview::
38587 * Protocol Basics::
38588 * The F Request Packet::
38589 * The F Reply Packet::
38590 * The Ctrl-C Message::
38592 * List of Supported Calls::
38593 * Protocol-specific Representation of Datatypes::
38595 * File-I/O Examples::
38598 @node File-I/O Overview
38599 @subsection File-I/O Overview
38600 @cindex file-i/o overview
38602 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
38603 target to use the host's file system and console I/O to perform various
38604 system calls. System calls on the target system are translated into a
38605 remote protocol packet to the host system, which then performs the needed
38606 actions and returns a response packet to the target system.
38607 This simulates file system operations even on targets that lack file systems.
38609 The protocol is defined to be independent of both the host and target systems.
38610 It uses its own internal representation of datatypes and values. Both
38611 @value{GDBN} and the target's @value{GDBN} stub are responsible for
38612 translating the system-dependent value representations into the internal
38613 protocol representations when data is transmitted.
38615 The communication is synchronous. A system call is possible only when
38616 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
38617 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
38618 the target is stopped to allow deterministic access to the target's
38619 memory. Therefore File-I/O is not interruptible by target signals. On
38620 the other hand, it is possible to interrupt File-I/O by a user interrupt
38621 (@samp{Ctrl-C}) within @value{GDBN}.
38623 The target's request to perform a host system call does not finish
38624 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
38625 after finishing the system call, the target returns to continuing the
38626 previous activity (continue, step). No additional continue or step
38627 request from @value{GDBN} is required.
38630 (@value{GDBP}) continue
38631 <- target requests 'system call X'
38632 target is stopped, @value{GDBN} executes system call
38633 -> @value{GDBN} returns result
38634 ... target continues, @value{GDBN} returns to wait for the target
38635 <- target hits breakpoint and sends a Txx packet
38638 The protocol only supports I/O on the console and to regular files on
38639 the host file system. Character or block special devices, pipes,
38640 named pipes, sockets or any other communication method on the host
38641 system are not supported by this protocol.
38643 File I/O is not supported in non-stop mode.
38645 @node Protocol Basics
38646 @subsection Protocol Basics
38647 @cindex protocol basics, file-i/o
38649 The File-I/O protocol uses the @code{F} packet as the request as well
38650 as reply packet. Since a File-I/O system call can only occur when
38651 @value{GDBN} is waiting for a response from the continuing or stepping target,
38652 the File-I/O request is a reply that @value{GDBN} has to expect as a result
38653 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
38654 This @code{F} packet contains all information needed to allow @value{GDBN}
38655 to call the appropriate host system call:
38659 A unique identifier for the requested system call.
38662 All parameters to the system call. Pointers are given as addresses
38663 in the target memory address space. Pointers to strings are given as
38664 pointer/length pair. Numerical values are given as they are.
38665 Numerical control flags are given in a protocol-specific representation.
38669 At this point, @value{GDBN} has to perform the following actions.
38673 If the parameters include pointer values to data needed as input to a
38674 system call, @value{GDBN} requests this data from the target with a
38675 standard @code{m} packet request. This additional communication has to be
38676 expected by the target implementation and is handled as any other @code{m}
38680 @value{GDBN} translates all value from protocol representation to host
38681 representation as needed. Datatypes are coerced into the host types.
38684 @value{GDBN} calls the system call.
38687 It then coerces datatypes back to protocol representation.
38690 If the system call is expected to return data in buffer space specified
38691 by pointer parameters to the call, the data is transmitted to the
38692 target using a @code{M} or @code{X} packet. This packet has to be expected
38693 by the target implementation and is handled as any other @code{M} or @code{X}
38698 Eventually @value{GDBN} replies with another @code{F} packet which contains all
38699 necessary information for the target to continue. This at least contains
38706 @code{errno}, if has been changed by the system call.
38713 After having done the needed type and value coercion, the target continues
38714 the latest continue or step action.
38716 @node The F Request Packet
38717 @subsection The @code{F} Request Packet
38718 @cindex file-i/o request packet
38719 @cindex @code{F} request packet
38721 The @code{F} request packet has the following format:
38724 @item F@var{call-id},@var{parameter@dots{}}
38726 @var{call-id} is the identifier to indicate the host system call to be called.
38727 This is just the name of the function.
38729 @var{parameter@dots{}} are the parameters to the system call.
38730 Parameters are hexadecimal integer values, either the actual values in case
38731 of scalar datatypes, pointers to target buffer space in case of compound
38732 datatypes and unspecified memory areas, or pointer/length pairs in case
38733 of string parameters. These are appended to the @var{call-id} as a
38734 comma-delimited list. All values are transmitted in ASCII
38735 string representation, pointer/length pairs separated by a slash.
38741 @node The F Reply Packet
38742 @subsection The @code{F} Reply Packet
38743 @cindex file-i/o reply packet
38744 @cindex @code{F} reply packet
38746 The @code{F} reply packet has the following format:
38750 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
38752 @var{retcode} is the return code of the system call as hexadecimal value.
38754 @var{errno} is the @code{errno} set by the call, in protocol-specific
38756 This parameter can be omitted if the call was successful.
38758 @var{Ctrl-C flag} is only sent if the user requested a break. In this
38759 case, @var{errno} must be sent as well, even if the call was successful.
38760 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
38767 or, if the call was interrupted before the host call has been performed:
38774 assuming 4 is the protocol-specific representation of @code{EINTR}.
38779 @node The Ctrl-C Message
38780 @subsection The @samp{Ctrl-C} Message
38781 @cindex ctrl-c message, in file-i/o protocol
38783 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
38784 reply packet (@pxref{The F Reply Packet}),
38785 the target should behave as if it had
38786 gotten a break message. The meaning for the target is ``system call
38787 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
38788 (as with a break message) and return to @value{GDBN} with a @code{T02}
38791 It's important for the target to know in which
38792 state the system call was interrupted. There are two possible cases:
38796 The system call hasn't been performed on the host yet.
38799 The system call on the host has been finished.
38803 These two states can be distinguished by the target by the value of the
38804 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
38805 call hasn't been performed. This is equivalent to the @code{EINTR} handling
38806 on POSIX systems. In any other case, the target may presume that the
38807 system call has been finished --- successfully or not --- and should behave
38808 as if the break message arrived right after the system call.
38810 @value{GDBN} must behave reliably. If the system call has not been called
38811 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
38812 @code{errno} in the packet. If the system call on the host has been finished
38813 before the user requests a break, the full action must be finished by
38814 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
38815 The @code{F} packet may only be sent when either nothing has happened
38816 or the full action has been completed.
38819 @subsection Console I/O
38820 @cindex console i/o as part of file-i/o
38822 By default and if not explicitly closed by the target system, the file
38823 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
38824 on the @value{GDBN} console is handled as any other file output operation
38825 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
38826 by @value{GDBN} so that after the target read request from file descriptor
38827 0 all following typing is buffered until either one of the following
38832 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
38834 system call is treated as finished.
38837 The user presses @key{RET}. This is treated as end of input with a trailing
38841 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
38842 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
38846 If the user has typed more characters than fit in the buffer given to
38847 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
38848 either another @code{read(0, @dots{})} is requested by the target, or debugging
38849 is stopped at the user's request.
38852 @node List of Supported Calls
38853 @subsection List of Supported Calls
38854 @cindex list of supported file-i/o calls
38871 @unnumberedsubsubsec open
38872 @cindex open, file-i/o system call
38877 int open(const char *pathname, int flags);
38878 int open(const char *pathname, int flags, mode_t mode);
38882 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
38885 @var{flags} is the bitwise @code{OR} of the following values:
38889 If the file does not exist it will be created. The host
38890 rules apply as far as file ownership and time stamps
38894 When used with @code{O_CREAT}, if the file already exists it is
38895 an error and open() fails.
38898 If the file already exists and the open mode allows
38899 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
38900 truncated to zero length.
38903 The file is opened in append mode.
38906 The file is opened for reading only.
38909 The file is opened for writing only.
38912 The file is opened for reading and writing.
38916 Other bits are silently ignored.
38920 @var{mode} is the bitwise @code{OR} of the following values:
38924 User has read permission.
38927 User has write permission.
38930 Group has read permission.
38933 Group has write permission.
38936 Others have read permission.
38939 Others have write permission.
38943 Other bits are silently ignored.
38946 @item Return value:
38947 @code{open} returns the new file descriptor or -1 if an error
38954 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
38957 @var{pathname} refers to a directory.
38960 The requested access is not allowed.
38963 @var{pathname} was too long.
38966 A directory component in @var{pathname} does not exist.
38969 @var{pathname} refers to a device, pipe, named pipe or socket.
38972 @var{pathname} refers to a file on a read-only filesystem and
38973 write access was requested.
38976 @var{pathname} is an invalid pointer value.
38979 No space on device to create the file.
38982 The process already has the maximum number of files open.
38985 The limit on the total number of files open on the system
38989 The call was interrupted by the user.
38995 @unnumberedsubsubsec close
38996 @cindex close, file-i/o system call
39005 @samp{Fclose,@var{fd}}
39007 @item Return value:
39008 @code{close} returns zero on success, or -1 if an error occurred.
39014 @var{fd} isn't a valid open file descriptor.
39017 The call was interrupted by the user.
39023 @unnumberedsubsubsec read
39024 @cindex read, file-i/o system call
39029 int read(int fd, void *buf, unsigned int count);
39033 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
39035 @item Return value:
39036 On success, the number of bytes read is returned.
39037 Zero indicates end of file. If count is zero, read
39038 returns zero as well. On error, -1 is returned.
39044 @var{fd} is not a valid file descriptor or is not open for
39048 @var{bufptr} is an invalid pointer value.
39051 The call was interrupted by the user.
39057 @unnumberedsubsubsec write
39058 @cindex write, file-i/o system call
39063 int write(int fd, const void *buf, unsigned int count);
39067 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
39069 @item Return value:
39070 On success, the number of bytes written are returned.
39071 Zero indicates nothing was written. On error, -1
39078 @var{fd} is not a valid file descriptor or is not open for
39082 @var{bufptr} is an invalid pointer value.
