1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988-2015 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{} 1988-2015 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) 1988-2015 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}).
1239 @c @cindex @code{--xdb}
1240 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1241 @c For information, see the file @file{xdb_trans.html}, which is usually
1242 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1245 @item -interpreter @var{interp}
1246 @cindex @code{--interpreter}
1247 Use the interpreter @var{interp} for interface with the controlling
1248 program or device. This option is meant to be set by programs which
1249 communicate with @value{GDBN} using it as a back end.
1250 @xref{Interpreters, , Command Interpreters}.
1252 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1253 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1254 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1255 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1256 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1257 @sc{gdb/mi} interfaces are no longer supported.
1260 @cindex @code{--write}
1261 Open the executable and core files for both reading and writing. This
1262 is equivalent to the @samp{set write on} command inside @value{GDBN}
1266 @cindex @code{--statistics}
1267 This option causes @value{GDBN} to print statistics about time and
1268 memory usage after it completes each command and returns to the prompt.
1271 @cindex @code{--version}
1272 This option causes @value{GDBN} to print its version number and
1273 no-warranty blurb, and exit.
1275 @item -configuration
1276 @cindex @code{--configuration}
1277 This option causes @value{GDBN} to print details about its build-time
1278 configuration parameters, and then exit. These details can be
1279 important when reporting @value{GDBN} bugs (@pxref{GDB Bugs}).
1284 @subsection What @value{GDBN} Does During Startup
1285 @cindex @value{GDBN} startup
1287 Here's the description of what @value{GDBN} does during session startup:
1291 Sets up the command interpreter as specified by the command line
1292 (@pxref{Mode Options, interpreter}).
1296 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1297 used when building @value{GDBN}; @pxref{System-wide configuration,
1298 ,System-wide configuration and settings}) and executes all the commands in
1301 @anchor{Home Directory Init File}
1303 Reads the init file (if any) in your home directory@footnote{On
1304 DOS/Windows systems, the home directory is the one pointed to by the
1305 @code{HOME} environment variable.} and executes all the commands in
1308 @anchor{Option -init-eval-command}
1310 Executes commands and command files specified by the @samp{-iex} and
1311 @samp{-ix} options in their specified order. Usually you should use the
1312 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1313 settings before @value{GDBN} init files get executed and before inferior
1317 Processes command line options and operands.
1319 @anchor{Init File in the Current Directory during Startup}
1321 Reads and executes the commands from init file (if any) in the current
1322 working directory as long as @samp{set auto-load local-gdbinit} is set to
1323 @samp{on} (@pxref{Init File in the Current Directory}).
1324 This is only done if the current directory is
1325 different from your home directory. Thus, you can have more than one
1326 init file, one generic in your home directory, and another, specific
1327 to the program you are debugging, in the directory where you invoke
1331 If the command line specified a program to debug, or a process to
1332 attach to, or a core file, @value{GDBN} loads any auto-loaded
1333 scripts provided for the program or for its loaded shared libraries.
1334 @xref{Auto-loading}.
1336 If you wish to disable the auto-loading during startup,
1337 you must do something like the following:
1340 $ gdb -iex "set auto-load python-scripts off" myprogram
1343 Option @samp{-ex} does not work because the auto-loading is then turned
1347 Executes commands and command files specified by the @samp{-ex} and
1348 @samp{-x} options in their specified order. @xref{Command Files}, for
1349 more details about @value{GDBN} command files.
1352 Reads the command history recorded in the @dfn{history file}.
1353 @xref{Command History}, for more details about the command history and the
1354 files where @value{GDBN} records it.
1357 Init files use the same syntax as @dfn{command files} (@pxref{Command
1358 Files}) and are processed by @value{GDBN} in the same way. The init
1359 file in your home directory can set options (such as @samp{set
1360 complaints}) that affect subsequent processing of command line options
1361 and operands. Init files are not executed if you use the @samp{-nx}
1362 option (@pxref{Mode Options, ,Choosing Modes}).
1364 To display the list of init files loaded by gdb at startup, you
1365 can use @kbd{gdb --help}.
1367 @cindex init file name
1368 @cindex @file{.gdbinit}
1369 @cindex @file{gdb.ini}
1370 The @value{GDBN} init files are normally called @file{.gdbinit}.
1371 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1372 the limitations of file names imposed by DOS filesystems. The Windows
1373 port of @value{GDBN} uses the standard name, but if it finds a
1374 @file{gdb.ini} file in your home directory, it warns you about that
1375 and suggests to rename the file to the standard name.
1379 @section Quitting @value{GDBN}
1380 @cindex exiting @value{GDBN}
1381 @cindex leaving @value{GDBN}
1384 @kindex quit @r{[}@var{expression}@r{]}
1385 @kindex q @r{(@code{quit})}
1386 @item quit @r{[}@var{expression}@r{]}
1388 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1389 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1390 do not supply @var{expression}, @value{GDBN} will terminate normally;
1391 otherwise it will terminate using the result of @var{expression} as the
1396 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1397 terminates the action of any @value{GDBN} command that is in progress and
1398 returns to @value{GDBN} command level. It is safe to type the interrupt
1399 character at any time because @value{GDBN} does not allow it to take effect
1400 until a time when it is safe.
1402 If you have been using @value{GDBN} to control an attached process or
1403 device, you can release it with the @code{detach} command
1404 (@pxref{Attach, ,Debugging an Already-running Process}).
1406 @node Shell Commands
1407 @section Shell Commands
1409 If you need to execute occasional shell commands during your
1410 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1411 just use the @code{shell} command.
1416 @cindex shell escape
1417 @item shell @var{command-string}
1418 @itemx !@var{command-string}
1419 Invoke a standard shell to execute @var{command-string}.
1420 Note that no space is needed between @code{!} and @var{command-string}.
1421 If it exists, the environment variable @code{SHELL} determines which
1422 shell to run. Otherwise @value{GDBN} uses the default shell
1423 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1426 The utility @code{make} is often needed in development environments.
1427 You do not have to use the @code{shell} command for this purpose in
1432 @cindex calling make
1433 @item make @var{make-args}
1434 Execute the @code{make} program with the specified
1435 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1438 @node Logging Output
1439 @section Logging Output
1440 @cindex logging @value{GDBN} output
1441 @cindex save @value{GDBN} output to a file
1443 You may want to save the output of @value{GDBN} commands to a file.
1444 There are several commands to control @value{GDBN}'s logging.
1448 @item set logging on
1450 @item set logging off
1452 @cindex logging file name
1453 @item set logging file @var{file}
1454 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1455 @item set logging overwrite [on|off]
1456 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1457 you want @code{set logging on} to overwrite the logfile instead.
1458 @item set logging redirect [on|off]
1459 By default, @value{GDBN} output will go to both the terminal and the logfile.
1460 Set @code{redirect} if you want output to go only to the log file.
1461 @kindex show logging
1463 Show the current values of the logging settings.
1467 @chapter @value{GDBN} Commands
1469 You can abbreviate a @value{GDBN} command to the first few letters of the command
1470 name, if that abbreviation is unambiguous; and you can repeat certain
1471 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1472 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1473 show you the alternatives available, if there is more than one possibility).
1476 * Command Syntax:: How to give commands to @value{GDBN}
1477 * Completion:: Command completion
1478 * Help:: How to ask @value{GDBN} for help
1481 @node Command Syntax
1482 @section Command Syntax
1484 A @value{GDBN} command is a single line of input. There is no limit on
1485 how long it can be. It starts with a command name, which is followed by
1486 arguments whose meaning depends on the command name. For example, the
1487 command @code{step} accepts an argument which is the number of times to
1488 step, as in @samp{step 5}. You can also use the @code{step} command
1489 with no arguments. Some commands do not allow any arguments.
1491 @cindex abbreviation
1492 @value{GDBN} command names may always be truncated if that abbreviation is
1493 unambiguous. Other possible command abbreviations are listed in the
1494 documentation for individual commands. In some cases, even ambiguous
1495 abbreviations are allowed; for example, @code{s} is specially defined as
1496 equivalent to @code{step} even though there are other commands whose
1497 names start with @code{s}. You can test abbreviations by using them as
1498 arguments to the @code{help} command.
1500 @cindex repeating commands
1501 @kindex RET @r{(repeat last command)}
1502 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1503 repeat the previous command. Certain commands (for example, @code{run})
1504 will not repeat this way; these are commands whose unintentional
1505 repetition might cause trouble and which you are unlikely to want to
1506 repeat. User-defined commands can disable this feature; see
1507 @ref{Define, dont-repeat}.
1509 The @code{list} and @code{x} commands, when you repeat them with
1510 @key{RET}, construct new arguments rather than repeating
1511 exactly as typed. This permits easy scanning of source or memory.
1513 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1514 output, in a way similar to the common utility @code{more}
1515 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1516 @key{RET} too many in this situation, @value{GDBN} disables command
1517 repetition after any command that generates this sort of display.
1519 @kindex # @r{(a comment)}
1521 Any text from a @kbd{#} to the end of the line is a comment; it does
1522 nothing. This is useful mainly in command files (@pxref{Command
1523 Files,,Command Files}).
1525 @cindex repeating command sequences
1526 @kindex Ctrl-o @r{(operate-and-get-next)}
1527 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1528 commands. This command accepts the current line, like @key{RET}, and
1529 then fetches the next line relative to the current line from the history
1533 @section Command Completion
1536 @cindex word completion
1537 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1538 only one possibility; it can also show you what the valid possibilities
1539 are for the next word in a command, at any time. This works for @value{GDBN}
1540 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1542 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1543 of a word. If there is only one possibility, @value{GDBN} fills in the
1544 word, and waits for you to finish the command (or press @key{RET} to
1545 enter it). For example, if you type
1547 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1548 @c complete accuracy in these examples; space introduced for clarity.
1549 @c If texinfo enhancements make it unnecessary, it would be nice to
1550 @c replace " @key" by "@key" in the following...
1552 (@value{GDBP}) info bre @key{TAB}
1556 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1557 the only @code{info} subcommand beginning with @samp{bre}:
1560 (@value{GDBP}) info breakpoints
1564 You can either press @key{RET} at this point, to run the @code{info
1565 breakpoints} command, or backspace and enter something else, if
1566 @samp{breakpoints} does not look like the command you expected. (If you
1567 were sure you wanted @code{info breakpoints} in the first place, you
1568 might as well just type @key{RET} immediately after @samp{info bre},
1569 to exploit command abbreviations rather than command completion).
1571 If there is more than one possibility for the next word when you press
1572 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1573 characters and try again, or just press @key{TAB} a second time;
1574 @value{GDBN} displays all the possible completions for that word. For
1575 example, you might want to set a breakpoint on a subroutine whose name
1576 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1577 just sounds the bell. Typing @key{TAB} again displays all the
1578 function names in your program that begin with those characters, for
1582 (@value{GDBP}) b make_ @key{TAB}
1583 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1584 make_a_section_from_file make_environ
1585 make_abs_section make_function_type
1586 make_blockvector make_pointer_type
1587 make_cleanup make_reference_type
1588 make_command make_symbol_completion_list
1589 (@value{GDBP}) b make_
1593 After displaying the available possibilities, @value{GDBN} copies your
1594 partial input (@samp{b make_} in the example) so you can finish the
1597 If you just want to see the list of alternatives in the first place, you
1598 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1599 means @kbd{@key{META} ?}. You can type this either by holding down a
1600 key designated as the @key{META} shift on your keyboard (if there is
1601 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1603 @cindex quotes in commands
1604 @cindex completion of quoted strings
1605 Sometimes the string you need, while logically a ``word'', may contain
1606 parentheses or other characters that @value{GDBN} normally excludes from
1607 its notion of a word. To permit word completion to work in this
1608 situation, you may enclose words in @code{'} (single quote marks) in
1609 @value{GDBN} commands.
1611 The most likely situation where you might need this is in typing the
1612 name of a C@t{++} function. This is because C@t{++} allows function
1613 overloading (multiple definitions of the same function, distinguished
1614 by argument type). For example, when you want to set a breakpoint you
1615 may need to distinguish whether you mean the version of @code{name}
1616 that takes an @code{int} parameter, @code{name(int)}, or the version
1617 that takes a @code{float} parameter, @code{name(float)}. To use the
1618 word-completion facilities in this situation, type a single quote
1619 @code{'} at the beginning of the function name. This alerts
1620 @value{GDBN} that it may need to consider more information than usual
1621 when you press @key{TAB} or @kbd{M-?} to request word completion:
1624 (@value{GDBP}) b 'bubble( @kbd{M-?}
1625 bubble(double,double) bubble(int,int)
1626 (@value{GDBP}) b 'bubble(
1629 In some cases, @value{GDBN} can tell that completing a name requires using
1630 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1631 completing as much as it can) if you do not type the quote in the first
1635 (@value{GDBP}) b bub @key{TAB}
1636 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1637 (@value{GDBP}) b 'bubble(
1641 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1642 you have not yet started typing the argument list when you ask for
1643 completion on an overloaded symbol.