39085 An attempt was made to write a file that exceeds the
39086 host-specific maximum file size allowed.
39089 No space on device to write the data.
39092 The call was interrupted by the user.
39098 @unnumberedsubsubsec lseek
39099 @cindex lseek, file-i/o system call
39104 long lseek (int fd, long offset, int flag);
39108 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
39110 @var{flag} is one of:
39114 The offset is set to @var{offset} bytes.
39117 The offset is set to its current location plus @var{offset}
39121 The offset is set to the size of the file plus @var{offset}
39125 @item Return value:
39126 On success, the resulting unsigned offset in bytes from
39127 the beginning of the file is returned. Otherwise, a
39128 value of -1 is returned.
39134 @var{fd} is not a valid open file descriptor.
39137 @var{fd} is associated with the @value{GDBN} console.
39140 @var{flag} is not a proper value.
39143 The call was interrupted by the user.
39149 @unnumberedsubsubsec rename
39150 @cindex rename, file-i/o system call
39155 int rename(const char *oldpath, const char *newpath);
39159 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
39161 @item Return value:
39162 On success, zero is returned. On error, -1 is returned.
39168 @var{newpath} is an existing directory, but @var{oldpath} is not a
39172 @var{newpath} is a non-empty directory.
39175 @var{oldpath} or @var{newpath} is a directory that is in use by some
39179 An attempt was made to make a directory a subdirectory
39183 A component used as a directory in @var{oldpath} or new
39184 path is not a directory. Or @var{oldpath} is a directory
39185 and @var{newpath} exists but is not a directory.
39188 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
39191 No access to the file or the path of the file.
39195 @var{oldpath} or @var{newpath} was too long.
39198 A directory component in @var{oldpath} or @var{newpath} does not exist.
39201 The file is on a read-only filesystem.
39204 The device containing the file has no room for the new
39208 The call was interrupted by the user.
39214 @unnumberedsubsubsec unlink
39215 @cindex unlink, file-i/o system call
39220 int unlink(const char *pathname);
39224 @samp{Funlink,@var{pathnameptr}/@var{len}}
39226 @item Return value:
39227 On success, zero is returned. On error, -1 is returned.
39233 No access to the file or the path of the file.
39236 The system does not allow unlinking of directories.
39239 The file @var{pathname} cannot be unlinked because it's
39240 being used by another process.
39243 @var{pathnameptr} is an invalid pointer value.
39246 @var{pathname} was too long.
39249 A directory component in @var{pathname} does not exist.
39252 A component of the path is not a directory.
39255 The file is on a read-only filesystem.
39258 The call was interrupted by the user.
39264 @unnumberedsubsubsec stat/fstat
39265 @cindex fstat, file-i/o system call
39266 @cindex stat, file-i/o system call
39271 int stat(const char *pathname, struct stat *buf);
39272 int fstat(int fd, struct stat *buf);
39276 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
39277 @samp{Ffstat,@var{fd},@var{bufptr}}
39279 @item Return value:
39280 On success, zero is returned. On error, -1 is returned.
39286 @var{fd} is not a valid open file.
39289 A directory component in @var{pathname} does not exist or the
39290 path is an empty string.
39293 A component of the path is not a directory.
39296 @var{pathnameptr} is an invalid pointer value.
39299 No access to the file or the path of the file.
39302 @var{pathname} was too long.
39305 The call was interrupted by the user.
39311 @unnumberedsubsubsec gettimeofday
39312 @cindex gettimeofday, file-i/o system call
39317 int gettimeofday(struct timeval *tv, void *tz);
39321 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
39323 @item Return value:
39324 On success, 0 is returned, -1 otherwise.
39330 @var{tz} is a non-NULL pointer.
39333 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
39339 @unnumberedsubsubsec isatty
39340 @cindex isatty, file-i/o system call
39345 int isatty(int fd);
39349 @samp{Fisatty,@var{fd}}
39351 @item Return value:
39352 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
39358 The call was interrupted by the user.
39363 Note that the @code{isatty} call is treated as a special case: it returns
39364 1 to the target if the file descriptor is attached
39365 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
39366 would require implementing @code{ioctl} and would be more complex than
39371 @unnumberedsubsubsec system
39372 @cindex system, file-i/o system call
39377 int system(const char *command);
39381 @samp{Fsystem,@var{commandptr}/@var{len}}
39383 @item Return value:
39384 If @var{len} is zero, the return value indicates whether a shell is
39385 available. A zero return value indicates a shell is not available.
39386 For non-zero @var{len}, the value returned is -1 on error and the
39387 return status of the command otherwise. Only the exit status of the
39388 command is returned, which is extracted from the host's @code{system}
39389 return value by calling @code{WEXITSTATUS(retval)}. In case
39390 @file{/bin/sh} could not be executed, 127 is returned.
39396 The call was interrupted by the user.
39401 @value{GDBN} takes over the full task of calling the necessary host calls
39402 to perform the @code{system} call. The return value of @code{system} on
39403 the host is simplified before it's returned
39404 to the target. Any termination signal information from the child process
39405 is discarded, and the return value consists
39406 entirely of the exit status of the called command.
39408 Due to security concerns, the @code{system} call is by default refused
39409 by @value{GDBN}. The user has to allow this call explicitly with the
39410 @code{set remote system-call-allowed 1} command.
39413 @item set remote system-call-allowed
39414 @kindex set remote system-call-allowed
39415 Control whether to allow the @code{system} calls in the File I/O
39416 protocol for the remote target. The default is zero (disabled).
39418 @item show remote system-call-allowed
39419 @kindex show remote system-call-allowed
39420 Show whether the @code{system} calls are allowed in the File I/O
39424 @node Protocol-specific Representation of Datatypes
39425 @subsection Protocol-specific Representation of Datatypes
39426 @cindex protocol-specific representation of datatypes, in file-i/o protocol
39429 * Integral Datatypes::
39431 * Memory Transfer::
39436 @node Integral Datatypes
39437 @unnumberedsubsubsec Integral Datatypes
39438 @cindex integral datatypes, in file-i/o protocol
39440 The integral datatypes used in the system calls are @code{int},
39441 @code{unsigned int}, @code{long}, @code{unsigned long},
39442 @code{mode_t}, and @code{time_t}.
39444 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
39445 implemented as 32 bit values in this protocol.
39447 @code{long} and @code{unsigned long} are implemented as 64 bit types.
39449 @xref{Limits}, for corresponding MIN and MAX values (similar to those
39450 in @file{limits.h}) to allow range checking on host and target.
39452 @code{time_t} datatypes are defined as seconds since the Epoch.
39454 All integral datatypes transferred as part of a memory read or write of a
39455 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
39458 @node Pointer Values
39459 @unnumberedsubsubsec Pointer Values
39460 @cindex pointer values, in file-i/o protocol
39462 Pointers to target data are transmitted as they are. An exception
39463 is made for pointers to buffers for which the length isn't
39464 transmitted as part of the function call, namely strings. Strings
39465 are transmitted as a pointer/length pair, both as hex values, e.g.@:
39472 which is a pointer to data of length 18 bytes at position 0x1aaf.
39473 The length is defined as the full string length in bytes, including
39474 the trailing null byte. For example, the string @code{"hello world"}
39475 at address 0x123456 is transmitted as
39481 @node Memory Transfer
39482 @unnumberedsubsubsec Memory Transfer
39483 @cindex memory transfer, in file-i/o protocol
39485 Structured data which is transferred using a memory read or write (for
39486 example, a @code{struct stat}) is expected to be in a protocol-specific format
39487 with all scalar multibyte datatypes being big endian. Translation to
39488 this representation needs to be done both by the target before the @code{F}
39489 packet is sent, and by @value{GDBN} before
39490 it transfers memory to the target. Transferred pointers to structured
39491 data should point to the already-coerced data at any time.
39495 @unnumberedsubsubsec struct stat
39496 @cindex struct stat, in file-i/o protocol
39498 The buffer of type @code{struct stat} used by the target and @value{GDBN}
39499 is defined as follows:
39503 unsigned int st_dev; /* device */
39504 unsigned int st_ino; /* inode */
39505 mode_t st_mode; /* protection */
39506 unsigned int st_nlink; /* number of hard links */
39507 unsigned int st_uid; /* user ID of owner */
39508 unsigned int st_gid; /* group ID of owner */
39509 unsigned int st_rdev; /* device type (if inode device) */
39510 unsigned long st_size; /* total size, in bytes */
39511 unsigned long st_blksize; /* blocksize for filesystem I/O */
39512 unsigned long st_blocks; /* number of blocks allocated */
39513 time_t st_atime; /* time of last access */
39514 time_t st_mtime; /* time of last modification */
39515 time_t st_ctime; /* time of last change */
39519 The integral datatypes conform to the definitions given in the
39520 appropriate section (see @ref{Integral Datatypes}, for details) so this
39521 structure is of size 64 bytes.
39523 The values of several fields have a restricted meaning and/or
39529 A value of 0 represents a file, 1 the console.
39532 No valid meaning for the target. Transmitted unchanged.
39535 Valid mode bits are described in @ref{Constants}. Any other
39536 bits have currently no meaning for the target.
39541 No valid meaning for the target. Transmitted unchanged.
39546 These values have a host and file system dependent
39547 accuracy. Especially on Windows hosts, the file system may not
39548 support exact timing values.
39551 The target gets a @code{struct stat} of the above representation and is
39552 responsible for coercing it to the target representation before
39555 Note that due to size differences between the host, target, and protocol
39556 representations of @code{struct stat} members, these members could eventually
39557 get truncated on the target.
39559 @node struct timeval
39560 @unnumberedsubsubsec struct timeval
39561 @cindex struct timeval, in file-i/o protocol
39563 The buffer of type @code{struct timeval} used by the File-I/O protocol
39564 is defined as follows:
39568 time_t tv_sec; /* second */
39569 long tv_usec; /* microsecond */
39573 The integral datatypes conform to the definitions given in the
39574 appropriate section (see @ref{Integral Datatypes}, for details) so this
39575 structure is of size 8 bytes.
39578 @subsection Constants
39579 @cindex constants, in file-i/o protocol
39581 The following values are used for the constants inside of the
39582 protocol. @value{GDBN} and target are responsible for translating these
39583 values before and after the call as needed.
39594 @unnumberedsubsubsec Open Flags
39595 @cindex open flags, in file-i/o protocol
39597 All values are given in hexadecimal representation.
39609 @node mode_t Values
39610 @unnumberedsubsubsec mode_t Values
39611 @cindex mode_t values, in file-i/o protocol
39613 All values are given in octal representation.
39630 @unnumberedsubsubsec Errno Values
39631 @cindex errno values, in file-i/o protocol
39633 All values are given in decimal representation.
39658 @code{EUNKNOWN} is used as a fallback error value if a host system returns
39659 any error value not in the list of supported error numbers.
39662 @unnumberedsubsubsec Lseek Flags
39663 @cindex lseek flags, in file-i/o protocol
39672 @unnumberedsubsubsec Limits
39673 @cindex limits, in file-i/o protocol
39675 All values are given in decimal representation.
39678 INT_MIN -2147483648
39680 UINT_MAX 4294967295
39681 LONG_MIN -9223372036854775808
39682 LONG_MAX 9223372036854775807
39683 ULONG_MAX 18446744073709551615
39686 @node File-I/O Examples
39687 @subsection File-I/O Examples
39688 @cindex file-i/o examples
39690 Example sequence of a write call, file descriptor 3, buffer is at target
39691 address 0x1234, 6 bytes should be written:
39694 <- @code{Fwrite,3,1234,6}
39695 @emph{request memory read from target}
39698 @emph{return "6 bytes written"}
39702 Example sequence of a read call, file descriptor 3, buffer is at target
39703 address 0x1234, 6 bytes should be read:
39706 <- @code{Fread,3,1234,6}
39707 @emph{request memory write to target}
39708 -> @code{X1234,6:XXXXXX}
39709 @emph{return "6 bytes read"}
39713 Example sequence of a read call, call fails on the host due to invalid
39714 file descriptor (@code{EBADF}):
39717 <- @code{Fread,3,1234,6}
39721 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
39725 <- @code{Fread,3,1234,6}
39730 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
39734 <- @code{Fread,3,1234,6}
39735 -> @code{X1234,6:XXXXXX}
39739 @node Library List Format
39740 @section Library List Format
39741 @cindex library list format, remote protocol
39743 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
39744 same process as your application to manage libraries. In this case,
39745 @value{GDBN} can use the loader's symbol table and normal memory
39746 operations to maintain a list of shared libraries. On other
39747 platforms, the operating system manages loaded libraries.
39748 @value{GDBN} can not retrieve the list of currently loaded libraries
39749 through memory operations, so it uses the @samp{qXfer:libraries:read}
39750 packet (@pxref{qXfer library list read}) instead. The remote stub
39751 queries the target's operating system and reports which libraries
39754 The @samp{qXfer:libraries:read} packet returns an XML document which
39755 lists loaded libraries and their offsets. Each library has an
39756 associated name and one or more segment or section base addresses,
39757 which report where the library was loaded in memory.
39759 For the common case of libraries that are fully linked binaries, the
39760 library should have a list of segments. If the target supports
39761 dynamic linking of a relocatable object file, its library XML element
39762 should instead include a list of allocated sections. The segment or
39763 section bases are start addresses, not relocation offsets; they do not
39764 depend on the library's link-time base addresses.
39766 @value{GDBN} must be linked with the Expat library to support XML
39767 library lists. @xref{Expat}.
39769 A simple memory map, with one loaded library relocated by a single
39770 offset, looks like this:
39774 <library name="/lib/libc.so.6">
39775 <segment address="0x10000000"/>
39780 Another simple memory map, with one loaded library with three
39781 allocated sections (.text, .data, .bss), looks like this:
39785 <library name="sharedlib.o">
39786 <section address="0x10000000"/>
39787 <section address="0x20000000"/>
39788 <section address="0x30000000"/>
39793 The format of a library list is described by this DTD:
39796 <!-- library-list: Root element with versioning -->
39797 <!ELEMENT library-list (library)*>
39798 <!ATTLIST library-list version CDATA #FIXED "1.0">
39799 <!ELEMENT library (segment*, section*)>
39800 <!ATTLIST library name CDATA #REQUIRED>
39801 <!ELEMENT segment EMPTY>
39802 <!ATTLIST segment address CDATA #REQUIRED>
39803 <!ELEMENT section EMPTY>
39804 <!ATTLIST section address CDATA #REQUIRED>
39807 In addition, segments and section descriptors cannot be mixed within a
39808 single library element, and you must supply at least one segment or
39809 section for each library.
39811 @node Library List Format for SVR4 Targets
39812 @section Library List Format for SVR4 Targets
39813 @cindex library list format, remote protocol
39815 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
39816 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
39817 shared libraries. Still a special library list provided by this packet is
39818 more efficient for the @value{GDBN} remote protocol.
39820 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
39821 loaded libraries and their SVR4 linker parameters. For each library on SVR4
39822 target, the following parameters are reported:
39826 @code{name}, the absolute file name from the @code{l_name} field of
39827 @code{struct link_map}.
39829 @code{lm} with address of @code{struct link_map} used for TLS
39830 (Thread Local Storage) access.
39832 @code{l_addr}, the displacement as read from the field @code{l_addr} of
39833 @code{struct link_map}. For prelinked libraries this is not an absolute
39834 memory address. It is a displacement of absolute memory address against
39835 address the file was prelinked to during the library load.
39837 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
39840 Additionally the single @code{main-lm} attribute specifies address of
39841 @code{struct link_map} used for the main executable. This parameter is used
39842 for TLS access and its presence is optional.
39844 @value{GDBN} must be linked with the Expat library to support XML
39845 SVR4 library lists. @xref{Expat}.
39847 A simple memory map, with two loaded libraries (which do not use prelink),
39851 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
39852 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
39854 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
39856 </library-list-svr>
39859 The format of an SVR4 library list is described by this DTD:
39862 <!-- library-list-svr4: Root element with versioning -->
39863 <!ELEMENT library-list-svr4 (library)*>
39864 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
39865 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
39866 <!ELEMENT library EMPTY>
39867 <!ATTLIST library name CDATA #REQUIRED>
39868 <!ATTLIST library lm CDATA #REQUIRED>
39869 <!ATTLIST library l_addr CDATA #REQUIRED>
39870 <!ATTLIST library l_ld CDATA #REQUIRED>
39873 @node Memory Map Format
39874 @section Memory Map Format
39875 @cindex memory map format
39877 To be able to write into flash memory, @value{GDBN} needs to obtain a
39878 memory map from the target. This section describes the format of the
39881 The memory map is obtained using the @samp{qXfer:memory-map:read}
39882 (@pxref{qXfer memory map read}) packet and is an XML document that
39883 lists memory regions.
39885 @value{GDBN} must be linked with the Expat library to support XML
39886 memory maps. @xref{Expat}.
39888 The top-level structure of the document is shown below:
39891 <?xml version="1.0"?>
39892 <!DOCTYPE memory-map
39893 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
39894 "http://sourceware.org/gdb/gdb-memory-map.dtd">
39900 Each region can be either:
39905 A region of RAM starting at @var{addr} and extending for @var{length}
39909 <memory type="ram" start="@var{addr}" length="@var{length}"/>
39914 A region of read-only memory:
39917 <memory type="rom" start="@var{addr}" length="@var{length}"/>
39922 A region of flash memory, with erasure blocks @var{blocksize}
39926 <memory type="flash" start="@var{addr}" length="@var{length}">
39927 <property name="blocksize">@var{blocksize}</property>
39933 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
39934 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
39935 packets to write to addresses in such ranges.
39937 The formal DTD for memory map format is given below:
39940 <!-- ................................................... -->
39941 <!-- Memory Map XML DTD ................................ -->
39942 <!-- File: memory-map.dtd .............................. -->
39943 <!-- .................................... .............. -->
39944 <!-- memory-map.dtd -->
39945 <!-- memory-map: Root element with versioning -->
39946 <!ELEMENT memory-map (memory | property)>
39947 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
39948 <!ELEMENT memory (property)>
39949 <!-- memory: Specifies a memory region,
39950 and its type, or device. -->
39951 <!ATTLIST memory type CDATA #REQUIRED
39952 start CDATA #REQUIRED
39953 length CDATA #REQUIRED
39954 device CDATA #IMPLIED>
39955 <!-- property: Generic attribute tag -->
39956 <!ELEMENT property (#PCDATA | property)*>
39957 <!ATTLIST property name CDATA #REQUIRED>
39960 @node Thread List Format
39961 @section Thread List Format
39962 @cindex thread list format
39964 To efficiently update the list of threads and their attributes,
39965 @value{GDBN} issues the @samp{qXfer:threads:read} packet
39966 (@pxref{qXfer threads read}) and obtains the XML document with
39967 the following structure:
39970 <?xml version="1.0"?>
39972 <thread id="id" core="0" name="name">
39973 ... description ...
39978 Each @samp{thread} element must have the @samp{id} attribute that
39979 identifies the thread (@pxref{thread-id syntax}). The
39980 @samp{core} attribute, if present, specifies which processor core
39981 the thread was last executing on. The @samp{name} attribute, if
39982 present, specifies the human-readable name of the thread. The content
39983 of the of @samp{thread} element is interpreted as human-readable
39984 auxiliary information.
39986 @node Traceframe Info Format
39987 @section Traceframe Info Format
39988 @cindex traceframe info format
39990 To be able to know which objects in the inferior can be examined when
39991 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
39992 memory ranges, registers and trace state variables that have been
39993 collected in a traceframe.
39995 This list is obtained using the @samp{qXfer:traceframe-info:read}
39996 (@pxref{qXfer traceframe info read}) packet and is an XML document.
39998 @value{GDBN} must be linked with the Expat library to support XML
39999 traceframe info discovery. @xref{Expat}.