1645 For more information about overloaded functions, see @ref{C Plus Plus
1646 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1647 overload-resolution off} to disable overload resolution;
1648 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1650 @cindex completion of structure field names
1651 @cindex structure field name completion
1652 @cindex completion of union field names
1653 @cindex union field name completion
1654 When completing in an expression which looks up a field in a
1655 structure, @value{GDBN} also tries@footnote{The completer can be
1656 confused by certain kinds of invalid expressions. Also, it only
1657 examines the static type of the expression, not the dynamic type.} to
1658 limit completions to the field names available in the type of the
1662 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1663 magic to_fputs to_rewind
1664 to_data to_isatty to_write
1665 to_delete to_put to_write_async_safe
1670 This is because the @code{gdb_stdout} is a variable of the type
1671 @code{struct ui_file} that is defined in @value{GDBN} sources as
1678 ui_file_flush_ftype *to_flush;
1679 ui_file_write_ftype *to_write;
1680 ui_file_write_async_safe_ftype *to_write_async_safe;
1681 ui_file_fputs_ftype *to_fputs;
1682 ui_file_read_ftype *to_read;
1683 ui_file_delete_ftype *to_delete;
1684 ui_file_isatty_ftype *to_isatty;
1685 ui_file_rewind_ftype *to_rewind;
1686 ui_file_put_ftype *to_put;
1693 @section Getting Help
1694 @cindex online documentation
1697 You can always ask @value{GDBN} itself for information on its commands,
1698 using the command @code{help}.
1701 @kindex h @r{(@code{help})}
1704 You can use @code{help} (abbreviated @code{h}) with no arguments to
1705 display a short list of named classes of commands:
1709 List of classes of commands:
1711 aliases -- Aliases of other commands
1712 breakpoints -- Making program stop at certain points
1713 data -- Examining data
1714 files -- Specifying and examining files
1715 internals -- Maintenance commands
1716 obscure -- Obscure features
1717 running -- Running the program
1718 stack -- Examining the stack
1719 status -- Status inquiries
1720 support -- Support facilities
1721 tracepoints -- Tracing of program execution without
1722 stopping the program
1723 user-defined -- User-defined commands
1725 Type "help" followed by a class name for a list of
1726 commands in that class.
1727 Type "help" followed by command name for full
1729 Command name abbreviations are allowed if unambiguous.
1732 @c the above line break eliminates huge line overfull...
1734 @item help @var{class}
1735 Using one of the general help classes as an argument, you can get a
1736 list of the individual commands in that class. For example, here is the
1737 help display for the class @code{status}:
1740 (@value{GDBP}) help status
1745 @c Line break in "show" line falsifies real output, but needed
1746 @c to fit in smallbook page size.
1747 info -- Generic command for showing things
1748 about the program being debugged
1749 show -- Generic command for showing things
1752 Type "help" followed by command name for full
1754 Command name abbreviations are allowed if unambiguous.
1758 @item help @var{command}
1759 With a command name as @code{help} argument, @value{GDBN} displays a
1760 short paragraph on how to use that command.
1763 @item apropos @var{args}
1764 The @code{apropos} command searches through all of the @value{GDBN}
1765 commands, and their documentation, for the regular expression specified in
1766 @var{args}. It prints out all matches found. For example:
1777 alias -- Define a new command that is an alias of an existing command
1778 aliases -- Aliases of other commands
1779 d -- Delete some breakpoints or auto-display expressions
1780 del -- Delete some breakpoints or auto-display expressions
1781 delete -- Delete some breakpoints or auto-display expressions
1786 @item complete @var{args}
1787 The @code{complete @var{args}} command lists all the possible completions
1788 for the beginning of a command. Use @var{args} to specify the beginning of the
1789 command you want completed. For example:
1795 @noindent results in:
1806 @noindent This is intended for use by @sc{gnu} Emacs.
1809 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1810 and @code{show} to inquire about the state of your program, or the state
1811 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1812 manual introduces each of them in the appropriate context. The listings
1813 under @code{info} and under @code{show} in the Command, Variable, and
1814 Function Index point to all the sub-commands. @xref{Command and Variable
1820 @kindex i @r{(@code{info})}
1822 This command (abbreviated @code{i}) is for describing the state of your
1823 program. For example, you can show the arguments passed to a function
1824 with @code{info args}, list the registers currently in use with @code{info
1825 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1826 You can get a complete list of the @code{info} sub-commands with
1827 @w{@code{help info}}.
1831 You can assign the result of an expression to an environment variable with
1832 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1833 @code{set prompt $}.
1837 In contrast to @code{info}, @code{show} is for describing the state of
1838 @value{GDBN} itself.
1839 You can change most of the things you can @code{show}, by using the
1840 related command @code{set}; for example, you can control what number
1841 system is used for displays with @code{set radix}, or simply inquire
1842 which is currently in use with @code{show radix}.
1845 To display all the settable parameters and their current
1846 values, you can use @code{show} with no arguments; you may also use
1847 @code{info set}. Both commands produce the same display.
1848 @c FIXME: "info set" violates the rule that "info" is for state of
1849 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1850 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1854 Here are several miscellaneous @code{show} subcommands, all of which are
1855 exceptional in lacking corresponding @code{set} commands:
1858 @kindex show version
1859 @cindex @value{GDBN} version number
1861 Show what version of @value{GDBN} is running. You should include this
1862 information in @value{GDBN} bug-reports. If multiple versions of
1863 @value{GDBN} are in use at your site, you may need to determine which
1864 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1865 commands are introduced, and old ones may wither away. Also, many
1866 system vendors ship variant versions of @value{GDBN}, and there are
1867 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1868 The version number is the same as the one announced when you start
1871 @kindex show copying
1872 @kindex info copying
1873 @cindex display @value{GDBN} copyright
1876 Display information about permission for copying @value{GDBN}.
1878 @kindex show warranty
1879 @kindex info warranty
1881 @itemx info warranty
1882 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1883 if your version of @value{GDBN} comes with one.
1885 @kindex show configuration
1886 @item show configuration
1887 Display detailed information about the way @value{GDBN} was configured
1888 when it was built. This displays the optional arguments passed to the
1889 @file{configure} script and also configuration parameters detected
1890 automatically by @command{configure}. When reporting a @value{GDBN}
1891 bug (@pxref{GDB Bugs}), it is important to include this information in
1897 @chapter Running Programs Under @value{GDBN}
1899 When you run a program under @value{GDBN}, you must first generate
1900 debugging information when you compile it.
1902 You may start @value{GDBN} with its arguments, if any, in an environment
1903 of your choice. If you are doing native debugging, you may redirect
1904 your program's input and output, debug an already running process, or
1905 kill a child process.
1908 * Compilation:: Compiling for debugging
1909 * Starting:: Starting your program
1910 * Arguments:: Your program's arguments
1911 * Environment:: Your program's environment
1913 * Working Directory:: Your program's working directory
1914 * Input/Output:: Your program's input and output
1915 * Attach:: Debugging an already-running process
1916 * Kill Process:: Killing the child process
1918 * Inferiors and Programs:: Debugging multiple inferiors and programs
1919 * Threads:: Debugging programs with multiple threads
1920 * Forks:: Debugging forks
1921 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1925 @section Compiling for Debugging
1927 In order to debug a program effectively, you need to generate
1928 debugging information when you compile it. This debugging information
1929 is stored in the object file; it describes the data type of each
1930 variable or function and the correspondence between source line numbers
1931 and addresses in the executable code.
1933 To request debugging information, specify the @samp{-g} option when you run
1936 Programs that are to be shipped to your customers are compiled with
1937 optimizations, using the @samp{-O} compiler option. However, some
1938 compilers are unable to handle the @samp{-g} and @samp{-O} options
1939 together. Using those compilers, you cannot generate optimized
1940 executables containing debugging information.
1942 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1943 without @samp{-O}, making it possible to debug optimized code. We
1944 recommend that you @emph{always} use @samp{-g} whenever you compile a
1945 program. You may think your program is correct, but there is no sense
1946 in pushing your luck. For more information, see @ref{Optimized Code}.
1948 Older versions of the @sc{gnu} C compiler permitted a variant option
1949 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1950 format; if your @sc{gnu} C compiler has this option, do not use it.
1952 @value{GDBN} knows about preprocessor macros and can show you their
1953 expansion (@pxref{Macros}). Most compilers do not include information
1954 about preprocessor macros in the debugging information if you specify
1955 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
1956 the @sc{gnu} C compiler, provides macro information if you are using
1957 the DWARF debugging format, and specify the option @option{-g3}.
1959 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
1960 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}, for more
1961 information on @value{NGCC} options affecting debug information.
1963 You will have the best debugging experience if you use the latest
1964 version of the DWARF debugging format that your compiler supports.
1965 DWARF is currently the most expressive and best supported debugging
1966 format in @value{GDBN}.
1970 @section Starting your Program
1976 @kindex r @r{(@code{run})}
1979 Use the @code{run} command to start your program under @value{GDBN}.
1980 You must first specify the program name with an argument to
1981 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1982 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
1983 command (@pxref{Files, ,Commands to Specify Files}).
1987 If you are running your program in an execution environment that
1988 supports processes, @code{run} creates an inferior process and makes
1989 that process run your program. In some environments without processes,
1990 @code{run} jumps to the start of your program. Other targets,
1991 like @samp{remote}, are always running. If you get an error
1992 message like this one:
1995 The "remote" target does not support "run".
1996 Try "help target" or "continue".
2000 then use @code{continue} to run your program. You may need @code{load}
2001 first (@pxref{load}).
2003 The execution of a program is affected by certain information it
2004 receives from its superior. @value{GDBN} provides ways to specify this
2005 information, which you must do @emph{before} starting your program. (You
2006 can change it after starting your program, but such changes only affect
2007 your program the next time you start it.) This information may be
2008 divided into four categories:
2011 @item The @emph{arguments.}
2012 Specify the arguments to give your program as the arguments of the
2013 @code{run} command. If a shell is available on your target, the shell
2014 is used to pass the arguments, so that you may use normal conventions
2015 (such as wildcard expansion or variable substitution) in describing
2017 In Unix systems, you can control which shell is used with the
2018 @code{SHELL} environment variable. If you do not define @code{SHELL},
2019 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2020 use of any shell with the @code{set startup-with-shell} command (see
2023 @item The @emph{environment.}
2024 Your program normally inherits its environment from @value{GDBN}, but you can
2025 use the @value{GDBN} commands @code{set environment} and @code{unset
2026 environment} to change parts of the environment that affect
2027 your program. @xref{Environment, ,Your Program's Environment}.
2029 @item The @emph{working directory.}
2030 Your program inherits its working directory from @value{GDBN}. You can set
2031 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
2032 @xref{Working Directory, ,Your Program's Working Directory}.
2034 @item The @emph{standard input and output.}
2035 Your program normally uses the same device for standard input and
2036 standard output as @value{GDBN} is using. You can redirect input and output
2037 in the @code{run} command line, or you can use the @code{tty} command to
2038 set a different device for your program.
2039 @xref{Input/Output, ,Your Program's Input and Output}.
2042 @emph{Warning:} While input and output redirection work, you cannot use
2043 pipes to pass the output of the program you are debugging to another
2044 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2048 When you issue the @code{run} command, your program begins to execute
2049 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2050 of how to arrange for your program to stop. Once your program has
2051 stopped, you may call functions in your program, using the @code{print}
2052 or @code{call} commands. @xref{Data, ,Examining Data}.
2054 If the modification time of your symbol file has changed since the last
2055 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2056 table, and reads it again. When it does this, @value{GDBN} tries to retain
2057 your current breakpoints.
2062 @cindex run to main procedure
2063 The name of the main procedure can vary from language to language.
2064 With C or C@t{++}, the main procedure name is always @code{main}, but
2065 other languages such as Ada do not require a specific name for their
2066 main procedure. The debugger provides a convenient way to start the
2067 execution of the program and to stop at the beginning of the main
2068 procedure, depending on the language used.
2070 The @samp{start} command does the equivalent of setting a temporary
2071 breakpoint at the beginning of the main procedure and then invoking
2072 the @samp{run} command.
2074 @cindex elaboration phase
2075 Some programs contain an @dfn{elaboration} phase where some startup code is
2076 executed before the main procedure is called. This depends on the
2077 languages used to write your program. In C@t{++}, for instance,
2078 constructors for static and global objects are executed before
2079 @code{main} is called. It is therefore possible that the debugger stops
2080 before reaching the main procedure. However, the temporary breakpoint
2081 will remain to halt execution.
2083 Specify the arguments to give to your program as arguments to the
2084 @samp{start} command. These arguments will be given verbatim to the
2085 underlying @samp{run} command. Note that the same arguments will be
2086 reused if no argument is provided during subsequent calls to
2087 @samp{start} or @samp{run}.
2089 It is sometimes necessary to debug the program during elaboration. In
2090 these cases, using the @code{start} command would stop the execution of
2091 your program too late, as the program would have already completed the
2092 elaboration phase. Under these circumstances, insert breakpoints in your
2093 elaboration code before running your program.
2095 @anchor{set exec-wrapper}
2096 @kindex set exec-wrapper
2097 @item set exec-wrapper @var{wrapper}
2098 @itemx show exec-wrapper
2099 @itemx unset exec-wrapper
2100 When @samp{exec-wrapper} is set, the specified wrapper is used to
2101 launch programs for debugging. @value{GDBN} starts your program
2102 with a shell command of the form @kbd{exec @var{wrapper}
2103 @var{program}}. Quoting is added to @var{program} and its
2104 arguments, but not to @var{wrapper}, so you should add quotes if
2105 appropriate for your shell. The wrapper runs until it executes
2106 your program, and then @value{GDBN} takes control.
2108 You can use any program that eventually calls @code{execve} with
2109 its arguments as a wrapper. Several standard Unix utilities do
2110 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2111 with @code{exec "$@@"} will also work.
2113 For example, you can use @code{env} to pass an environment variable to
2114 the debugged program, without setting the variable in your shell's
2118 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2122 This command is available when debugging locally on most targets, excluding
2123 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2125 @kindex set startup-with-shell
2126 @item set startup-with-shell
2127 @itemx set startup-with-shell on
2128 @itemx set startup-with-shell off
2129 @itemx show set startup-with-shell
2130 On Unix systems, by default, if a shell is available on your target,
2131 @value{GDBN}) uses it to start your program. Arguments of the
2132 @code{run} command are passed to the shell, which does variable
2133 substitution, expands wildcard characters and performs redirection of
2134 I/O. In some circumstances, it may be useful to disable such use of a
2135 shell, for example, when debugging the shell itself or diagnosing
2136 startup failures such as:
2140 Starting program: ./a.out
2141 During startup program terminated with signal SIGSEGV, Segmentation fault.
2145 which indicates the shell or the wrapper specified with
2146 @samp{exec-wrapper} crashed, not your program. Most often, this is
2147 caused by something odd in your shell's non-interactive mode
2148 initialization file---such as @file{.cshrc} for C-shell,
2149 $@file{.zshenv} for the Z shell, or the file specified in the
2150 @samp{BASH_ENV} environment variable for BASH.
2152 @anchor{set auto-connect-native-target}
2153 @kindex set auto-connect-native-target
2154 @item set auto-connect-native-target
2155 @itemx set auto-connect-native-target on
2156 @itemx set auto-connect-native-target off
2157 @itemx show auto-connect-native-target
2159 By default, if not connected to any target yet (e.g., with
2160 @code{target remote}), the @code{run} command starts your program as a
2161 native process under @value{GDBN}, on your local machine. If you're
2162 sure you don't want to debug programs on your local machine, you can
2163 tell @value{GDBN} to not connect to the native target automatically
2164 with the @code{set auto-connect-native-target off} command.
2166 If @code{on}, which is the default, and if @value{GDBN} is not
2167 connected to a target already, the @code{run} command automaticaly
2168 connects to the native target, if one is available.
2170 If @code{off}, and if @value{GDBN} is not connected to a target
2171 already, the @code{run} command fails with an error:
2175 Don't know how to run. Try "help target".
2178 If @value{GDBN} is already connected to a target, @value{GDBN} always
2179 uses it with the @code{run} command.
2181 In any case, you can explicitly connect to the native target with the
2182 @code{target native} command. For example,
2185 (@value{GDBP}) set auto-connect-native-target off
2187 Don't know how to run. Try "help target".
2188 (@value{GDBP}) target native
2190 Starting program: ./a.out
2191 [Inferior 1 (process 10421) exited normally]
2194 In case you connected explicitly to the @code{native} target,
2195 @value{GDBN} remains connected even if all inferiors exit, ready for
2196 the next @code{run} command. Use the @code{disconnect} command to
2199 Examples of other commands that likewise respect the
2200 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2201 proc}, @code{info os}.
2203 @kindex set disable-randomization
2204 @item set disable-randomization
2205 @itemx set disable-randomization on
2206 This option (enabled by default in @value{GDBN}) will turn off the native
2207 randomization of the virtual address space of the started program. This option
2208 is useful for multiple debugging sessions to make the execution better
2209 reproducible and memory addresses reusable across debugging sessions.
2211 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2212 On @sc{gnu}/Linux you can get the same behavior using
2215 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2218 @item set disable-randomization off
2219 Leave the behavior of the started executable unchanged. Some bugs rear their
2220 ugly heads only when the program is loaded at certain addresses. If your bug
2221 disappears when you run the program under @value{GDBN}, that might be because
2222 @value{GDBN} by default disables the address randomization on platforms, such
2223 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2224 disable-randomization off} to try to reproduce such elusive bugs.
2226 On targets where it is available, virtual address space randomization
2227 protects the programs against certain kinds of security attacks. In these
2228 cases the attacker needs to know the exact location of a concrete executable
2229 code. Randomizing its location makes it impossible to inject jumps misusing
2230 a code at its expected addresses.
2232 Prelinking shared libraries provides a startup performance advantage but it
2233 makes addresses in these libraries predictable for privileged processes by
2234 having just unprivileged access at the target system. Reading the shared
2235 library binary gives enough information for assembling the malicious code
2236 misusing it. Still even a prelinked shared library can get loaded at a new
2237 random address just requiring the regular relocation process during the
2238 startup. Shared libraries not already prelinked are always loaded at
2239 a randomly chosen address.
2241 Position independent executables (PIE) contain position independent code
2242 similar to the shared libraries and therefore such executables get loaded at
2243 a randomly chosen address upon startup. PIE executables always load even
2244 already prelinked shared libraries at a random address. You can build such
2245 executable using @command{gcc -fPIE -pie}.
2247 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2248 (as long as the randomization is enabled).
2250 @item show disable-randomization
2251 Show the current setting of the explicit disable of the native randomization of
2252 the virtual address space of the started program.
2257 @section Your Program's Arguments
2259 @cindex arguments (to your program)
2260 The arguments to your program can be specified by the arguments of the
2262 They are passed to a shell, which expands wildcard characters and
2263 performs redirection of I/O, and thence to your program. Your
2264 @code{SHELL} environment variable (if it exists) specifies what shell
2265 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2266 the default shell (@file{/bin/sh} on Unix).
2268 On non-Unix systems, the program is usually invoked directly by
2269 @value{GDBN}, which emulates I/O redirection via the appropriate system
2270 calls, and the wildcard characters are expanded by the startup code of
2271 the program, not by the shell.
2273 @code{run} with no arguments uses the same arguments used by the previous
2274 @code{run}, or those set by the @code{set args} command.
2279 Specify the arguments to be used the next time your program is run. If
2280 @code{set args} has no arguments, @code{run} executes your program
2281 with no arguments. Once you have run your program with arguments,
2282 using @code{set args} before the next @code{run} is the only way to run
2283 it again without arguments.
2287 Show the arguments to give your program when it is started.
2291 @section Your Program's Environment
2293 @cindex environment (of your program)
2294 The @dfn{environment} consists of a set of environment variables and
2295 their values. Environment variables conventionally record such things as
2296 your user name, your home directory, your terminal type, and your search
2297 path for programs to run. Usually you set up environment variables with
2298 the shell and they are inherited by all the other programs you run. When
2299 debugging, it can be useful to try running your program with a modified
2300 environment without having to start @value{GDBN} over again.
2304 @item path @var{directory}
2305 Add @var{directory} to the front of the @code{PATH} environment variable
2306 (the search path for executables) that will be passed to your program.
2307 The value of @code{PATH} used by @value{GDBN} does not change.
2308 You may specify several directory names, separated by whitespace or by a
2309 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2310 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2311 is moved to the front, so it is searched sooner.
2313 You can use the string @samp{$cwd} to refer to whatever is the current
2314 working directory at the time @value{GDBN} searches the path. If you
2315 use @samp{.} instead, it refers to the directory where you executed the
2316 @code{path} command. @value{GDBN} replaces @samp{.} in the
2317 @var{directory} argument (with the current path) before adding
2318 @var{directory} to the search path.
2319 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2320 @c document that, since repeating it would be a no-op.
2324 Display the list of search paths for executables (the @code{PATH}
2325 environment variable).
2327 @kindex show environment
2328 @item show environment @r{[}@var{varname}@r{]}
2329 Print the value of environment variable @var{varname} to be given to
2330 your program when it starts. If you do not supply @var{varname},
2331 print the names and values of all environment variables to be given to
2332 your program. You can abbreviate @code{environment} as @code{env}.
2334 @kindex set environment
2335 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2336 Set environment variable @var{varname} to @var{value}. The value
2337 changes for your program (and the shell @value{GDBN} uses to launch
2338 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2339 values of environment variables are just strings, and any
2340 interpretation is supplied by your program itself. The @var{value}
2341 parameter is optional; if it is eliminated, the variable is set to a
2343 @c "any string" here does not include leading, trailing
2344 @c blanks. Gnu asks: does anyone care?
2346 For example, this command:
2353 tells the debugged program, when subsequently run, that its user is named
2354 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2355 are not actually required.)
2357 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2358 which also inherits the environment set with @code{set environment}.
2359 If necessary, you can avoid that by using the @samp{env} program as a
2360 wrapper instead of using @code{set environment}. @xref{set
2361 exec-wrapper}, for an example doing just that.
2363 @kindex unset environment
2364 @item unset environment @var{varname}
2365 Remove variable @var{varname} from the environment to be passed to your
2366 program. This is different from @samp{set env @var{varname} =};
2367 @code{unset environment} removes the variable from the environment,
2368 rather than assigning it an empty value.
2371 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2372 the shell indicated by your @code{SHELL} environment variable if it
2373 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2374 names a shell that runs an initialization file when started
2375 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2376 for the Z shell, or the file specified in the @samp{BASH_ENV}
2377 environment variable for BASH---any variables you set in that file
2378 affect your program. You may wish to move setting of environment
2379 variables to files that are only run when you sign on, such as
2380 @file{.login} or @file{.profile}.
2382 @node Working Directory
2383 @section Your Program's Working Directory
2385 @cindex working directory (of your program)
2386 Each time you start your program with @code{run}, it inherits its
2387 working directory from the current working directory of @value{GDBN}.
2388 The @value{GDBN} working directory is initially whatever it inherited
2389 from its parent process (typically the shell), but you can specify a new
2390 working directory in @value{GDBN} with the @code{cd} command.
2392 The @value{GDBN} working directory also serves as a default for the commands
2393 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2398 @cindex change working directory
2399 @item cd @r{[}@var{directory}@r{]}
2400 Set the @value{GDBN} working directory to @var{directory}. If not
2401 given, @var{directory} uses @file{'~'}.
2405 Print the @value{GDBN} working directory.
2408 It is generally impossible to find the current working directory of
2409 the process being debugged (since a program can change its directory
2410 during its run). If you work on a system where @value{GDBN} is
2411 configured with the @file{/proc} support, you can use the @code{info
2412 proc} command (@pxref{SVR4 Process Information}) to find out the
2413 current working directory of the debuggee.
2416 @section Your Program's Input and Output
2421 By default, the program you run under @value{GDBN} does input and output to
2422 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2423 to its own terminal modes to interact with you, but it records the terminal
2424 modes your program was using and switches back to them when you continue
2425 running your program.
2428 @kindex info terminal
2430 Displays information recorded by @value{GDBN} about the terminal modes your
2434 You can redirect your program's input and/or output using shell
2435 redirection with the @code{run} command. For example,
2442 starts your program, diverting its output to the file @file{outfile}.
2445 @cindex controlling terminal
2446 Another way to specify where your program should do input and output is
2447 with the @code{tty} command. This command accepts a file name as
2448 argument, and causes this file to be the default for future @code{run}
2449 commands. It also resets the controlling terminal for the child
2450 process, for future @code{run} commands. For example,
2457 directs that processes started with subsequent @code{run} commands
2458 default to do input and output on the terminal @file{/dev/ttyb} and have
2459 that as their controlling terminal.
2461 An explicit redirection in @code{run} overrides the @code{tty} command's
2462 effect on the input/output device, but not its effect on the controlling
2465 When you use the @code{tty} command or redirect input in the @code{run}
2466 command, only the input @emph{for your program} is affected. The input
2467 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2468 for @code{set inferior-tty}.
2470 @cindex inferior tty
2471 @cindex set inferior controlling terminal
2472 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2473 display the name of the terminal that will be used for future runs of your
2477 @item set inferior-tty /dev/ttyb
2478 @kindex set inferior-tty
2479 Set the tty for the program being debugged to /dev/ttyb.
2481 @item show inferior-tty
2482 @kindex show inferior-tty
2483 Show the current tty for the program being debugged.
2487 @section Debugging an Already-running Process
2492 @item attach @var{process-id}
2493 This command attaches to a running process---one that was started
2494 outside @value{GDBN}. (@code{info files} shows your active
2495 targets.) The command takes as argument a process ID. The usual way to
2496 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2497 or with the @samp{jobs -l} shell command.
2499 @code{attach} does not repeat if you press @key{RET} a second time after
2500 executing the command.
2503 To use @code{attach}, your program must be running in an environment
2504 which supports processes; for example, @code{attach} does not work for
2505 programs on bare-board targets that lack an operating system. You must
2506 also have permission to send the process a signal.
2508 When you use @code{attach}, the debugger finds the program running in
2509 the process first by looking in the current working directory, then (if
2510 the program is not found) by using the source file search path
2511 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2512 the @code{file} command to load the program. @xref{Files, ,Commands to
2515 The first thing @value{GDBN} does after arranging to debug the specified
2516 process is to stop it. You can examine and modify an attached process
2517 with all the @value{GDBN} commands that are ordinarily available when
2518 you start processes with @code{run}. You can insert breakpoints; you
2519 can step and continue; you can modify storage. If you would rather the
2520 process continue running, you may use the @code{continue} command after
2521 attaching @value{GDBN} to the process.
2526 When you have finished debugging the attached process, you can use the
2527 @code{detach} command to release it from @value{GDBN} control. Detaching
2528 the process continues its execution. After the @code{detach} command,
2529 that process and @value{GDBN} become completely independent once more, and you
2530 are ready to @code{attach} another process or start one with @code{run}.
2531 @code{detach} does not repeat if you press @key{RET} again after
2532 executing the command.
2535 If you exit @value{GDBN} while you have an attached process, you detach
2536 that process. If you use the @code{run} command, you kill that process.
2537 By default, @value{GDBN} asks for confirmation if you try to do either of these
2538 things; you can control whether or not you need to confirm by using the
2539 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2543 @section Killing the Child Process
2548 Kill the child process in which your program is running under @value{GDBN}.
2551 This command is useful if you wish to debug a core dump instead of a
2552 running process. @value{GDBN} ignores any core dump file while your program
2555 On some operating systems, a program cannot be executed outside @value{GDBN}
2556 while you have breakpoints set on it inside @value{GDBN}. You can use the
2557 @code{kill} command in this situation to permit running your program
2558 outside the debugger.