40001 The top-level structure of the document is shown below:
40004 <?xml version="1.0"?>
40005 <!DOCTYPE traceframe-info
40006 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
40007 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
40013 Each traceframe block can be either:
40018 A region of collected memory starting at @var{addr} and extending for
40019 @var{length} bytes from there:
40022 <memory start="@var{addr}" length="@var{length}"/>
40026 A block indicating trace state variable numbered @var{number} has been
40030 <tvar id="@var{number}"/>
40035 The formal DTD for the traceframe info format is given below:
40038 <!ELEMENT traceframe-info (memory | tvar)* >
40039 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
40041 <!ELEMENT memory EMPTY>
40042 <!ATTLIST memory start CDATA #REQUIRED
40043 length CDATA #REQUIRED>
40045 <!ATTLIST tvar id CDATA #REQUIRED>
40048 @node Branch Trace Format
40049 @section Branch Trace Format
40050 @cindex branch trace format
40052 In order to display the branch trace of an inferior thread,
40053 @value{GDBN} needs to obtain the list of branches. This list is
40054 represented as list of sequential code blocks that are connected via
40055 branches. The code in each block has been executed sequentially.
40057 This list is obtained using the @samp{qXfer:btrace:read}
40058 (@pxref{qXfer btrace read}) packet and is an XML document.
40060 @value{GDBN} must be linked with the Expat library to support XML
40061 traceframe info discovery. @xref{Expat}.
40063 The top-level structure of the document is shown below:
40066 <?xml version="1.0"?>
40068 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
40069 "http://sourceware.org/gdb/gdb-btrace.dtd">
40078 A block of sequentially executed instructions starting at @var{begin}
40079 and ending at @var{end}:
40082 <block begin="@var{begin}" end="@var{end}"/>
40087 The formal DTD for the branch trace format is given below:
40090 <!ELEMENT btrace (block* | pt) >
40091 <!ATTLIST btrace version CDATA #FIXED "1.0">
40093 <!ELEMENT block EMPTY>
40094 <!ATTLIST block begin CDATA #REQUIRED
40095 end CDATA #REQUIRED>
40097 <!ELEMENT pt (pt-config?, raw?)>
40099 <!ELEMENT pt-config (cpu?)>
40101 <!ELEMENT cpu EMPTY>
40102 <!ATTLIST cpu vendor CDATA #REQUIRED
40103 family CDATA #REQUIRED
40104 model CDATA #REQUIRED
40105 stepping CDATA #REQUIRED>
40107 <!ELEMENT raw (#PCDATA)>
40110 @node Branch Trace Configuration Format
40111 @section Branch Trace Configuration Format
40112 @cindex branch trace configuration format
40114 For each inferior thread, @value{GDBN} can obtain the branch trace
40115 configuration using the @samp{qXfer:btrace-conf:read}
40116 (@pxref{qXfer btrace-conf read}) packet.
40118 The configuration describes the branch trace format and configuration
40119 settings for that format. The following information is described:
40123 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
40126 The size of the @acronym{BTS} ring buffer in bytes.
40129 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
40133 The size of the @acronym{Intel PT} ring buffer in bytes.
40137 @value{GDBN} must be linked with the Expat library to support XML
40138 branch trace configuration discovery. @xref{Expat}.
40140 The formal DTD for the branch trace configuration format is given below:
40143 <!ELEMENT btrace-conf (bts?, pt?)>
40144 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
40146 <!ELEMENT bts EMPTY>
40147 <!ATTLIST bts size CDATA #IMPLIED>
40149 <!ELEMENT pt EMPTY>
40150 <!ATTLIST pt size CDATA #IMPLIED>
40153 @include agentexpr.texi
40155 @node Target Descriptions
40156 @appendix Target Descriptions
40157 @cindex target descriptions
40159 One of the challenges of using @value{GDBN} to debug embedded systems
40160 is that there are so many minor variants of each processor
40161 architecture in use. It is common practice for vendors to start with
40162 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
40163 and then make changes to adapt it to a particular market niche. Some
40164 architectures have hundreds of variants, available from dozens of
40165 vendors. This leads to a number of problems:
40169 With so many different customized processors, it is difficult for
40170 the @value{GDBN} maintainers to keep up with the changes.
40172 Since individual variants may have short lifetimes or limited
40173 audiences, it may not be worthwhile to carry information about every
40174 variant in the @value{GDBN} source tree.
40176 When @value{GDBN} does support the architecture of the embedded system
40177 at hand, the task of finding the correct architecture name to give the
40178 @command{set architecture} command can be error-prone.
40181 To address these problems, the @value{GDBN} remote protocol allows a
40182 target system to not only identify itself to @value{GDBN}, but to
40183 actually describe its own features. This lets @value{GDBN} support
40184 processor variants it has never seen before --- to the extent that the
40185 descriptions are accurate, and that @value{GDBN} understands them.
40187 @value{GDBN} must be linked with the Expat library to support XML
40188 target descriptions. @xref{Expat}.
40191 * Retrieving Descriptions:: How descriptions are fetched from a target.
40192 * Target Description Format:: The contents of a target description.
40193 * Predefined Target Types:: Standard types available for target
40195 * Standard Target Features:: Features @value{GDBN} knows about.
40198 @node Retrieving Descriptions
40199 @section Retrieving Descriptions
40201 Target descriptions can be read from the target automatically, or
40202 specified by the user manually. The default behavior is to read the
40203 description from the target. @value{GDBN} retrieves it via the remote
40204 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
40205 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
40206 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
40207 XML document, of the form described in @ref{Target Description
40210 Alternatively, you can specify a file to read for the target description.
40211 If a file is set, the target will not be queried. The commands to
40212 specify a file are:
40215 @cindex set tdesc filename
40216 @item set tdesc filename @var{path}
40217 Read the target description from @var{path}.
40219 @cindex unset tdesc filename
40220 @item unset tdesc filename
40221 Do not read the XML target description from a file. @value{GDBN}
40222 will use the description supplied by the current target.
40224 @cindex show tdesc filename
40225 @item show tdesc filename
40226 Show the filename to read for a target description, if any.
40230 @node Target Description Format
40231 @section Target Description Format
40232 @cindex target descriptions, XML format
40234 A target description annex is an @uref{http://www.w3.org/XML/, XML}
40235 document which complies with the Document Type Definition provided in
40236 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
40237 means you can use generally available tools like @command{xmllint} to
40238 check that your feature descriptions are well-formed and valid.
40239 However, to help people unfamiliar with XML write descriptions for
40240 their targets, we also describe the grammar here.
40242 Target descriptions can identify the architecture of the remote target
40243 and (for some architectures) provide information about custom register
40244 sets. They can also identify the OS ABI of the remote target.
40245 @value{GDBN} can use this information to autoconfigure for your
40246 target, or to warn you if you connect to an unsupported target.
40248 Here is a simple target description:
40251 <target version="1.0">
40252 <architecture>i386:x86-64</architecture>
40257 This minimal description only says that the target uses
40258 the x86-64 architecture.
40260 A target description has the following overall form, with [ ] marking
40261 optional elements and @dots{} marking repeatable elements. The elements
40262 are explained further below.
40265 <?xml version="1.0"?>
40266 <!DOCTYPE target SYSTEM "gdb-target.dtd">
40267 <target version="1.0">
40268 @r{[}@var{architecture}@r{]}
40269 @r{[}@var{osabi}@r{]}
40270 @r{[}@var{compatible}@r{]}
40271 @r{[}@var{feature}@dots{}@r{]}
40276 The description is generally insensitive to whitespace and line
40277 breaks, under the usual common-sense rules. The XML version
40278 declaration and document type declaration can generally be omitted
40279 (@value{GDBN} does not require them), but specifying them may be
40280 useful for XML validation tools. The @samp{version} attribute for
40281 @samp{<target>} may also be omitted, but we recommend
40282 including it; if future versions of @value{GDBN} use an incompatible
40283 revision of @file{gdb-target.dtd}, they will detect and report
40284 the version mismatch.
40286 @subsection Inclusion
40287 @cindex target descriptions, inclusion
40290 @cindex <xi:include>
40293 It can sometimes be valuable to split a target description up into
40294 several different annexes, either for organizational purposes, or to
40295 share files between different possible target descriptions. You can
40296 divide a description into multiple files by replacing any element of
40297 the target description with an inclusion directive of the form:
40300 <xi:include href="@var{document}"/>
40304 When @value{GDBN} encounters an element of this form, it will retrieve
40305 the named XML @var{document}, and replace the inclusion directive with
40306 the contents of that document. If the current description was read
40307 using @samp{qXfer}, then so will be the included document;
40308 @var{document} will be interpreted as the name of an annex. If the
40309 current description was read from a file, @value{GDBN} will look for
40310 @var{document} as a file in the same directory where it found the
40311 original description.
40313 @subsection Architecture
40314 @cindex <architecture>
40316 An @samp{<architecture>} element has this form:
40319 <architecture>@var{arch}</architecture>
40322 @var{arch} is one of the architectures from the set accepted by
40323 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
40326 @cindex @code{<osabi>}
40328 This optional field was introduced in @value{GDBN} version 7.0.
40329 Previous versions of @value{GDBN} ignore it.
40331 An @samp{<osabi>} element has this form:
40334 <osabi>@var{abi-name}</osabi>
40337 @var{abi-name} is an OS ABI name from the same selection accepted by
40338 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
40340 @subsection Compatible Architecture
40341 @cindex @code{<compatible>}
40343 This optional field was introduced in @value{GDBN} version 7.0.
40344 Previous versions of @value{GDBN} ignore it.
40346 A @samp{<compatible>} element has this form:
40349 <compatible>@var{arch}</compatible>
40352 @var{arch} is one of the architectures from the set accepted by
40353 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
40355 A @samp{<compatible>} element is used to specify that the target
40356 is able to run binaries in some other than the main target architecture
40357 given by the @samp{<architecture>} element. For example, on the
40358 Cell Broadband Engine, the main architecture is @code{powerpc:common}
40359 or @code{powerpc:common64}, but the system is able to run binaries
40360 in the @code{spu} architecture as well. The way to describe this
40361 capability with @samp{<compatible>} is as follows:
40364 <architecture>powerpc:common</architecture>
40365 <compatible>spu</compatible>
40368 @subsection Features
40371 Each @samp{<feature>} describes some logical portion of the target
40372 system. Features are currently used to describe available CPU
40373 registers and the types of their contents. A @samp{<feature>} element
40377 <feature name="@var{name}">
40378 @r{[}@var{type}@dots{}@r{]}
40384 Each feature's name should be unique within the description. The name
40385 of a feature does not matter unless @value{GDBN} has some special
40386 knowledge of the contents of that feature; if it does, the feature
40387 should have its standard name. @xref{Standard Target Features}.
40391 Any register's value is a collection of bits which @value{GDBN} must
40392 interpret. The default interpretation is a two's complement integer,
40393 but other types can be requested by name in the register description.
40394 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
40395 Target Types}), and the description can define additional composite types.
40397 Each type element must have an @samp{id} attribute, which gives
40398 a unique (within the containing @samp{<feature>}) name to the type.
40399 Types must be defined before they are used.
40402 Some targets offer vector registers, which can be treated as arrays
40403 of scalar elements. These types are written as @samp{<vector>} elements,
40404 specifying the array element type, @var{type}, and the number of elements,
40408 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
40412 If a register's value is usefully viewed in multiple ways, define it
40413 with a union type containing the useful representations. The
40414 @samp{<union>} element contains one or more @samp{<field>} elements,
40415 each of which has a @var{name} and a @var{type}:
40418 <union id="@var{id}">
40419 <field name="@var{name}" type="@var{type}"/>
40425 If a register's value is composed from several separate values, define
40426 it with a structure type. There are two forms of the @samp{<struct>}
40427 element; a @samp{<struct>} element must either contain only bitfields
40428 or contain no bitfields. If the structure contains only bitfields,
40429 its total size in bytes must be specified, each bitfield must have an
40430 explicit start and end, and bitfields are automatically assigned an
40431 integer type. The field's @var{start} should be less than or
40432 equal to its @var{end}, and zero represents the least significant bit.
40435 <struct id="@var{id}" size="@var{size}">
40436 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
40441 If the structure contains no bitfields, then each field has an
40442 explicit type, and no implicit padding is added.
40445 <struct id="@var{id}">
40446 <field name="@var{name}" type="@var{type}"/>
40452 If a register's value is a series of single-bit flags, define it with
40453 a flags type. The @samp{<flags>} element has an explicit @var{size}
40454 and contains one or more @samp{<field>} elements. Each field has a
40455 @var{name}, a @var{start}, and an @var{end}. Only single-bit flags
40459 <flags id="@var{id}" size="@var{size}">
40460 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
40465 @subsection Registers
40468 Each register is represented as an element with this form:
40471 <reg name="@var{name}"
40472 bitsize="@var{size}"
40473 @r{[}regnum="@var{num}"@r{]}
40474 @r{[}save-restore="@var{save-restore}"@r{]}
40475 @r{[}type="@var{type}"@r{]}
40476 @r{[}group="@var{group}"@r{]}/>
40480 The components are as follows:
40485 The register's name; it must be unique within the target description.
40488 The register's size, in bits.
40491 The register's number. If omitted, a register's number is one greater
40492 than that of the previous register (either in the current feature or in
40493 a preceding feature); the first register in the target description
40494 defaults to zero. This register number is used to read or write
40495 the register; e.g.@: it is used in the remote @code{p} and @code{P}
40496 packets, and registers appear in the @code{g} and @code{G} packets
40497 in order of increasing register number.
40500 Whether the register should be preserved across inferior function
40501 calls; this must be either @code{yes} or @code{no}. The default is
40502 @code{yes}, which is appropriate for most registers except for
40503 some system control registers; this is not related to the target's
40507 The type of the register. It may be a predefined type, a type
40508 defined in the current feature, or one of the special types @code{int}
40509 and @code{float}. @code{int} is an integer type of the correct size
40510 for @var{bitsize}, and @code{float} is a floating point type (in the
40511 architecture's normal floating point format) of the correct size for
40512 @var{bitsize}. The default is @code{int}.
40515 The register group to which this register belongs. It must
40516 be either @code{general}, @code{float}, or @code{vector}. If no
40517 @var{group} is specified, @value{GDBN} will not display the register
40518 in @code{info registers}.
40522 @node Predefined Target Types
40523 @section Predefined Target Types
40524 @cindex target descriptions, predefined types
40526 Type definitions in the self-description can build up composite types
40527 from basic building blocks, but can not define fundamental types. Instead,
40528 standard identifiers are provided by @value{GDBN} for the fundamental
40529 types. The currently supported types are:
40538 Signed integer types holding the specified number of bits.
40545 Unsigned integer types holding the specified number of bits.
40549 Pointers to unspecified code and data. The program counter and
40550 any dedicated return address register may be marked as code
40551 pointers; printing a code pointer converts it into a symbolic
40552 address. The stack pointer and any dedicated address registers
40553 may be marked as data pointers.
40556 Single precision IEEE floating point.
40559 Double precision IEEE floating point.
40562 The 12-byte extended precision format used by ARM FPA registers.
40565 The 10-byte extended precision format used by x87 registers.
40568 32bit @sc{eflags} register used by x86.
40571 32bit @sc{mxcsr} register used by x86.
40575 @node Standard Target Features
40576 @section Standard Target Features
40577 @cindex target descriptions, standard features
40579 A target description must contain either no registers or all the
40580 target's registers. If the description contains no registers, then
40581 @value{GDBN} will assume a default register layout, selected based on
40582 the architecture. If the description contains any registers, the
40583 default layout will not be used; the standard registers must be
40584 described in the target description, in such a way that @value{GDBN}
40585 can recognize them.
40587 This is accomplished by giving specific names to feature elements
40588 which contain standard registers. @value{GDBN} will look for features
40589 with those names and verify that they contain the expected registers;
40590 if any known feature is missing required registers, or if any required
40591 feature is missing, @value{GDBN} will reject the target
40592 description. You can add additional registers to any of the
40593 standard features --- @value{GDBN} will display them just as if
40594 they were added to an unrecognized feature.
40596 This section lists the known features and their expected contents.
40597 Sample XML documents for these features are included in the
40598 @value{GDBN} source tree, in the directory @file{gdb/features}.
40600 Names recognized by @value{GDBN} should include the name of the
40601 company or organization which selected the name, and the overall
40602 architecture to which the feature applies; so e.g.@: the feature
40603 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
40605 The names of registers are not case sensitive for the purpose
40606 of recognizing standard features, but @value{GDBN} will only display
40607 registers using the capitalization used in the description.
40610 * AArch64 Features::
40613 * MicroBlaze Features::
40616 * Nios II Features::
40617 * PowerPC Features::
40618 * S/390 and System z Features::
40623 @node AArch64 Features
40624 @subsection AArch64 Features
40625 @cindex target descriptions, AArch64 features
40627 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
40628 targets. It should contain registers @samp{x0} through @samp{x30},
40629 @samp{sp}, @samp{pc}, and @samp{cpsr}.
40631 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
40632 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
40636 @subsection ARM Features
40637 @cindex target descriptions, ARM features
40639 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
40641 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
40642 @samp{lr}, @samp{pc}, and @samp{cpsr}.
40644 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
40645 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
40646 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
40649 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
40650 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
40652 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
40653 it should contain at least registers @samp{wR0} through @samp{wR15} and
40654 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
40655 @samp{wCSSF}, and @samp{wCASF} registers are optional.
40657 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
40658 should contain at least registers @samp{d0} through @samp{d15}. If
40659 they are present, @samp{d16} through @samp{d31} should also be included.
40660 @value{GDBN} will synthesize the single-precision registers from
40661 halves of the double-precision registers.
40663 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
40664 need to contain registers; it instructs @value{GDBN} to display the
40665 VFP double-precision registers as vectors and to synthesize the
40666 quad-precision registers from pairs of double-precision registers.
40667 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
40668 be present and include 32 double-precision registers.
40670 @node i386 Features
40671 @subsection i386 Features
40672 @cindex target descriptions, i386 features
40674 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
40675 targets. It should describe the following registers:
40679 @samp{eax} through @samp{edi} plus @samp{eip} for i386
40681 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
40683 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
40684 @samp{fs}, @samp{gs}
40686 @samp{st0} through @samp{st7}
40688 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
40689 @samp{foseg}, @samp{fooff} and @samp{fop}
40692 The register sets may be different, depending on the target.
40694 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
40695 describe registers:
40699 @samp{xmm0} through @samp{xmm7} for i386
40701 @samp{xmm0} through @samp{xmm15} for amd64
40706 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
40707 @samp{org.gnu.gdb.i386.sse} feature. It should
40708 describe the upper 128 bits of @sc{ymm} registers:
40712 @samp{ymm0h} through @samp{ymm7h} for i386
40714 @samp{ymm0h} through @samp{ymm15h} for amd64
40717 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
40718 Memory Protection Extension (MPX). It should describe the following registers:
40722 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
40724 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
40727 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
40728 describe a single register, @samp{orig_eax}.
40730 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
40731 @samp{org.gnu.gdb.i386.avx} feature. It should
40732 describe additional @sc{xmm} registers:
40736 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
40739 It should describe the upper 128 bits of additional @sc{ymm} registers:
40743 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
40747 describe the upper 256 bits of @sc{zmm} registers:
40751 @samp{zmm0h} through @samp{zmm7h} for i386.
40753 @samp{zmm0h} through @samp{zmm15h} for amd64.
40757 describe the additional @sc{zmm} registers:
40761 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
40764 @node MicroBlaze Features
40765 @subsection MicroBlaze Features
40766 @cindex target descriptions, MicroBlaze features
40768 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
40769 targets. It should contain registers @samp{r0} through @samp{r31},
40770 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
40771 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
40772 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
40774 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
40775 If present, it should contain registers @samp{rshr} and @samp{rslr}
40777 @node MIPS Features
40778 @subsection @acronym{MIPS} Features
40779 @cindex target descriptions, @acronym{MIPS} features
40781 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
40782 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
40783 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
40786 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
40787 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
40788 registers. They may be 32-bit or 64-bit depending on the target.