2560 The @code{kill} command is also useful if you wish to recompile and
2561 relink your program, since on many systems it is impossible to modify an
2562 executable file while it is running in a process. In this case, when you
2563 next type @code{run}, @value{GDBN} notices that the file has changed, and
2564 reads the symbol table again (while trying to preserve your current
2565 breakpoint settings).
2567 @node Inferiors and Programs
2568 @section Debugging Multiple Inferiors and Programs
2570 @value{GDBN} lets you run and debug multiple programs in a single
2571 session. In addition, @value{GDBN} on some systems may let you run
2572 several programs simultaneously (otherwise you have to exit from one
2573 before starting another). In the most general case, you can have
2574 multiple threads of execution in each of multiple processes, launched
2575 from multiple executables.
2578 @value{GDBN} represents the state of each program execution with an
2579 object called an @dfn{inferior}. An inferior typically corresponds to
2580 a process, but is more general and applies also to targets that do not
2581 have processes. Inferiors may be created before a process runs, and
2582 may be retained after a process exits. Inferiors have unique
2583 identifiers that are different from process ids. Usually each
2584 inferior will also have its own distinct address space, although some
2585 embedded targets may have several inferiors running in different parts
2586 of a single address space. Each inferior may in turn have multiple
2587 threads running in it.
2589 To find out what inferiors exist at any moment, use @w{@code{info
2593 @kindex info inferiors
2594 @item info inferiors
2595 Print a list of all inferiors currently being managed by @value{GDBN}.
2597 @value{GDBN} displays for each inferior (in this order):
2601 the inferior number assigned by @value{GDBN}
2604 the target system's inferior identifier
2607 the name of the executable the inferior is running.
2612 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2613 indicates the current inferior.
2617 @c end table here to get a little more width for example
2620 (@value{GDBP}) info inferiors
2621 Num Description Executable
2622 2 process 2307 hello
2623 * 1 process 3401 goodbye
2626 To switch focus between inferiors, use the @code{inferior} command:
2629 @kindex inferior @var{infno}
2630 @item inferior @var{infno}
2631 Make inferior number @var{infno} the current inferior. The argument
2632 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2633 in the first field of the @samp{info inferiors} display.
2637 You can get multiple executables into a debugging session via the
2638 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2639 systems @value{GDBN} can add inferiors to the debug session
2640 automatically by following calls to @code{fork} and @code{exec}. To
2641 remove inferiors from the debugging session use the
2642 @w{@code{remove-inferiors}} command.
2645 @kindex add-inferior
2646 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2647 Adds @var{n} inferiors to be run using @var{executable} as the
2648 executable; @var{n} defaults to 1. If no executable is specified,
2649 the inferiors begins empty, with no program. You can still assign or
2650 change the program assigned to the inferior at any time by using the
2651 @code{file} command with the executable name as its argument.
2653 @kindex clone-inferior
2654 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2655 Adds @var{n} inferiors ready to execute the same program as inferior
2656 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
2657 number of the current inferior. This is a convenient command when you
2658 want to run another instance of the inferior you are debugging.
2661 (@value{GDBP}) info inferiors
2662 Num Description Executable
2663 * 1 process 29964 helloworld
2664 (@value{GDBP}) clone-inferior
2667 (@value{GDBP}) info inferiors
2668 Num Description Executable
2670 * 1 process 29964 helloworld
2673 You can now simply switch focus to inferior 2 and run it.
2675 @kindex remove-inferiors
2676 @item remove-inferiors @var{infno}@dots{}
2677 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2678 possible to remove an inferior that is running with this command. For
2679 those, use the @code{kill} or @code{detach} command first.
2683 To quit debugging one of the running inferiors that is not the current
2684 inferior, you can either detach from it by using the @w{@code{detach
2685 inferior}} command (allowing it to run independently), or kill it
2686 using the @w{@code{kill inferiors}} command:
2689 @kindex detach inferiors @var{infno}@dots{}
2690 @item detach inferior @var{infno}@dots{}
2691 Detach from the inferior or inferiors identified by @value{GDBN}
2692 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2693 still stays on the list of inferiors shown by @code{info inferiors},
2694 but its Description will show @samp{<null>}.
2696 @kindex kill inferiors @var{infno}@dots{}
2697 @item kill inferiors @var{infno}@dots{}
2698 Kill the inferior or inferiors identified by @value{GDBN} inferior
2699 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2700 stays on the list of inferiors shown by @code{info inferiors}, but its
2701 Description will show @samp{<null>}.
2704 After the successful completion of a command such as @code{detach},
2705 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2706 a normal process exit, the inferior is still valid and listed with
2707 @code{info inferiors}, ready to be restarted.
2710 To be notified when inferiors are started or exit under @value{GDBN}'s
2711 control use @w{@code{set print inferior-events}}:
2714 @kindex set print inferior-events
2715 @cindex print messages on inferior start and exit
2716 @item set print inferior-events
2717 @itemx set print inferior-events on
2718 @itemx set print inferior-events off
2719 The @code{set print inferior-events} command allows you to enable or
2720 disable printing of messages when @value{GDBN} notices that new
2721 inferiors have started or that inferiors have exited or have been
2722 detached. By default, these messages will not be printed.
2724 @kindex show print inferior-events
2725 @item show print inferior-events
2726 Show whether messages will be printed when @value{GDBN} detects that
2727 inferiors have started, exited or have been detached.
2730 Many commands will work the same with multiple programs as with a
2731 single program: e.g., @code{print myglobal} will simply display the
2732 value of @code{myglobal} in the current inferior.
2735 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2736 get more info about the relationship of inferiors, programs, address
2737 spaces in a debug session. You can do that with the @w{@code{maint
2738 info program-spaces}} command.
2741 @kindex maint info program-spaces
2742 @item maint info program-spaces
2743 Print a list of all program spaces currently being managed by
2746 @value{GDBN} displays for each program space (in this order):
2750 the program space number assigned by @value{GDBN}
2753 the name of the executable loaded into the program space, with e.g.,
2754 the @code{file} command.
2759 An asterisk @samp{*} preceding the @value{GDBN} program space number
2760 indicates the current program space.
2762 In addition, below each program space line, @value{GDBN} prints extra
2763 information that isn't suitable to display in tabular form. For
2764 example, the list of inferiors bound to the program space.
2767 (@value{GDBP}) maint info program-spaces
2770 Bound inferiors: ID 1 (process 21561)
2774 Here we can see that no inferior is running the program @code{hello},
2775 while @code{process 21561} is running the program @code{goodbye}. On
2776 some targets, it is possible that multiple inferiors are bound to the
2777 same program space. The most common example is that of debugging both
2778 the parent and child processes of a @code{vfork} call. For example,
2781 (@value{GDBP}) maint info program-spaces
2784 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2787 Here, both inferior 2 and inferior 1 are running in the same program
2788 space as a result of inferior 1 having executed a @code{vfork} call.
2792 @section Debugging Programs with Multiple Threads
2794 @cindex threads of execution
2795 @cindex multiple threads
2796 @cindex switching threads
2797 In some operating systems, such as HP-UX and Solaris, a single program
2798 may have more than one @dfn{thread} of execution. The precise semantics
2799 of threads differ from one operating system to another, but in general
2800 the threads of a single program are akin to multiple processes---except
2801 that they share one address space (that is, they can all examine and
2802 modify the same variables). On the other hand, each thread has its own
2803 registers and execution stack, and perhaps private memory.
2805 @value{GDBN} provides these facilities for debugging multi-thread
2809 @item automatic notification of new threads
2810 @item @samp{thread @var{threadno}}, a command to switch among threads
2811 @item @samp{info threads}, a command to inquire about existing threads
2812 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2813 a command to apply a command to a list of threads
2814 @item thread-specific breakpoints
2815 @item @samp{set print thread-events}, which controls printing of
2816 messages on thread start and exit.
2817 @item @samp{set libthread-db-search-path @var{path}}, which lets
2818 the user specify which @code{libthread_db} to use if the default choice
2819 isn't compatible with the program.
2823 @emph{Warning:} These facilities are not yet available on every
2824 @value{GDBN} configuration where the operating system supports threads.
2825 If your @value{GDBN} does not support threads, these commands have no
2826 effect. For example, a system without thread support shows no output
2827 from @samp{info threads}, and always rejects the @code{thread} command,
2831 (@value{GDBP}) info threads
2832 (@value{GDBP}) thread 1
2833 Thread ID 1 not known. Use the "info threads" command to
2834 see the IDs of currently known threads.
2836 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2837 @c doesn't support threads"?
2840 @cindex focus of debugging
2841 @cindex current thread
2842 The @value{GDBN} thread debugging facility allows you to observe all
2843 threads while your program runs---but whenever @value{GDBN} takes
2844 control, one thread in particular is always the focus of debugging.
2845 This thread is called the @dfn{current thread}. Debugging commands show
2846 program information from the perspective of the current thread.
2848 @cindex @code{New} @var{systag} message
2849 @cindex thread identifier (system)
2850 @c FIXME-implementors!! It would be more helpful if the [New...] message
2851 @c included GDB's numeric thread handle, so you could just go to that
2852 @c thread without first checking `info threads'.
2853 Whenever @value{GDBN} detects a new thread in your program, it displays
2854 the target system's identification for the thread with a message in the
2855 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
2856 whose form varies depending on the particular system. For example, on
2857 @sc{gnu}/Linux, you might see
2860 [New Thread 0x41e02940 (LWP 25582)]
2864 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2865 the @var{systag} is simply something like @samp{process 368}, with no
2868 @c FIXME!! (1) Does the [New...] message appear even for the very first
2869 @c thread of a program, or does it only appear for the
2870 @c second---i.e.@: when it becomes obvious we have a multithread
2872 @c (2) *Is* there necessarily a first thread always? Or do some
2873 @c multithread systems permit starting a program with multiple
2874 @c threads ab initio?
2876 @cindex thread number
2877 @cindex thread identifier (GDB)
2878 For debugging purposes, @value{GDBN} associates its own thread
2879 number---always a single integer---with each thread in your program.
2882 @kindex info threads
2883 @item info threads @r{[}@var{id}@dots{}@r{]}
2884 Display a summary of all threads currently in your program. Optional
2885 argument @var{id}@dots{} is one or more thread ids separated by spaces, and
2886 means to print information only about the specified thread or threads.
2887 @value{GDBN} displays for each thread (in this order):
2891 the thread number assigned by @value{GDBN}
2894 the target system's thread identifier (@var{systag})
2897 the thread's name, if one is known. A thread can either be named by
2898 the user (see @code{thread name}, below), or, in some cases, by the
2902 the current stack frame summary for that thread
2906 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2907 indicates the current thread.
2911 @c end table here to get a little more width for example
2914 (@value{GDBP}) info threads
2916 3 process 35 thread 27 0x34e5 in sigpause ()
2917 2 process 35 thread 23 0x34e5 in sigpause ()
2918 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2922 On Solaris, you can display more information about user threads with a
2923 Solaris-specific command:
2926 @item maint info sol-threads
2927 @kindex maint info sol-threads
2928 @cindex thread info (Solaris)
2929 Display info on Solaris user threads.
2933 @kindex thread @var{threadno}
2934 @item thread @var{threadno}
2935 Make thread number @var{threadno} the current thread. The command
2936 argument @var{threadno} is the internal @value{GDBN} thread number, as
2937 shown in the first field of the @samp{info threads} display.
2938 @value{GDBN} responds by displaying the system identifier of the thread
2939 you selected, and its current stack frame summary:
2942 (@value{GDBP}) thread 2
2943 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
2944 #0 some_function (ignore=0x0) at example.c:8
2945 8 printf ("hello\n");
2949 As with the @samp{[New @dots{}]} message, the form of the text after
2950 @samp{Switching to} depends on your system's conventions for identifying
2953 @vindex $_thread@r{, convenience variable}
2954 The debugger convenience variable @samp{$_thread} contains the number
2955 of the current thread. You may find this useful in writing breakpoint
2956 conditional expressions, command scripts, and so forth. See
2957 @xref{Convenience Vars,, Convenience Variables}, for general
2958 information on convenience variables.
2960 @kindex thread apply
2961 @cindex apply command to several threads
2962 @item thread apply [@var{threadno} | all] @var{command}
2963 The @code{thread apply} command allows you to apply the named
2964 @var{command} to one or more threads. Specify the numbers of the
2965 threads that you want affected with the command argument
2966 @var{threadno}. It can be a single thread number, one of the numbers
2967 shown in the first field of the @samp{info threads} display; or it
2968 could be a range of thread numbers, as in @code{2-4}. To apply a
2969 command to all threads, type @kbd{thread apply all @var{command}}.
2972 @cindex name a thread
2973 @item thread name [@var{name}]
2974 This command assigns a name to the current thread. If no argument is
2975 given, any existing user-specified name is removed. The thread name
2976 appears in the @samp{info threads} display.
2978 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
2979 determine the name of the thread as given by the OS. On these
2980 systems, a name specified with @samp{thread name} will override the
2981 system-give name, and removing the user-specified name will cause
2982 @value{GDBN} to once again display the system-specified name.
2985 @cindex search for a thread
2986 @item thread find [@var{regexp}]
2987 Search for and display thread ids whose name or @var{systag}
2988 matches the supplied regular expression.