40790 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
40791 it may be optional in a future version of @value{GDBN}. It should
40792 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
40793 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
40795 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
40796 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
40797 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
40798 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
40800 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
40801 contain a single register, @samp{restart}, which is used by the
40802 Linux kernel to control restartable syscalls.
40804 @node M68K Features
40805 @subsection M68K Features
40806 @cindex target descriptions, M68K features
40809 @item @samp{org.gnu.gdb.m68k.core}
40810 @itemx @samp{org.gnu.gdb.coldfire.core}
40811 @itemx @samp{org.gnu.gdb.fido.core}
40812 One of those features must be always present.
40813 The feature that is present determines which flavor of m68k is
40814 used. The feature that is present should contain registers
40815 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
40816 @samp{sp}, @samp{ps} and @samp{pc}.
40818 @item @samp{org.gnu.gdb.coldfire.fp}
40819 This feature is optional. If present, it should contain registers
40820 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
40824 @node Nios II Features
40825 @subsection Nios II Features
40826 @cindex target descriptions, Nios II features
40828 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
40829 targets. It should contain the 32 core registers (@samp{zero},
40830 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
40831 @samp{pc}, and the 16 control registers (@samp{status} through
40834 @node PowerPC Features
40835 @subsection PowerPC Features
40836 @cindex target descriptions, PowerPC features
40838 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
40839 targets. It should contain registers @samp{r0} through @samp{r31},
40840 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
40841 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
40843 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
40844 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
40846 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
40847 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
40850 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
40851 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
40852 will combine these registers with the floating point registers
40853 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
40854 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
40855 through @samp{vs63}, the set of vector registers for POWER7.
40857 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
40858 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
40859 @samp{spefscr}. SPE targets should provide 32-bit registers in
40860 @samp{org.gnu.gdb.power.core} and provide the upper halves in
40861 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
40862 these to present registers @samp{ev0} through @samp{ev31} to the
40865 @node S/390 and System z Features
40866 @subsection S/390 and System z Features
40867 @cindex target descriptions, S/390 features
40868 @cindex target descriptions, System z features
40870 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
40871 System z targets. It should contain the PSW and the 16 general
40872 registers. In particular, System z targets should provide the 64-bit
40873 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
40874 S/390 targets should provide the 32-bit versions of these registers.
40875 A System z target that runs in 31-bit addressing mode should provide
40876 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
40877 register's upper halves @samp{r0h} through @samp{r15h}, and their
40878 lower halves @samp{r0l} through @samp{r15l}.
40880 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
40881 contain the 64-bit registers @samp{f0} through @samp{f15}, and
40884 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
40885 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
40887 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
40888 contain the register @samp{orig_r2}, which is 64-bit wide on System z
40889 targets and 32-bit otherwise. In addition, the feature may contain
40890 the @samp{last_break} register, whose width depends on the addressing
40891 mode, as well as the @samp{system_call} register, which is always
40894 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
40895 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
40896 @samp{atia}, and @samp{tr0} through @samp{tr15}.
40898 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
40899 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
40900 combined by @value{GDBN} with the floating point registers @samp{f0}
40901 through @samp{f15} to present the 128-bit wide vector registers
40902 @samp{v0} through @samp{v15}. In addition, this feature should
40903 contain the 128-bit wide vector registers @samp{v16} through
40906 @node TIC6x Features
40907 @subsection TMS320C6x Features
40908 @cindex target descriptions, TIC6x features
40909 @cindex target descriptions, TMS320C6x features
40910 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
40911 targets. It should contain registers @samp{A0} through @samp{A15},
40912 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
40914 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
40915 contain registers @samp{A16} through @samp{A31} and @samp{B16}
40916 through @samp{B31}.
40918 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
40919 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
40921 @node Operating System Information
40922 @appendix Operating System Information
40923 @cindex operating system information
40929 Users of @value{GDBN} often wish to obtain information about the state of
40930 the operating system running on the target---for example the list of
40931 processes, or the list of open files. This section describes the
40932 mechanism that makes it possible. This mechanism is similar to the
40933 target features mechanism (@pxref{Target Descriptions}), but focuses
40934 on a different aspect of target.
40936 Operating system information is retrived from the target via the
40937 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
40938 read}). The object name in the request should be @samp{osdata}, and
40939 the @var{annex} identifies the data to be fetched.
40942 @appendixsection Process list
40943 @cindex operating system information, process list
40945 When requesting the process list, the @var{annex} field in the
40946 @samp{qXfer} request should be @samp{processes}. The returned data is
40947 an XML document. The formal syntax of this document is defined in
40948 @file{gdb/features/osdata.dtd}.
40950 An example document is:
40953 <?xml version="1.0"?>
40954 <!DOCTYPE target SYSTEM "osdata.dtd">
40955 <osdata type="processes">
40957 <column name="pid">1</column>
40958 <column name="user">root</column>
40959 <column name="command">/sbin/init</column>
40960 <column name="cores">1,2,3</column>
40965 Each item should include a column whose name is @samp{pid}. The value
40966 of that column should identify the process on the target. The
40967 @samp{user} and @samp{command} columns are optional, and will be
40968 displayed by @value{GDBN}. The @samp{cores} column, if present,
40969 should contain a comma-separated list of cores that this process
40970 is running on. Target may provide additional columns,
40971 which @value{GDBN} currently ignores.
40973 @node Trace File Format
40974 @appendix Trace File Format
40975 @cindex trace file format
40977 The trace file comes in three parts: a header, a textual description
40978 section, and a trace frame section with binary data.
40980 The header has the form @code{\x7fTRACE0\n}. The first byte is
40981 @code{0x7f} so as to indicate that the file contains binary data,
40982 while the @code{0} is a version number that may have different values
40985 The description section consists of multiple lines of @sc{ascii} text
40986 separated by newline characters (@code{0xa}). The lines may include a
40987 variety of optional descriptive or context-setting information, such
40988 as tracepoint definitions or register set size. @value{GDBN} will
40989 ignore any line that it does not recognize. An empty line marks the end
40992 @c FIXME add some specific types of data
40994 The trace frame section consists of a number of consecutive frames.
40995 Each frame begins with a two-byte tracepoint number, followed by a
40996 four-byte size giving the amount of data in the frame. The data in
40997 the frame consists of a number of blocks, each introduced by a
40998 character indicating its type (at least register, memory, and trace
40999 state variable). The data in this section is raw binary, not a
41000 hexadecimal or other encoding; its endianness matches the target's
41003 @c FIXME bi-arch may require endianness/arch info in description section
41006 @item R @var{bytes}
41007 Register block. The number and ordering of bytes matches that of a
41008 @code{g} packet in the remote protocol. Note that these are the
41009 actual bytes, in target order and @value{GDBN} register order, not a
41010 hexadecimal encoding.
41012 @item M @var{address} @var{length} @var{bytes}...
41013 Memory block. This is a contiguous block of memory, at the 8-byte
41014 address @var{address}, with a 2-byte length @var{length}, followed by
41015 @var{length} bytes.
41017 @item V @var{number} @var{value}
41018 Trace state variable block. This records the 8-byte signed value
41019 @var{value} of trace state variable numbered @var{number}.
41023 Future enhancements of the trace file format may include additional types
41026 @node Index Section Format
41027 @appendix @code{.gdb_index} section format
41028 @cindex .gdb_index section format
41029 @cindex index section format
41031 This section documents the index section that is created by @code{save
41032 gdb-index} (@pxref{Index Files}). The index section is
41033 DWARF-specific; some knowledge of DWARF is assumed in this
41036 The mapped index file format is designed to be directly
41037 @code{mmap}able on any architecture. In most cases, a datum is
41038 represented using a little-endian 32-bit integer value, called an
41039 @code{offset_type}. Big endian machines must byte-swap the values
41040 before using them. Exceptions to this rule are noted. The data is
41041 laid out such that alignment is always respected.
41043 A mapped index consists of several areas, laid out in order.
41047 The file header. This is a sequence of values, of @code{offset_type}
41048 unless otherwise noted:
41052 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
41053 Version 4 uses a different hashing function from versions 5 and 6.
41054 Version 6 includes symbols for inlined functions, whereas versions 4
41055 and 5 do not. Version 7 adds attributes to the CU indices in the
41056 symbol table. Version 8 specifies that symbols from DWARF type units
41057 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
41058 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
41060 @value{GDBN} will only read version 4, 5, or 6 indices
41061 by specifying @code{set use-deprecated-index-sections on}.
41062 GDB has a workaround for potentially broken version 7 indices so it is
41063 currently not flagged as deprecated.
41066 The offset, from the start of the file, of the CU list.
41069 The offset, from the start of the file, of the types CU list. Note
41070 that this area can be empty, in which case this offset will be equal
41071 to the next offset.
41074 The offset, from the start of the file, of the address area.
41077 The offset, from the start of the file, of the symbol table.
41080 The offset, from the start of the file, of the constant pool.
41084 The CU list. This is a sequence of pairs of 64-bit little-endian
41085 values, sorted by the CU offset. The first element in each pair is
41086 the offset of a CU in the @code{.debug_info} section. The second
41087 element in each pair is the length of that CU. References to a CU
41088 elsewhere in the map are done using a CU index, which is just the
41089 0-based index into this table. Note that if there are type CUs, then
41090 conceptually CUs and type CUs form a single list for the purposes of
41094 The types CU list. This is a sequence of triplets of 64-bit
41095 little-endian values. In a triplet, the first value is the CU offset,
41096 the second value is the type offset in the CU, and the third value is
41097 the type signature. The types CU list is not sorted.
41100 The address area. The address area consists of a sequence of address
41101 entries. Each address entry has three elements:
41105 The low address. This is a 64-bit little-endian value.
41108 The high address. This is a 64-bit little-endian value. Like
41109 @code{DW_AT_high_pc}, the value is one byte beyond the end.
41112 The CU index. This is an @code{offset_type} value.
41116 The symbol table. This is an open-addressed hash table. The size of
41117 the hash table is always a power of 2.
41119 Each slot in the hash table consists of a pair of @code{offset_type}
41120 values. The first value is the offset of the symbol's name in the
41121 constant pool. The second value is the offset of the CU vector in the
41124 If both values are 0, then this slot in the hash table is empty. This
41125 is ok because while 0 is a valid constant pool index, it cannot be a
41126 valid index for both a string and a CU vector.
41128 The hash value for a table entry is computed by applying an
41129 iterative hash function to the symbol's name. Starting with an
41130 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
41131 the string is incorporated into the hash using the formula depending on the
41136 The formula is @code{r = r * 67 + c - 113}.
41138 @item Versions 5 to 7
41139 The formula is @code{r = r * 67 + tolower (c) - 113}.
41142 The terminating @samp{\0} is not incorporated into the hash.
41144 The step size used in the hash table is computed via
41145 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
41146 value, and @samp{size} is the size of the hash table. The step size
41147 is used to find the next candidate slot when handling a hash
41150 The names of C@t{++} symbols in the hash table are canonicalized. We
41151 don't currently have a simple description of the canonicalization
41152 algorithm; if you intend to create new index sections, you must read
41156 The constant pool. This is simply a bunch of bytes. It is organized
41157 so that alignment is correct: CU vectors are stored first, followed by
41160 A CU vector in the constant pool is a sequence of @code{offset_type}
41161 values. The first value is the number of CU indices in the vector.
41162 Each subsequent value is the index and symbol attributes of a CU in
41163 the CU list. This element in the hash table is used to indicate which
41164 CUs define the symbol and how the symbol is used.
41165 See below for the format of each CU index+attributes entry.
41167 A string in the constant pool is zero-terminated.
41170 Attributes were added to CU index values in @code{.gdb_index} version 7.
41171 If a symbol has multiple uses within a CU then there is one
41172 CU index+attributes value for each use.
41174 The format of each CU index+attributes entry is as follows
41180 This is the index of the CU in the CU list.
41182 These bits are reserved for future purposes and must be zero.
41184 The kind of the symbol in the CU.
41188 This value is reserved and should not be used.
41189 By reserving zero the full @code{offset_type} value is backwards compatible
41190 with previous versions of the index.
41192 The symbol is a type.
41194 The symbol is a variable or an enum value.
41196 The symbol is a function.
41198 Any other kind of symbol.
41200 These values are reserved.
41204 This bit is zero if the value is global and one if it is static.
41206 The determination of whether a symbol is global or static is complicated.
41207 The authorative reference is the file @file{dwarf2read.c} in
41208 @value{GDBN} sources.
41212 This pseudo-code describes the computation of a symbol's kind and
41213 global/static attributes in the index.
41216 is_external = get_attribute (die, DW_AT_external);
41217 language = get_attribute (cu_die, DW_AT_language);
41220 case DW_TAG_typedef:
41221 case DW_TAG_base_type:
41222 case DW_TAG_subrange_type:
41226 case DW_TAG_enumerator:
41228 is_static = (language != CPLUS && language != JAVA);
41230 case DW_TAG_subprogram:
41232 is_static = ! (is_external || language == ADA);
41234 case DW_TAG_constant:
41236 is_static = ! is_external;
41238 case DW_TAG_variable:
41240 is_static = ! is_external;
41242 case DW_TAG_namespace:
41246 case DW_TAG_class_type:
41247 case DW_TAG_interface_type:
41248 case DW_TAG_structure_type:
41249 case DW_TAG_union_type:
41250 case DW_TAG_enumeration_type:
41252 is_static = (language != CPLUS && language != JAVA);
41260 @appendix Manual pages
41264 * gdb man:: The GNU Debugger man page
41265 * gdbserver man:: Remote Server for the GNU Debugger man page
41266 * gcore man:: Generate a core file of a running program
41267 * gdbinit man:: gdbinit scripts
41273 @c man title gdb The GNU Debugger
41275 @c man begin SYNOPSIS gdb
41276 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
41277 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
41278 [@option{-b}@w{ }@var{bps}]
41279 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
41280 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
41281 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
41282 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
41283 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
41286 @c man begin DESCRIPTION gdb
41287 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
41288 going on ``inside'' another program while it executes -- or what another
41289 program was doing at the moment it crashed.
41291 @value{GDBN} can do four main kinds of things (plus other things in support of
41292 these) to help you catch bugs in the act:
41296 Start your program, specifying anything that might affect its behavior.
41299 Make your program stop on specified conditions.
41302 Examine what has happened, when your program has stopped.
41305 Change things in your program, so you can experiment with correcting the
41306 effects of one bug and go on to learn about another.
41309 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
41312 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
41313 commands from the terminal until you tell it to exit with the @value{GDBN}
41314 command @code{quit}. You can get online help from @value{GDBN} itself
41315 by using the command @code{help}.
41317 You can run @code{gdb} with no arguments or options; but the most
41318 usual way to start @value{GDBN} is with one argument or two, specifying an
41319 executable program as the argument:
41325 You can also start with both an executable program and a core file specified:
41331 You can, instead, specify a process ID as a second argument, if you want
41332 to debug a running process:
41340 would attach @value{GDBN} to process @code{1234} (unless you also have a file
41341 named @file{1234}; @value{GDBN} does check for a core file first).
41342 With option @option{-p} you can omit the @var{program} filename.
41344 Here are some of the most frequently needed @value{GDBN} commands:
41346 @c pod2man highlights the right hand side of the @item lines.
41348 @item break [@var{file}:]@var{functiop}
41349 Set a breakpoint at @var{function} (in @var{file}).
41351 @item run [@var{arglist}]
41352 Start your program (with @var{arglist}, if specified).
41355 Backtrace: display the program stack.
41357 @item print @var{expr}
41358 Display the value of an expression.
41361 Continue running your program (after stopping, e.g. at a breakpoint).
41364 Execute next program line (after stopping); step @emph{over} any
41365 function calls in the line.
41367 @item edit [@var{file}:]@var{function}
41368 look at the program line where it is presently stopped.
41370 @item list [@var{file}:]@var{function}
41371 type the text of the program in the vicinity of where it is presently stopped.
41374 Execute next program line (after stopping); step @emph{into} any
41375 function calls in the line.
41377 @item help [@var{name}]
41378 Show information about @value{GDBN} command @var{name}, or general information
41379 about using @value{GDBN}.
41382 Exit from @value{GDBN}.
41386 For full details on @value{GDBN},
41387 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
41388 by Richard M. Stallman and Roland H. Pesch. The same text is available online
41389 as the @code{gdb} entry in the @code{info} program.
41393 @c man begin OPTIONS gdb
41394 Any arguments other than options specify an executable
41395 file and core file (or process ID); that is, the first argument
41396 encountered with no
41397 associated option flag is equivalent to a @option{-se} option, and the second,
41398 if any, is equivalent to a @option{-c} option if it's the name of a file.
41400 both long and short forms; both are shown here. The long forms are also
41401 recognized if you truncate them, so long as enough of the option is
41402 present to be unambiguous. (If you prefer, you can flag option
41403 arguments with @option{+} rather than @option{-}, though we illustrate the
41404 more usual convention.)
41406 All the options and command line arguments you give are processed
41407 in sequential order. The order makes a difference when the @option{-x}
41413 List all options, with brief explanations.
41415 @item -symbols=@var{file}
41416 @itemx -s @var{file}
41417 Read symbol table from file @var{file}.
41420 Enable writing into executable and core files.
41422 @item -exec=@var{file}
41423 @itemx -e @var{file}
41424 Use file @var{file} as the executable file to execute when
41425 appropriate, and for examining pure data in conjunction with a core
41428 @item -se=@var{file}
41429 Read symbol table from file @var{file} and use it as the executable
41432 @item -core=@var{file}
41433 @itemx -c @var{file}
41434 Use file @var{file} as a core dump to examine.
41436 @item -command=@var{file}
41437 @itemx -x @var{file}
41438 Execute @value{GDBN} commands from file @var{file}.
41440 @item -ex @var{command}
41441 Execute given @value{GDBN} @var{command}.
41443 @item -directory=@var{directory}
41444 @itemx -d @var{directory}
41445 Add @var{directory} to the path to search for source files.
41448 Do not execute commands from @file{~/.gdbinit}.
41452 Do not execute commands from any @file{.gdbinit} initialization files.
41456 ``Quiet''. Do not print the introductory and copyright messages. These
41457 messages are also suppressed in batch mode.
41460 Run in batch mode. Exit with status @code{0} after processing all the command
41461 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
41462 Exit with nonzero status if an error occurs in executing the @value{GDBN}
41463 commands in the command files.
41465 Batch mode may be useful for running @value{GDBN} as a filter, for example to
41466 download and run a program on another computer; in order to make this
41467 more useful, the message
41470 Program exited normally.
41474 (which is ordinarily issued whenever a program running under @value{GDBN} control
41475 terminates) is not issued when running in batch mode.
41477 @item -cd=@var{directory}
41478 Run @value{GDBN} using @var{directory} as its working directory,
41479 instead of the current directory.
41483 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
41484 @value{GDBN} to output the full file name and line number in a standard,
41485 recognizable fashion each time a stack frame is displayed (which
41486 includes each time the program stops). This recognizable format looks
41487 like two @samp{\032} characters, followed by the file name, line number
41488 and character position separated by colons, and a newline. The
41489 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
41490 characters as a signal to display the source code for the frame.
41493 Set the line speed (baud rate or bits per second) of any serial
41494 interface used by @value{GDBN} for remote debugging.
41496 @item -tty=@var{device}
41497 Run using @var{device} for your program's standard input and output.
41501 @c man begin SEEALSO gdb
41503 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
41504 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
41505 documentation are properly installed at your site, the command
41512 should give you access to the complete manual.
41514 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
41515 Richard M. Stallman and Roland H. Pesch, July 1991.
41519 @node gdbserver man
41520 @heading gdbserver man
41522 @c man title gdbserver Remote Server for the GNU Debugger
41524 @c man begin SYNOPSIS gdbserver
41525 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
41527 gdbserver --attach @var{comm} @var{pid}
41529 gdbserver --multi @var{comm}
41533 @c man begin DESCRIPTION gdbserver
41534 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
41535 than the one which is running the program being debugged.