2990 As well as being the complement to the @samp{thread name} command,
2991 this command also allows you to identify a thread by its target
2992 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
2996 (@value{GDBN}) thread find 26688
2997 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
2998 (@value{GDBN}) info thread 4
3000 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3003 @kindex set print thread-events
3004 @cindex print messages on thread start and exit
3005 @item set print thread-events
3006 @itemx set print thread-events on
3007 @itemx set print thread-events off
3008 The @code{set print thread-events} command allows you to enable or
3009 disable printing of messages when @value{GDBN} notices that new threads have
3010 started or that threads have exited. By default, these messages will
3011 be printed if detection of these events is supported by the target.
3012 Note that these messages cannot be disabled on all targets.
3014 @kindex show print thread-events
3015 @item show print thread-events
3016 Show whether messages will be printed when @value{GDBN} detects that threads
3017 have started and exited.
3020 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3021 more information about how @value{GDBN} behaves when you stop and start
3022 programs with multiple threads.
3024 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3025 watchpoints in programs with multiple threads.
3027 @anchor{set libthread-db-search-path}
3029 @kindex set libthread-db-search-path
3030 @cindex search path for @code{libthread_db}
3031 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3032 If this variable is set, @var{path} is a colon-separated list of
3033 directories @value{GDBN} will use to search for @code{libthread_db}.
3034 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3035 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3036 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3039 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3040 @code{libthread_db} library to obtain information about threads in the
3041 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3042 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3043 specific thread debugging library loading is enabled
3044 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3046 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3047 refers to the default system directories that are
3048 normally searched for loading shared libraries. The @samp{$sdir} entry
3049 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3050 (@pxref{libthread_db.so.1 file}).
3052 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3053 refers to the directory from which @code{libpthread}
3054 was loaded in the inferior process.
3056 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3057 @value{GDBN} attempts to initialize it with the current inferior process.
3058 If this initialization fails (which could happen because of a version
3059 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3060 will unload @code{libthread_db}, and continue with the next directory.
3061 If none of @code{libthread_db} libraries initialize successfully,
3062 @value{GDBN} will issue a warning and thread debugging will be disabled.
3064 Setting @code{libthread-db-search-path} is currently implemented
3065 only on some platforms.
3067 @kindex show libthread-db-search-path
3068 @item show libthread-db-search-path
3069 Display current libthread_db search path.
3071 @kindex set debug libthread-db
3072 @kindex show debug libthread-db
3073 @cindex debugging @code{libthread_db}
3074 @item set debug libthread-db
3075 @itemx show debug libthread-db
3076 Turns on or off display of @code{libthread_db}-related events.
3077 Use @code{1} to enable, @code{0} to disable.
3081 @section Debugging Forks
3083 @cindex fork, debugging programs which call
3084 @cindex multiple processes
3085 @cindex processes, multiple
3086 On most systems, @value{GDBN} has no special support for debugging
3087 programs which create additional processes using the @code{fork}
3088 function. When a program forks, @value{GDBN} will continue to debug the
3089 parent process and the child process will run unimpeded. If you have
3090 set a breakpoint in any code which the child then executes, the child
3091 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3092 will cause it to terminate.
3094 However, if you want to debug the child process there is a workaround
3095 which isn't too painful. Put a call to @code{sleep} in the code which
3096 the child process executes after the fork. It may be useful to sleep
3097 only if a certain environment variable is set, or a certain file exists,
3098 so that the delay need not occur when you don't want to run @value{GDBN}
3099 on the child. While the child is sleeping, use the @code{ps} program to
3100 get its process ID. Then tell @value{GDBN} (a new invocation of
3101 @value{GDBN} if you are also debugging the parent process) to attach to
3102 the child process (@pxref{Attach}). From that point on you can debug
3103 the child process just like any other process which you attached to.
3105 On some systems, @value{GDBN} provides support for debugging programs that
3106 create additional processes using the @code{fork} or @code{vfork} functions.
3107 Currently, the only platforms with this feature are HP-UX (11.x and later
3108 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
3110 By default, when a program forks, @value{GDBN} will continue to debug
3111 the parent process and the child process will run unimpeded.
3113 If you want to follow the child process instead of the parent process,
3114 use the command @w{@code{set follow-fork-mode}}.
3117 @kindex set follow-fork-mode
3118 @item set follow-fork-mode @var{mode}
3119 Set the debugger response to a program call of @code{fork} or
3120 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3121 process. The @var{mode} argument can be:
3125 The original process is debugged after a fork. The child process runs
3126 unimpeded. This is the default.
3129 The new process is debugged after a fork. The parent process runs
3134 @kindex show follow-fork-mode
3135 @item show follow-fork-mode
3136 Display the current debugger response to a @code{fork} or @code{vfork} call.
3139 @cindex debugging multiple processes
3140 On Linux, if you want to debug both the parent and child processes, use the
3141 command @w{@code{set detach-on-fork}}.
3144 @kindex set detach-on-fork
3145 @item set detach-on-fork @var{mode}
3146 Tells gdb whether to detach one of the processes after a fork, or
3147 retain debugger control over them both.
3151 The child process (or parent process, depending on the value of
3152 @code{follow-fork-mode}) will be detached and allowed to run
3153 independently. This is the default.
3156 Both processes will be held under the control of @value{GDBN}.
3157 One process (child or parent, depending on the value of
3158 @code{follow-fork-mode}) is debugged as usual, while the other
3163 @kindex show detach-on-fork
3164 @item show detach-on-fork
3165 Show whether detach-on-fork mode is on/off.
3168 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3169 will retain control of all forked processes (including nested forks).
3170 You can list the forked processes under the control of @value{GDBN} by
3171 using the @w{@code{info inferiors}} command, and switch from one fork
3172 to another by using the @code{inferior} command (@pxref{Inferiors and
3173 Programs, ,Debugging Multiple Inferiors and Programs}).
3175 To quit debugging one of the forked processes, you can either detach
3176 from it by using the @w{@code{detach inferiors}} command (allowing it
3177 to run independently), or kill it using the @w{@code{kill inferiors}}
3178 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3181 If you ask to debug a child process and a @code{vfork} is followed by an
3182 @code{exec}, @value{GDBN} executes the new target up to the first
3183 breakpoint in the new target. If you have a breakpoint set on
3184 @code{main} in your original program, the breakpoint will also be set on
3185 the child process's @code{main}.
3187 On some systems, when a child process is spawned by @code{vfork}, you
3188 cannot debug the child or parent until an @code{exec} call completes.
3190 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3191 call executes, the new target restarts. To restart the parent
3192 process, use the @code{file} command with the parent executable name
3193 as its argument. By default, after an @code{exec} call executes,
3194 @value{GDBN} discards the symbols of the previous executable image.
3195 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3199 @kindex set follow-exec-mode
3200 @item set follow-exec-mode @var{mode}
3202 Set debugger response to a program call of @code{exec}. An
3203 @code{exec} call replaces the program image of a process.
3205 @code{follow-exec-mode} can be:
3209 @value{GDBN} creates a new inferior and rebinds the process to this
3210 new inferior. The program the process was running before the
3211 @code{exec} call can be restarted afterwards by restarting the
3217 (@value{GDBP}) info inferiors
3219 Id Description Executable
3222 process 12020 is executing new program: prog2
3223 Program exited normally.
3224 (@value{GDBP}) info inferiors
3225 Id Description Executable
3231 @value{GDBN} keeps the process bound to the same inferior. The new
3232 executable image replaces the previous executable loaded in the
3233 inferior. Restarting the inferior after the @code{exec} call, with
3234 e.g., the @code{run} command, restarts the executable the process was
3235 running after the @code{exec} call. This is the default mode.
3240 (@value{GDBP}) info inferiors
3241 Id Description Executable
3244 process 12020 is executing new program: prog2
3245 Program exited normally.
3246 (@value{GDBP}) info inferiors
3247 Id Description Executable
3254 You can use the @code{catch} command to make @value{GDBN} stop whenever
3255 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3256 Catchpoints, ,Setting Catchpoints}.
3258 @node Checkpoint/Restart
3259 @section Setting a @emph{Bookmark} to Return to Later
3264 @cindex snapshot of a process
3265 @cindex rewind program state
3267 On certain operating systems@footnote{Currently, only
3268 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3269 program's state, called a @dfn{checkpoint}, and come back to it
3272 Returning to a checkpoint effectively undoes everything that has
3273 happened in the program since the @code{checkpoint} was saved. This
3274 includes changes in memory, registers, and even (within some limits)
3275 system state. Effectively, it is like going back in time to the
3276 moment when the checkpoint was saved.
3278 Thus, if you're stepping thru a program and you think you're
3279 getting close to the point where things go wrong, you can save
3280 a checkpoint. Then, if you accidentally go too far and miss
3281 the critical statement, instead of having to restart your program
3282 from the beginning, you can just go back to the checkpoint and
3283 start again from there.
3285 This can be especially useful if it takes a lot of time or
3286 steps to reach the point where you think the bug occurs.
3288 To use the @code{checkpoint}/@code{restart} method of debugging:
3293 Save a snapshot of the debugged program's current execution state.
3294 The @code{checkpoint} command takes no arguments, but each checkpoint
3295 is assigned a small integer id, similar to a breakpoint id.
3297 @kindex info checkpoints
3298 @item info checkpoints
3299 List the checkpoints that have been saved in the current debugging
3300 session. For each checkpoint, the following information will be
3307 @item Source line, or label
3310 @kindex restart @var{checkpoint-id}
3311 @item restart @var{checkpoint-id}
3312 Restore the program state that was saved as checkpoint number
3313 @var{checkpoint-id}. All program variables, registers, stack frames
3314 etc.@: will be returned to the values that they had when the checkpoint
3315 was saved. In essence, gdb will ``wind back the clock'' to the point
3316 in time when the checkpoint was saved.
3318 Note that breakpoints, @value{GDBN} variables, command history etc.
3319 are not affected by restoring a checkpoint. In general, a checkpoint
3320 only restores things that reside in the program being debugged, not in
3323 @kindex delete checkpoint @var{checkpoint-id}
3324 @item delete checkpoint @var{checkpoint-id}
3325 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3329 Returning to a previously saved checkpoint will restore the user state
3330 of the program being debugged, plus a significant subset of the system
3331 (OS) state, including file pointers. It won't ``un-write'' data from
3332 a file, but it will rewind the file pointer to the previous location,
3333 so that the previously written data can be overwritten. For files
3334 opened in read mode, the pointer will also be restored so that the
3335 previously read data can be read again.
3337 Of course, characters that have been sent to a printer (or other
3338 external device) cannot be ``snatched back'', and characters received
3339 from eg.@: a serial device can be removed from internal program buffers,
3340 but they cannot be ``pushed back'' into the serial pipeline, ready to
3341 be received again. Similarly, the actual contents of files that have
3342 been changed cannot be restored (at this time).
3344 However, within those constraints, you actually can ``rewind'' your
3345 program to a previously saved point in time, and begin debugging it
3346 again --- and you can change the course of events so as to debug a
3347 different execution path this time.
3349 @cindex checkpoints and process id
3350 Finally, there is one bit of internal program state that will be
3351 different when you return to a checkpoint --- the program's process
3352 id. Each checkpoint will have a unique process id (or @var{pid}),
3353 and each will be different from the program's original @var{pid}.
3354 If your program has saved a local copy of its process id, this could
3355 potentially pose a problem.
3357 @subsection A Non-obvious Benefit of Using Checkpoints
3359 On some systems such as @sc{gnu}/Linux, address space randomization
3360 is performed on new processes for security reasons. This makes it
3361 difficult or impossible to set a breakpoint, or watchpoint, on an
3362 absolute address if you have to restart the program, since the
3363 absolute location of a symbol will change from one execution to the
3366 A checkpoint, however, is an @emph{identical} copy of a process.
3367 Therefore if you create a checkpoint at (eg.@:) the start of main,
3368 and simply return to that checkpoint instead of restarting the
3369 process, you can avoid the effects of address randomization and
3370 your symbols will all stay in the same place.
3373 @chapter Stopping and Continuing
3375 The principal purposes of using a debugger are so that you can stop your
3376 program before it terminates; or so that, if your program runs into
3377 trouble, you can investigate and find out why.
3379 Inside @value{GDBN}, your program may stop for any of several reasons,
3380 such as a signal, a breakpoint, or reaching a new line after a
3381 @value{GDBN} command such as @code{step}. You may then examine and
3382 change variables, set new breakpoints or remove old ones, and then
3383 continue execution. Usually, the messages shown by @value{GDBN} provide
3384 ample explanation of the status of your program---but you can also
3385 explicitly request this information at any time.
3388 @kindex info program
3390 Display information about the status of your program: whether it is
3391 running or not, what process it is, and why it stopped.
3395 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3396 * Continuing and Stepping:: Resuming execution
3397 * Skipping Over Functions and Files::
3398 Skipping over functions and files
3400 * Thread Stops:: Stopping and starting multi-thread programs
3404 @section Breakpoints, Watchpoints, and Catchpoints
3407 A @dfn{breakpoint} makes your program stop whenever a certain point in
3408 the program is reached. For each breakpoint, you can add conditions to
3409 control in finer detail whether your program stops. You can set
3410 breakpoints with the @code{break} command and its variants (@pxref{Set
3411 Breaks, ,Setting Breakpoints}), to specify the place where your program
3412 should stop by line number, function name or exact address in the
3415 On some systems, you can set breakpoints in shared libraries before
3416 the executable is run. There is a minor limitation on HP-UX systems:
3417 you must wait until the executable is run in order to set breakpoints
3418 in shared library routines that are not called directly by the program
3419 (for example, routines that are arguments in a @code{pthread_create}
3423 @cindex data breakpoints
3424 @cindex memory tracing
3425 @cindex breakpoint on memory address
3426 @cindex breakpoint on variable modification
3427 A @dfn{watchpoint} is a special breakpoint that stops your program
3428 when the value of an expression changes. The expression may be a value
3429 of a variable, or it could involve values of one or more variables
3430 combined by operators, such as @samp{a + b}. This is sometimes called
3431 @dfn{data breakpoints}. You must use a different command to set
3432 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3433 from that, you can manage a watchpoint like any other breakpoint: you
3434 enable, disable, and delete both breakpoints and watchpoints using the
3437 You can arrange to have values from your program displayed automatically
3438 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3442 @cindex breakpoint on events
3443 A @dfn{catchpoint} is another special breakpoint that stops your program
3444 when a certain kind of event occurs, such as the throwing of a C@t{++}
3445 exception or the loading of a library. As with watchpoints, you use a
3446 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3447 Catchpoints}), but aside from that, you can manage a catchpoint like any
3448 other breakpoint. (To stop when your program receives a signal, use the
3449 @code{handle} command; see @ref{Signals, ,Signals}.)
3451 @cindex breakpoint numbers
3452 @cindex numbers for breakpoints
3453 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3454 catchpoint when you create it; these numbers are successive integers
3455 starting with one. In many of the commands for controlling various
3456 features of breakpoints you use the breakpoint number to say which
3457 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3458 @dfn{disabled}; if disabled, it has no effect on your program until you
3461 @cindex breakpoint ranges
3462 @cindex ranges of breakpoints
3463 Some @value{GDBN} commands accept a range of breakpoints on which to
3464 operate. A breakpoint range is either a single breakpoint number, like
3465 @samp{5}, or two such numbers, in increasing order, separated by a
3466 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3467 all breakpoints in that range are operated on.
3470 * Set Breaks:: Setting breakpoints
3471 * Set Watchpoints:: Setting watchpoints
3472 * Set Catchpoints:: Setting catchpoints
3473 * Delete Breaks:: Deleting breakpoints
3474 * Disabling:: Disabling breakpoints
3475 * Conditions:: Break conditions
3476 * Break Commands:: Breakpoint command lists
3477 * Dynamic Printf:: Dynamic printf
3478 * Save Breakpoints:: How to save breakpoints in a file
3479 * Static Probe Points:: Listing static probe points
3480 * Error in Breakpoints:: ``Cannot insert breakpoints''
3481 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3485 @subsection Setting Breakpoints
3487 @c FIXME LMB what does GDB do if no code on line of breakpt?
3488 @c consider in particular declaration with/without initialization.
3490 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3493 @kindex b @r{(@code{break})}
3494 @vindex $bpnum@r{, convenience variable}
3495 @cindex latest breakpoint
3496 Breakpoints are set with the @code{break} command (abbreviated
3497 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3498 number of the breakpoint you've set most recently; see @ref{Convenience
3499 Vars,, Convenience Variables}, for a discussion of what you can do with
3500 convenience variables.
3503 @item break @var{location}
3504 Set a breakpoint at the given @var{location}, which can specify a
3505 function name, a line number, or an address of an instruction.
3506 (@xref{Specify Location}, for a list of all the possible ways to
3507 specify a @var{location}.) The breakpoint will stop your program just
3508 before it executes any of the code in the specified @var{location}.
3510 When using source languages that permit overloading of symbols, such as
3511 C@t{++}, a function name may refer to more than one possible place to break.
3512 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3515 It is also possible to insert a breakpoint that will stop the program
3516 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3517 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3520 When called without any arguments, @code{break} sets a breakpoint at
3521 the next instruction to be executed in the selected stack frame
3522 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3523 innermost, this makes your program stop as soon as control
3524 returns to that frame. This is similar to the effect of a
3525 @code{finish} command in the frame inside the selected frame---except
3526 that @code{finish} does not leave an active breakpoint. If you use
3527 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3528 the next time it reaches the current location; this may be useful
3531 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3532 least one instruction has been executed. If it did not do this, you
3533 would be unable to proceed past a breakpoint without first disabling the
3534 breakpoint. This rule applies whether or not the breakpoint already
3535 existed when your program stopped.
3537 @item break @dots{} if @var{cond}
3538 Set a breakpoint with condition @var{cond}; evaluate the expression
3539 @var{cond} each time the breakpoint is reached, and stop only if the
3540 value is nonzero---that is, if @var{cond} evaluates as true.
3541 @samp{@dots{}} stands for one of the possible arguments described
3542 above (or no argument) specifying where to break. @xref{Conditions,
3543 ,Break Conditions}, for more information on breakpoint conditions.
3546 @item tbreak @var{args}
3547 Set a breakpoint enabled only for one stop. The @var{args} are the
3548 same as for the @code{break} command, and the breakpoint is set in the same
3549 way, but the breakpoint is automatically deleted after the first time your
3550 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3553 @cindex hardware breakpoints
3554 @item hbreak @var{args}
3555 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
3556 @code{break} command and the breakpoint is set in the same way, but the
3557 breakpoint requires hardware support and some target hardware may not
3558 have this support. The main purpose of this is EPROM/ROM code
3559 debugging, so you can set a breakpoint at an instruction without
3560 changing the instruction. This can be used with the new trap-generation
3561 provided by SPARClite DSU and most x86-based targets. These targets
3562 will generate traps when a program accesses some data or instruction
3563 address that is assigned to the debug registers. However the hardware
3564 breakpoint registers can take a limited number of breakpoints. For
3565 example, on the DSU, only two data breakpoints can be set at a time, and
3566 @value{GDBN} will reject this command if more than two are used. Delete
3567 or disable unused hardware breakpoints before setting new ones
3568 (@pxref{Disabling, ,Disabling Breakpoints}).
3569 @xref{Conditions, ,Break Conditions}.
3570 For remote targets, you can restrict the number of hardware
3571 breakpoints @value{GDBN} will use, see @ref{set remote
3572 hardware-breakpoint-limit}.
3575 @item thbreak @var{args}
3576 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
3577 are the same as for the @code{hbreak} command and the breakpoint is set in
3578 the same way. However, like the @code{tbreak} command,
3579 the breakpoint is automatically deleted after the
3580 first time your program stops there. Also, like the @code{hbreak}
3581 command, the breakpoint requires hardware support and some target hardware
3582 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3583 See also @ref{Conditions, ,Break Conditions}.
3586 @cindex regular expression
3587 @cindex breakpoints at functions matching a regexp
3588 @cindex set breakpoints in many functions
3589 @item rbreak @var{regex}
3590 Set breakpoints on all functions matching the regular expression
3591 @var{regex}. This command sets an unconditional breakpoint on all
3592 matches, printing a list of all breakpoints it set. Once these
3593 breakpoints are set, they are treated just like the breakpoints set with
3594 the @code{break} command. You can delete them, disable them, or make
3595 them conditional the same way as any other breakpoint.
3597 The syntax of the regular expression is the standard one used with tools
3598 like @file{grep}. Note that this is different from the syntax used by
3599 shells, so for instance @code{foo*} matches all functions that include
3600 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3601 @code{.*} leading and trailing the regular expression you supply, so to
3602 match only functions that begin with @code{foo}, use @code{^foo}.
3604 @cindex non-member C@t{++} functions, set breakpoint in
3605 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3606 breakpoints on overloaded functions that are not members of any special
3609 @cindex set breakpoints on all functions
3610 The @code{rbreak} command can be used to set breakpoints in
3611 @strong{all} the functions in a program, like this:
3614 (@value{GDBP}) rbreak .
3617 @item rbreak @var{file}:@var{regex}
3618 If @code{rbreak} is called with a filename qualification, it limits
3619 the search for functions matching the given regular expression to the
3620 specified @var{file}. This can be used, for example, to set breakpoints on
3621 every function in a given file:
3624 (@value{GDBP}) rbreak file.c:.
3627 The colon separating the filename qualifier from the regex may
3628 optionally be surrounded by spaces.
3630 @kindex info breakpoints
3631 @cindex @code{$_} and @code{info breakpoints}
3632 @item info breakpoints @r{[}@var{n}@dots{}@r{]}
3633 @itemx info break @r{[}@var{n}@dots{}@r{]}
3634 Print a table of all breakpoints, watchpoints, and catchpoints set and
3635 not deleted. Optional argument @var{n} means print information only
3636 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
3637 For each breakpoint, following columns are printed:
3640 @item Breakpoint Numbers
3642 Breakpoint, watchpoint, or catchpoint.
3644 Whether the breakpoint is marked to be disabled or deleted when hit.
3645 @item Enabled or Disabled
3646 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3647 that are not enabled.
3649 Where the breakpoint is in your program, as a memory address. For a
3650 pending breakpoint whose address is not yet known, this field will
3651 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3652 library that has the symbol or line referred by breakpoint is loaded.
3653 See below for details. A breakpoint with several locations will
3654 have @samp{<MULTIPLE>} in this field---see below for details.
3656 Where the breakpoint is in the source for your program, as a file and
3657 line number. For a pending breakpoint, the original string passed to
3658 the breakpoint command will be listed as it cannot be resolved until
3659 the appropriate shared library is loaded in the future.
3663 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
3664 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
3665 @value{GDBN} on the host's side. If it is ``target'', then the condition
3666 is evaluated by the target. The @code{info break} command shows
3667 the condition on the line following the affected breakpoint, together with
3668 its condition evaluation mode in between parentheses.
3670 Breakpoint commands, if any, are listed after that. A pending breakpoint is
3671 allowed to have a condition specified for it. The condition is not parsed for
3672 validity until a shared library is loaded that allows the pending
3673 breakpoint to resolve to a valid location.
3676 @code{info break} with a breakpoint
3677 number @var{n} as argument lists only that breakpoint. The
3678 convenience variable @code{$_} and the default examining-address for
3679 the @code{x} command are set to the address of the last breakpoint
3680 listed (@pxref{Memory, ,Examining Memory}).
3683 @code{info break} displays a count of the number of times the breakpoint
3684 has been hit. This is especially useful in conjunction with the
3685 @code{ignore} command. You can ignore a large number of breakpoint
3686 hits, look at the breakpoint info to see how many times the breakpoint
3687 was hit, and then run again, ignoring one less than that number. This
3688 will get you quickly to the last hit of that breakpoint.
3691 For a breakpoints with an enable count (xref) greater than 1,
3692 @code{info break} also displays that count.
3696 @value{GDBN} allows you to set any number of breakpoints at the same place in
3697 your program. There is nothing silly or meaningless about this. When
3698 the breakpoints are conditional, this is even useful
3699 (@pxref{Conditions, ,Break Conditions}).
3701 @cindex multiple locations, breakpoints
3702 @cindex breakpoints, multiple locations
3703 It is possible that a breakpoint corresponds to several locations
3704 in your program. Examples of this situation are:
3708 Multiple functions in the program may have the same name.
3711 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3712 instances of the function body, used in different cases.
3715 For a C@t{++} template function, a given line in the function can
3716 correspond to any number of instantiations.
3719 For an inlined function, a given source line can correspond to
3720 several places where that function is inlined.
3723 In all those cases, @value{GDBN} will insert a breakpoint at all
3724 the relevant locations.
3726 A breakpoint with multiple locations is displayed in the breakpoint
3727 table using several rows---one header row, followed by one row for
3728 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3729 address column. The rows for individual locations contain the actual
3730 addresses for locations, and show the functions to which those
3731 locations belong. The number column for a location is of the form
3732 @var{breakpoint-number}.@var{location-number}.
3737 Num Type Disp Enb Address What
3738 1 breakpoint keep y <MULTIPLE>
3740 breakpoint already hit 1 time
3741 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3742 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3745 Each location can be individually enabled or disabled by passing
3746 @var{breakpoint-number}.@var{location-number} as argument to the
3747 @code{enable} and @code{disable} commands. Note that you cannot
3748 delete the individual locations from the list, you can only delete the
3749 entire list of locations that belong to their parent breakpoint (with
3750 the @kbd{delete @var{num}} command, where @var{num} is the number of
3751 the parent breakpoint, 1 in the above example). Disabling or enabling
3752 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3753 that belong to that breakpoint.
3755 @cindex pending breakpoints
3756 It's quite common to have a breakpoint inside a shared library.
3757 Shared libraries can be loaded and unloaded explicitly,
3758 and possibly repeatedly, as the program is executed. To support
3759 this use case, @value{GDBN} updates breakpoint locations whenever
3760 any shared library is loaded or unloaded. Typically, you would
3761 set a breakpoint in a shared library at the beginning of your
3762 debugging session, when the library is not loaded, and when the
3763 symbols from the library are not available. When you try to set
3764 breakpoint, @value{GDBN} will ask you if you want to set
3765 a so called @dfn{pending breakpoint}---breakpoint whose address
3766 is not yet resolved.
3768 After the program is run, whenever a new shared library is loaded,
3769 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3770 shared library contains the symbol or line referred to by some
3771 pending breakpoint, that breakpoint is resolved and becomes an
3772 ordinary breakpoint. When a library is unloaded, all breakpoints
3773 that refer to its symbols or source lines become pending again.
3775 This logic works for breakpoints with multiple locations, too. For
3776 example, if you have a breakpoint in a C@t{++} template function, and
3777 a newly loaded shared library has an instantiation of that template,
3778 a new location is added to the list of locations for the breakpoint.
3780 Except for having unresolved address, pending breakpoints do not
3781 differ from regular breakpoints. You can set conditions or commands,
3782 enable and disable them and perform other breakpoint operations.
3784 @value{GDBN} provides some additional commands for controlling what
3785 happens when the @samp{break} command cannot resolve breakpoint
3786 address specification to an address:
3788 @kindex set breakpoint pending
3789 @kindex show breakpoint pending
3791 @item set breakpoint pending auto
3792 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3793 location, it queries you whether a pending breakpoint should be created.