41538 @subheading Usage (server (target) side)
41541 Usage (server (target) side):
41544 First, you need to have a copy of the program you want to debug put onto
41545 the target system. The program can be stripped to save space if needed, as
41546 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
41547 the @value{GDBN} running on the host system.
41549 To use the server, you log on to the target system, and run the @command{gdbserver}
41550 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
41551 your program, and (c) its arguments. The general syntax is:
41554 target> gdbserver @var{comm} @var{program} [@var{args} ...]
41557 For example, using a serial port, you might say:
41561 @c @file would wrap it as F</dev/com1>.
41562 target> gdbserver /dev/com1 emacs foo.txt
41565 target> gdbserver @file{/dev/com1} emacs foo.txt
41569 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
41570 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
41571 waits patiently for the host @value{GDBN} to communicate with it.
41573 To use a TCP connection, you could say:
41576 target> gdbserver host:2345 emacs foo.txt
41579 This says pretty much the same thing as the last example, except that we are
41580 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
41581 that we are expecting to see a TCP connection from @code{host} to local TCP port
41582 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
41583 want for the port number as long as it does not conflict with any existing TCP
41584 ports on the target system. This same port number must be used in the host
41585 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
41586 you chose a port number that conflicts with another service, @command{gdbserver} will
41587 print an error message and exit.
41589 @command{gdbserver} can also attach to running programs.
41590 This is accomplished via the @option{--attach} argument. The syntax is:
41593 target> gdbserver --attach @var{comm} @var{pid}
41596 @var{pid} is the process ID of a currently running process. It isn't
41597 necessary to point @command{gdbserver} at a binary for the running process.
41599 To start @code{gdbserver} without supplying an initial command to run
41600 or process ID to attach, use the @option{--multi} command line option.
41601 In such case you should connect using @kbd{target extended-remote} to start
41602 the program you want to debug.
41605 target> gdbserver --multi @var{comm}
41609 @subheading Usage (host side)
41615 You need an unstripped copy of the target program on your host system, since
41616 @value{GDBN} needs to examine it's symbol tables and such. Start up @value{GDBN} as you normally
41617 would, with the target program as the first argument. (You may need to use the
41618 @option{--baud} option if the serial line is running at anything except 9600 baud.)
41619 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
41620 new command you need to know about is @code{target remote}
41621 (or @code{target extended-remote}). Its argument is either
41622 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
41623 descriptor. For example:
41627 @c @file would wrap it as F</dev/ttyb>.
41628 (gdb) target remote /dev/ttyb
41631 (gdb) target remote @file{/dev/ttyb}
41636 communicates with the server via serial line @file{/dev/ttyb}, and:
41639 (gdb) target remote the-target:2345
41643 communicates via a TCP connection to port 2345 on host `the-target', where
41644 you previously started up @command{gdbserver} with the same port number. Note that for
41645 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
41646 command, otherwise you may get an error that looks something like
41647 `Connection refused'.
41649 @command{gdbserver} can also debug multiple inferiors at once,
41652 the @value{GDBN} manual in node @code{Inferiors and Programs}
41653 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
41656 @ref{Inferiors and Programs}.
41658 In such case use the @code{extended-remote} @value{GDBN} command variant:
41661 (gdb) target extended-remote the-target:2345
41664 The @command{gdbserver} option @option{--multi} may or may not be used in such
41668 @c man begin OPTIONS gdbserver
41669 There are three different modes for invoking @command{gdbserver}:
41674 Debug a specific program specified by its program name:
41677 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
41680 The @var{comm} parameter specifies how should the server communicate
41681 with @value{GDBN}; it is either a device name (to use a serial line),
41682 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
41683 stdin/stdout of @code{gdbserver}. Specify the name of the program to
41684 debug in @var{prog}. Any remaining arguments will be passed to the
41685 program verbatim. When the program exits, @value{GDBN} will close the
41686 connection, and @code{gdbserver} will exit.
41689 Debug a specific program by specifying the process ID of a running
41693 gdbserver --attach @var{comm} @var{pid}
41696 The @var{comm} parameter is as described above. Supply the process ID
41697 of a running program in @var{pid}; @value{GDBN} will do everything
41698 else. Like with the previous mode, when the process @var{pid} exits,
41699 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
41702 Multi-process mode -- debug more than one program/process:
41705 gdbserver --multi @var{comm}
41708 In this mode, @value{GDBN} can instruct @command{gdbserver} which
41709 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
41710 close the connection when a process being debugged exits, so you can
41711 debug several processes in the same session.
41714 In each of the modes you may specify these options:
41719 List all options, with brief explanations.
41722 This option causes @command{gdbserver} to print its version number and exit.
41725 @command{gdbserver} will attach to a running program. The syntax is:
41728 target> gdbserver --attach @var{comm} @var{pid}
41731 @var{pid} is the process ID of a currently running process. It isn't
41732 necessary to point @command{gdbserver} at a binary for the running process.
41735 To start @code{gdbserver} without supplying an initial command to run
41736 or process ID to attach, use this command line option.
41737 Then you can connect using @kbd{target extended-remote} and start
41738 the program you want to debug. The syntax is:
41741 target> gdbserver --multi @var{comm}
41745 Instruct @code{gdbserver} to display extra status information about the debugging
41747 This option is intended for @code{gdbserver} development and for bug reports to
41750 @item --remote-debug
41751 Instruct @code{gdbserver} to display remote protocol debug output.
41752 This option is intended for @code{gdbserver} development and for bug reports to
41755 @item --debug-format=option1@r{[},option2,...@r{]}
41756 Instruct @code{gdbserver} to include extra information in each line
41757 of debugging output.
41758 @xref{Other Command-Line Arguments for gdbserver}.
41761 Specify a wrapper to launch programs
41762 for debugging. The option should be followed by the name of the
41763 wrapper, then any command-line arguments to pass to the wrapper, then
41764 @kbd{--} indicating the end of the wrapper arguments.
41767 By default, @command{gdbserver} keeps the listening TCP port open, so that
41768 additional connections are possible. However, if you start @code{gdbserver}
41769 with the @option{--once} option, it will stop listening for any further
41770 connection attempts after connecting to the first @value{GDBN} session.
41772 @c --disable-packet is not documented for users.
41774 @c --disable-randomization and --no-disable-randomization are superseded by
41775 @c QDisableRandomization.
41780 @c man begin SEEALSO gdbserver
41782 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
41783 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
41784 documentation are properly installed at your site, the command
41790 should give you access to the complete manual.
41792 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
41793 Richard M. Stallman and Roland H. Pesch, July 1991.
41800 @c man title gcore Generate a core file of a running program
41803 @c man begin SYNOPSIS gcore
41804 gcore [-o @var{filename}] @var{pid}
41808 @c man begin DESCRIPTION gcore
41809 Generate a core dump of a running program with process ID @var{pid}.
41810 Produced file is equivalent to a kernel produced core file as if the process
41811 crashed (and if @kbd{ulimit -c} were used to set up an appropriate core dump
41812 limit). Unlike after a crash, after @command{gcore} the program remains
41813 running without any change.
41816 @c man begin OPTIONS gcore
41818 @item -o @var{filename}
41819 The optional argument
41820 @var{filename} specifies the file name where to put the core dump.
41821 If not specified, the file name defaults to @file{core.@var{pid}},
41822 where @var{pid} is the running program process ID.
41826 @c man begin SEEALSO gcore
41828 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
41829 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
41830 documentation are properly installed at your site, the command
41837 should give you access to the complete manual.
41839 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
41840 Richard M. Stallman and Roland H. Pesch, July 1991.
41847 @c man title gdbinit GDB initialization scripts
41850 @c man begin SYNOPSIS gdbinit
41851 @ifset SYSTEM_GDBINIT
41852 @value{SYSTEM_GDBINIT}
41861 @c man begin DESCRIPTION gdbinit
41862 These files contain @value{GDBN} commands to automatically execute during
41863 @value{GDBN} startup. The lines of contents are canned sequences of commands,
41866 the @value{GDBN} manual in node @code{Sequences}
41867 -- shell command @code{info -f gdb -n Sequences}.
41873 Please read more in
41875 the @value{GDBN} manual in node @code{Startup}
41876 -- shell command @code{info -f gdb -n Startup}.
41883 @ifset SYSTEM_GDBINIT
41884 @item @value{SYSTEM_GDBINIT}
41886 @ifclear SYSTEM_GDBINIT
41887 @item (not enabled with @code{--with-system-gdbinit} during compilation)
41889 System-wide initialization file. It is executed unless user specified
41890 @value{GDBN} option @code{-nx} or @code{-n}.
41893 the @value{GDBN} manual in node @code{System-wide configuration}
41894 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
41897 @ref{System-wide configuration}.
41901 User initialization file. It is executed unless user specified
41902 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
41905 Initialization file for current directory. It may need to be enabled with
41906 @value{GDBN} security command @code{set auto-load local-gdbinit}.
41909 the @value{GDBN} manual in node @code{Init File in the Current Directory}
41910 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
41913 @ref{Init File in the Current Directory}.
41918 @c man begin SEEALSO gdbinit
41920 gdb(1), @code{info -f gdb -n Startup}
41922 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
41923 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
41924 documentation are properly installed at your site, the command
41930 should give you access to the complete manual.
41932 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
41933 Richard M. Stallman and Roland H. Pesch, July 1991.
41939 @node GNU Free Documentation License
41940 @appendix GNU Free Documentation License
41943 @node Concept Index
41944 @unnumbered Concept Index
41948 @node Command and Variable Index
41949 @unnumbered Command, Variable, and Function Index
41954 % I think something like @@colophon should be in texinfo. In the
41956 \long\def\colophon{\hbox to0pt{}\vfill
41957 \centerline{The body of this manual is set in}
41958 \centerline{\fontname\tenrm,}
41959 \centerline{with headings in {\bf\fontname\tenbf}}
41960 \centerline{and examples in {\tt\fontname\tentt}.}
41961 \centerline{{\it\fontname\tenit\/},}
41962 \centerline{{\bf\fontname\tenbf}, and}
41963 \centerline{{\sl\fontname\tensl\/}}
41964 \centerline{are used for emphasis.}\vfill}
41966 % Blame: doc@@cygnus.com, 1991.