3795 @item set breakpoint pending on
3796 This indicates that an unrecognized breakpoint location should automatically
3797 result in a pending breakpoint being created.
3799 @item set breakpoint pending off
3800 This indicates that pending breakpoints are not to be created. Any
3801 unrecognized breakpoint location results in an error. This setting does
3802 not affect any pending breakpoints previously created.
3804 @item show breakpoint pending
3805 Show the current behavior setting for creating pending breakpoints.
3808 The settings above only affect the @code{break} command and its
3809 variants. Once breakpoint is set, it will be automatically updated
3810 as shared libraries are loaded and unloaded.
3812 @cindex automatic hardware breakpoints
3813 For some targets, @value{GDBN} can automatically decide if hardware or
3814 software breakpoints should be used, depending on whether the
3815 breakpoint address is read-only or read-write. This applies to
3816 breakpoints set with the @code{break} command as well as to internal
3817 breakpoints set by commands like @code{next} and @code{finish}. For
3818 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3821 You can control this automatic behaviour with the following commands::
3823 @kindex set breakpoint auto-hw
3824 @kindex show breakpoint auto-hw
3826 @item set breakpoint auto-hw on
3827 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3828 will try to use the target memory map to decide if software or hardware
3829 breakpoint must be used.
3831 @item set breakpoint auto-hw off
3832 This indicates @value{GDBN} should not automatically select breakpoint
3833 type. If the target provides a memory map, @value{GDBN} will warn when
3834 trying to set software breakpoint at a read-only address.
3837 @value{GDBN} normally implements breakpoints by replacing the program code
3838 at the breakpoint address with a special instruction, which, when
3839 executed, given control to the debugger. By default, the program
3840 code is so modified only when the program is resumed. As soon as
3841 the program stops, @value{GDBN} restores the original instructions. This
3842 behaviour guards against leaving breakpoints inserted in the
3843 target should gdb abrubptly disconnect. However, with slow remote
3844 targets, inserting and removing breakpoint can reduce the performance.
3845 This behavior can be controlled with the following commands::
3847 @kindex set breakpoint always-inserted
3848 @kindex show breakpoint always-inserted
3850 @item set breakpoint always-inserted off
3851 All breakpoints, including newly added by the user, are inserted in
3852 the target only when the target is resumed. All breakpoints are
3853 removed from the target when it stops. This is the default mode.
3855 @item set breakpoint always-inserted on
3856 Causes all breakpoints to be inserted in the target at all times. If
3857 the user adds a new breakpoint, or changes an existing breakpoint, the
3858 breakpoints in the target are updated immediately. A breakpoint is
3859 removed from the target only when breakpoint itself is deleted.
3862 @value{GDBN} handles conditional breakpoints by evaluating these conditions
3863 when a breakpoint breaks. If the condition is true, then the process being
3864 debugged stops, otherwise the process is resumed.
3866 If the target supports evaluating conditions on its end, @value{GDBN} may
3867 download the breakpoint, together with its conditions, to it.
3869 This feature can be controlled via the following commands:
3871 @kindex set breakpoint condition-evaluation
3872 @kindex show breakpoint condition-evaluation
3874 @item set breakpoint condition-evaluation host
3875 This option commands @value{GDBN} to evaluate the breakpoint
3876 conditions on the host's side. Unconditional breakpoints are sent to
3877 the target which in turn receives the triggers and reports them back to GDB
3878 for condition evaluation. This is the standard evaluation mode.
3880 @item set breakpoint condition-evaluation target
3881 This option commands @value{GDBN} to download breakpoint conditions
3882 to the target at the moment of their insertion. The target
3883 is responsible for evaluating the conditional expression and reporting
3884 breakpoint stop events back to @value{GDBN} whenever the condition
3885 is true. Due to limitations of target-side evaluation, some conditions
3886 cannot be evaluated there, e.g., conditions that depend on local data
3887 that is only known to the host. Examples include
3888 conditional expressions involving convenience variables, complex types
3889 that cannot be handled by the agent expression parser and expressions
3890 that are too long to be sent over to the target, specially when the
3891 target is a remote system. In these cases, the conditions will be
3892 evaluated by @value{GDBN}.
3894 @item set breakpoint condition-evaluation auto
3895 This is the default mode. If the target supports evaluating breakpoint
3896 conditions on its end, @value{GDBN} will download breakpoint conditions to
3897 the target (limitations mentioned previously apply). If the target does
3898 not support breakpoint condition evaluation, then @value{GDBN} will fallback
3899 to evaluating all these conditions on the host's side.
3903 @cindex negative breakpoint numbers
3904 @cindex internal @value{GDBN} breakpoints
3905 @value{GDBN} itself sometimes sets breakpoints in your program for
3906 special purposes, such as proper handling of @code{longjmp} (in C
3907 programs). These internal breakpoints are assigned negative numbers,
3908 starting with @code{-1}; @samp{info breakpoints} does not display them.
3909 You can see these breakpoints with the @value{GDBN} maintenance command
3910 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3913 @node Set Watchpoints
3914 @subsection Setting Watchpoints
3916 @cindex setting watchpoints
3917 You can use a watchpoint to stop execution whenever the value of an
3918 expression changes, without having to predict a particular place where
3919 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3920 The expression may be as simple as the value of a single variable, or
3921 as complex as many variables combined by operators. Examples include:
3925 A reference to the value of a single variable.
3928 An address cast to an appropriate data type. For example,
3929 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3930 address (assuming an @code{int} occupies 4 bytes).
3933 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3934 expression can use any operators valid in the program's native
3935 language (@pxref{Languages}).
3938 You can set a watchpoint on an expression even if the expression can
3939 not be evaluated yet. For instance, you can set a watchpoint on
3940 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3941 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3942 the expression produces a valid value. If the expression becomes
3943 valid in some other way than changing a variable (e.g.@: if the memory
3944 pointed to by @samp{*global_ptr} becomes readable as the result of a
3945 @code{malloc} call), @value{GDBN} may not stop until the next time
3946 the expression changes.
3948 @cindex software watchpoints
3949 @cindex hardware watchpoints
3950 Depending on your system, watchpoints may be implemented in software or
3951 hardware. @value{GDBN} does software watchpointing by single-stepping your
3952 program and testing the variable's value each time, which is hundreds of
3953 times slower than normal execution. (But this may still be worth it, to
3954 catch errors where you have no clue what part of your program is the
3957 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3958 x86-based targets, @value{GDBN} includes support for hardware
3959 watchpoints, which do not slow down the running of your program.
3963 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3964 Set a watchpoint for an expression. @value{GDBN} will break when the
3965 expression @var{expr} is written into by the program and its value
3966 changes. The simplest (and the most popular) use of this command is
3967 to watch the value of a single variable:
3970 (@value{GDBP}) watch foo
3973 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3974 argument, @value{GDBN} breaks only when the thread identified by
3975 @var{threadnum} changes the value of @var{expr}. If any other threads
3976 change the value of @var{expr}, @value{GDBN} will not break. Note
3977 that watchpoints restricted to a single thread in this way only work
3978 with Hardware Watchpoints.
3980 Ordinarily a watchpoint respects the scope of variables in @var{expr}
3981 (see below). The @code{-location} argument tells @value{GDBN} to
3982 instead watch the memory referred to by @var{expr}. In this case,
3983 @value{GDBN} will evaluate @var{expr}, take the address of the result,
3984 and watch the memory at that address. The type of the result is used
3985 to determine the size of the watched memory. If the expression's
3986 result does not have an address, then @value{GDBN} will print an
3989 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
3990 of masked watchpoints, if the current architecture supports this
3991 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
3992 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
3993 to an address to watch. The mask specifies that some bits of an address
3994 (the bits which are reset in the mask) should be ignored when matching
3995 the address accessed by the inferior against the watchpoint address.
3996 Thus, a masked watchpoint watches many addresses simultaneously---those
3997 addresses whose unmasked bits are identical to the unmasked bits in the
3998 watchpoint address. The @code{mask} argument implies @code{-location}.
4002 (@value{GDBP}) watch foo mask 0xffff00ff
4003 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4007 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
4008 Set a watchpoint that will break when the value of @var{expr} is read
4012 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
4013 Set a watchpoint that will break when @var{expr} is either read from
4014 or written into by the program.
4016 @kindex info watchpoints @r{[}@var{n}@dots{}@r{]}
4017 @item info watchpoints @r{[}@var{n}@dots{}@r{]}
4018 This command prints a list of watchpoints, using the same format as
4019 @code{info break} (@pxref{Set Breaks}).
4022 If you watch for a change in a numerically entered address you need to
4023 dereference it, as the address itself is just a constant number which will
4024 never change. @value{GDBN} refuses to create a watchpoint that watches
4025 a never-changing value:
4028 (@value{GDBP}) watch 0x600850
4029 Cannot watch constant value 0x600850.
4030 (@value{GDBP}) watch *(int *) 0x600850
4031 Watchpoint 1: *(int *) 6293584
4034 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4035 watchpoints execute very quickly, and the debugger reports a change in
4036 value at the exact instruction where the change occurs. If @value{GDBN}
4037 cannot set a hardware watchpoint, it sets a software watchpoint, which
4038 executes more slowly and reports the change in value at the next
4039 @emph{statement}, not the instruction, after the change occurs.
4041 @cindex use only software watchpoints
4042 You can force @value{GDBN} to use only software watchpoints with the
4043 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4044 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4045 the underlying system supports them. (Note that hardware-assisted
4046 watchpoints that were set @emph{before} setting
4047 @code{can-use-hw-watchpoints} to zero will still use the hardware
4048 mechanism of watching expression values.)
4051 @item set can-use-hw-watchpoints
4052 @kindex set can-use-hw-watchpoints
4053 Set whether or not to use hardware watchpoints.
4055 @item show can-use-hw-watchpoints
4056 @kindex show can-use-hw-watchpoints
4057 Show the current mode of using hardware watchpoints.
4060 For remote targets, you can restrict the number of hardware
4061 watchpoints @value{GDBN} will use, see @ref{set remote
4062 hardware-breakpoint-limit}.
4064 When you issue the @code{watch} command, @value{GDBN} reports
4067 Hardware watchpoint @var{num}: @var{expr}
4071 if it was able to set a hardware watchpoint.
4073 Currently, the @code{awatch} and @code{rwatch} commands can only set
4074 hardware watchpoints, because accesses to data that don't change the
4075 value of the watched expression cannot be detected without examining
4076 every instruction as it is being executed, and @value{GDBN} does not do
4077 that currently. If @value{GDBN} finds that it is unable to set a
4078 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4079 will print a message like this:
4082 Expression cannot be implemented with read/access watchpoint.
4085 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4086 data type of the watched expression is wider than what a hardware
4087 watchpoint on the target machine can handle. For example, some systems
4088 can only watch regions that are up to 4 bytes wide; on such systems you
4089 cannot set hardware watchpoints for an expression that yields a
4090 double-precision floating-point number (which is typically 8 bytes
4091 wide). As a work-around, it might be possible to break the large region
4092 into a series of smaller ones and watch them with separate watchpoints.
4094 If you set too many hardware watchpoints, @value{GDBN} might be unable
4095 to insert all of them when you resume the execution of your program.
4096 Since the precise number of active watchpoints is unknown until such
4097 time as the program is about to be resumed, @value{GDBN} might not be
4098 able to warn you about this when you set the watchpoints, and the
4099 warning will be printed only when the program is resumed:
4102 Hardware watchpoint @var{num}: Could not insert watchpoint
4106 If this happens, delete or disable some of the watchpoints.
4108 Watching complex expressions that reference many variables can also
4109 exhaust the resources available for hardware-assisted watchpoints.
4110 That's because @value{GDBN} needs to watch every variable in the
4111 expression with separately allocated resources.
4113 If you call a function interactively using @code{print} or @code{call},
4114 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4115 kind of breakpoint or the call completes.
4117 @value{GDBN} automatically deletes watchpoints that watch local
4118 (automatic) variables, or expressions that involve such variables, when
4119 they go out of scope, that is, when the execution leaves the block in
4120 which these variables were defined. In particular, when the program
4121 being debugged terminates, @emph{all} local variables go out of scope,
4122 and so only watchpoints that watch global variables remain set. If you
4123 rerun the program, you will need to set all such watchpoints again. One
4124 way of doing that would be to set a code breakpoint at the entry to the
4125 @code{main} function and when it breaks, set all the watchpoints.
4127 @cindex watchpoints and threads
4128 @cindex threads and watchpoints
4129 In multi-threaded programs, watchpoints will detect changes to the
4130 watched expression from every thread.
4133 @emph{Warning:} In multi-threaded programs, software watchpoints
4134 have only limited usefulness. If @value{GDBN} creates a software
4135 watchpoint, it can only watch the value of an expression @emph{in a
4136 single thread}. If you are confident that the expression can only
4137 change due to the current thread's activity (and if you are also
4138 confident that no other thread can become current), then you can use
4139 software watchpoints as usual. However, @value{GDBN} may not notice
4140 when a non-current thread's activity changes the expression. (Hardware
4141 watchpoints, in contrast, watch an expression in all threads.)
4144 @xref{set remote hardware-watchpoint-limit}.
4146 @node Set Catchpoints
4147 @subsection Setting Catchpoints
4148 @cindex catchpoints, setting
4149 @cindex exception handlers
4150 @cindex event handling
4152 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4153 kinds of program events, such as C@t{++} exceptions or the loading of a
4154 shared library. Use the @code{catch} command to set a catchpoint.
4158 @item catch @var{event}
4159 Stop when @var{event} occurs. The @var{event} can be any of the following:
4162 @item throw @r{[}@var{regexp}@r{]}
4163 @itemx rethrow @r{[}@var{regexp}@r{]}
4164 @itemx catch @r{[}@var{regexp}@r{]}
4166 @kindex catch rethrow
4168 @cindex stop on C@t{++} exceptions
4169 The throwing, re-throwing, or catching of a C@t{++} exception.
4171 If @var{regexp} is given, then only exceptions whose type matches the
4172 regular expression will be caught.
4174 @vindex $_exception@r{, convenience variable}
4175 The convenience variable @code{$_exception} is available at an
4176 exception-related catchpoint, on some systems. This holds the
4177 exception being thrown.
4179 There are currently some limitations to C@t{++} exception handling in
4184 The support for these commands is system-dependent. Currently, only
4185 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4189 The regular expression feature and the @code{$_exception} convenience
4190 variable rely on the presence of some SDT probes in @code{libstdc++}.
4191 If these probes are not present, then these features cannot be used.
4192 These probes were first available in the GCC 4.8 release, but whether
4193 or not they are available in your GCC also depends on how it was
4197 The @code{$_exception} convenience variable is only valid at the
4198 instruction at which an exception-related catchpoint is set.
4201 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4202 location in the system library which implements runtime exception
4203 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4204 (@pxref{Selection}) to get to your code.
4207 If you call a function interactively, @value{GDBN} normally returns
4208 control to you when the function has finished executing. If the call
4209 raises an exception, however, the call may bypass the mechanism that
4210 returns control to you and cause your program either to abort or to
4211 simply continue running until it hits a breakpoint, catches a signal
4212 that @value{GDBN} is listening for, or exits. This is the case even if
4213 you set a catchpoint for the exception; catchpoints on exceptions are
4214 disabled within interactive calls. @xref{Calling}, for information on
4215 controlling this with @code{set unwind-on-terminating-exception}.
4218 You cannot raise an exception interactively.
4221 You cannot install an exception handler interactively.
4225 @kindex catch exception
4226 @cindex Ada exception catching
4227 @cindex catch Ada exceptions
4228 An Ada exception being raised. If an exception name is specified
4229 at the end of the command (eg @code{catch exception Program_Error}),
4230 the debugger will stop only when this specific exception is raised.
4231 Otherwise, the debugger stops execution when any Ada exception is raised.
4233 When inserting an exception catchpoint on a user-defined exception whose
4234 name is identical to one of the exceptions defined by the language, the
4235 fully qualified name must be used as the exception name. Otherwise,
4236 @value{GDBN} will assume that it should stop on the pre-defined exception
4237 rather than the user-defined one. For instance, assuming an exception
4238 called @code{Constraint_Error} is defined in package @code{Pck}, then
4239 the command to use to catch such exceptions is @kbd{catch exception
4240 Pck.Constraint_Error}.
4242 @item exception unhandled
4243 @kindex catch exception unhandled
4244 An exception that was raised but is not handled by the program.
4247 @kindex catch assert
4248 A failed Ada assertion.
4252 @cindex break on fork/exec
4253 A call to @code{exec}. This is currently only available for HP-UX
4257 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @dots{}
4258 @kindex catch syscall
4259 @cindex break on a system call.
4260 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4261 syscall is a mechanism for application programs to request a service
4262 from the operating system (OS) or one of the OS system services.
4263 @value{GDBN} can catch some or all of the syscalls issued by the
4264 debuggee, and show the related information for each syscall. If no
4265 argument is specified, calls to and returns from all system calls
4268 @var{name} can be any system call name that is valid for the
4269 underlying OS. Just what syscalls are valid depends on the OS. On
4270 GNU and Unix systems, you can find the full list of valid syscall
4271 names on @file{/usr/include/asm/unistd.h}.
4273 @c For MS-Windows, the syscall names and the corresponding numbers
4274 @c can be found, e.g., on this URL:
4275 @c http://www.metasploit.com/users/opcode/syscalls.html
4276 @c but we don't support Windows syscalls yet.
4278 Normally, @value{GDBN} knows in advance which syscalls are valid for
4279 each OS, so you can use the @value{GDBN} command-line completion
4280 facilities (@pxref{Completion,, command completion}) to list the
4283 You may also specify the system call numerically. A syscall's
4284 number is the value passed to the OS's syscall dispatcher to
4285 identify the requested service. When you specify the syscall by its
4286 name, @value{GDBN} uses its database of syscalls to convert the name
4287 into the corresponding numeric code, but using the number directly
4288 may be useful if @value{GDBN}'s database does not have the complete
4289 list of syscalls on your system (e.g., because @value{GDBN} lags
4290 behind the OS upgrades).
4292 The example below illustrates how this command works if you don't provide
4296 (@value{GDBP}) catch syscall
4297 Catchpoint 1 (syscall)
4299 Starting program: /tmp/catch-syscall
4301 Catchpoint 1 (call to syscall 'close'), \
4302 0xffffe424 in __kernel_vsyscall ()
4306 Catchpoint 1 (returned from syscall 'close'), \
4307 0xffffe424 in __kernel_vsyscall ()
4311 Here is an example of catching a system call by name:
4314 (@value{GDBP}) catch syscall chroot
4315 Catchpoint 1 (syscall 'chroot' [61])
4317 Starting program: /tmp/catch-syscall
4319 Catchpoint 1 (call to syscall 'chroot'), \
4320 0xffffe424 in __kernel_vsyscall ()
4324 Catchpoint 1 (returned from syscall 'chroot'), \
4325 0xffffe424 in __kernel_vsyscall ()
4329 An example of specifying a system call numerically. In the case
4330 below, the syscall number has a corresponding entry in the XML
4331 file, so @value{GDBN} finds its name and prints it:
4334 (@value{GDBP}) catch syscall 252
4335 Catchpoint 1 (syscall(s) 'exit_group')
4337 Starting program: /tmp/catch-syscall
4339 Catchpoint 1 (call to syscall 'exit_group'), \
4340 0xffffe424 in __kernel_vsyscall ()
4344 Program exited normally.
4348 However, there can be situations when there is no corresponding name
4349 in XML file for that syscall number. In this case, @value{GDBN} prints
4350 a warning message saying that it was not able to find the syscall name,
4351 but the catchpoint will be set anyway. See the example below:
4354 (@value{GDBP}) catch syscall 764
4355 warning: The number '764' does not represent a known syscall.
4356 Catchpoint 2 (syscall 764)
4360 If you configure @value{GDBN} using the @samp{--without-expat} option,
4361 it will not be able to display syscall names. Also, if your
4362 architecture does not have an XML file describing its system calls,
4363 you will not be able to see the syscall names. It is important to
4364 notice that these two features are used for accessing the syscall
4365 name database. In either case, you will see a warning like this:
4368 (@value{GDBP}) catch syscall
4369 warning: Could not open "syscalls/i386-linux.xml"
4370 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4371 GDB will not be able to display syscall names.
4372 Catchpoint 1 (syscall)
4376 Of course, the file name will change depending on your architecture and system.
4378 Still using the example above, you can also try to catch a syscall by its
4379 number. In this case, you would see something like:
4382 (@value{GDBP}) catch syscall 252
4383 Catchpoint 1 (syscall(s) 252)
4386 Again, in this case @value{GDBN} would not be able to display syscall's names.
4390 A call to @code{fork}. This is currently only available for HP-UX
4395 A call to @code{vfork}. This is currently only available for HP-UX
4398 @item load @r{[}regexp@r{]}
4399 @itemx unload @r{[}regexp@r{]}
4401 @kindex catch unload
4402 The loading or unloading of a shared library. If @var{regexp} is
4403 given, then the catchpoint will stop only if the regular expression
4404 matches one of the affected libraries.
4406 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
4407 @kindex catch signal
4408 The delivery of a signal.
4410 With no arguments, this catchpoint will catch any signal that is not
4411 used internally by @value{GDBN}, specifically, all signals except
4412 @samp{SIGTRAP} and @samp{SIGINT}.
4414 With the argument @samp{all}, all signals, including those used by
4415 @value{GDBN}, will be caught. This argument cannot be used with other
4418 Otherwise, the arguments are a list of signal names as given to
4419 @code{handle} (@pxref{Signals}). Only signals specified in this list
4422 One reason that @code{catch signal} can be more useful than
4423 @code{handle} is that you can attach commands and conditions to the
4426 When a signal is caught by a catchpoint, the signal's @code{stop} and
4427 @code{print} settings, as specified by @code{handle}, are ignored.
4428 However, whether the signal is still delivered to the inferior depends
4429 on the @code{pass} setting; this can be changed in the catchpoint's
4434 @item tcatch @var{event}
4436 Set a catchpoint that is enabled only for one stop. The catchpoint is
4437 automatically deleted after the first time the event is caught.
4441 Use the @code{info break} command to list the current catchpoints.
4445 @subsection Deleting Breakpoints
4447 @cindex clearing breakpoints, watchpoints, catchpoints
4448 @cindex deleting breakpoints, watchpoints, catchpoints
4449 It is often necessary to eliminate a breakpoint, watchpoint, or
4450 catchpoint once it has done its job and you no longer want your program
4451 to stop there. This is called @dfn{deleting} the breakpoint. A
4452 breakpoint that has been deleted no longer exists; it is forgotten.
4454 With the @code{clear} command you can delete breakpoints according to
4455 where they are in your program. With the @code{delete} command you can
4456 delete individual breakpoints, watchpoints, or catchpoints by specifying
4457 their breakpoint numbers.
4459 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4460 automatically ignores breakpoints on the first instruction to be executed
4461 when you continue execution without changing the execution address.
4466 Delete any breakpoints at the next instruction to be executed in the
4467 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4468 the innermost frame is selected, this is a good way to delete a
4469 breakpoint where your program just stopped.
4471 @item clear @var{location}
4472 Delete any breakpoints set at the specified @var{location}.
4473 @xref{Specify Location}, for the various forms of @var{location}; the
4474 most useful ones are listed below:
4477 @item clear @var{function}
4478 @itemx clear @var{filename}:@var{function}
4479 Delete any breakpoints set at entry to the named @var{function}.
4481 @item clear @var{linenum}
4482 @itemx clear @var{filename}:@var{linenum}
4483 Delete any breakpoints set at or within the code of the specified
4484 @var{linenum} of the specified @var{filename}.
4487 @cindex delete breakpoints
4489 @kindex d @r{(@code{delete})}
4490 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4491 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4492 ranges specified as arguments. If no argument is specified, delete all
4493 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4494 confirm off}). You can abbreviate this command as @code{d}.
4498 @subsection Disabling Breakpoints
4500 @cindex enable/disable a breakpoint
4501 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4502 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4503 it had been deleted, but remembers the information on the breakpoint so
4504 that you can @dfn{enable} it again later.
4506 You disable and enable breakpoints, watchpoints, and catchpoints with
4507 the @code{enable} and @code{disable} commands, optionally specifying
4508 one or more breakpoint numbers as arguments. Use @code{info break} to
4509 print a list of all breakpoints, watchpoints, and catchpoints if you
4510 do not know which numbers to use.
4512 Disabling and enabling a breakpoint that has multiple locations
4513 affects all of its locations.
4515 A breakpoint, watchpoint, or catchpoint can have any of several
4516 different states of enablement:
4520 Enabled. The breakpoint stops your program. A breakpoint set
4521 with the @code{break} command starts out in this state.
4523 Disabled. The breakpoint has no effect on your program.
4525 Enabled once. The breakpoint stops your program, but then becomes
4528 Enabled for a count. The breakpoint stops your program for the next
4529 N times, then becomes disabled.
4531 Enabled for deletion. The breakpoint stops your program, but
4532 immediately after it does so it is deleted permanently. A breakpoint
4533 set with the @code{tbreak} command starts out in this state.
4536 You can use the following commands to enable or disable breakpoints,
4537 watchpoints, and catchpoints:
4541 @kindex dis @r{(@code{disable})}
4542 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4543 Disable the specified breakpoints---or all breakpoints, if none are
4544 listed. A disabled breakpoint has no effect but is not forgotten. All
4545 options such as ignore-counts, conditions and commands are remembered in
4546 case the breakpoint is enabled again later. You may abbreviate
4547 @code{disable} as @code{dis}.
4550 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4551 Enable the specified breakpoints (or all defined breakpoints). They
4552 become effective once again in stopping your program.
4554 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4555 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4556 of these breakpoints immediately after stopping your program.
4558 @item enable @r{[}breakpoints@r{]} count @var{count} @var{range}@dots{}
4559 Enable the specified breakpoints temporarily. @value{GDBN} records
4560 @var{count} with each of the specified breakpoints, and decrements a
4561 breakpoint's count when it is hit. When any count reaches 0,
4562 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
4563 count (@pxref{Conditions, ,Break Conditions}), that will be
4564 decremented to 0 before @var{count} is affected.
4566 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4567 Enable the specified breakpoints to work once, then die. @value{GDBN}
4568 deletes any of these breakpoints as soon as your program stops there.
4569 Breakpoints set by the @code{tbreak} command start out in this state.
4572 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4573 @c confusing: tbreak is also initially enabled.
4574 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4575 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4576 subsequently, they become disabled or enabled only when you use one of
4577 the commands above. (The command @code{until} can set and delete a
4578 breakpoint of its own, but it does not change the state of your other
4579 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4583 @subsection Break Conditions
4584 @cindex conditional breakpoints
4585 @cindex breakpoint conditions
4587 @c FIXME what is scope of break condition expr? Context where wanted?
4588 @c in particular for a watchpoint?
4589 The simplest sort of breakpoint breaks every time your program reaches a
4590 specified place. You can also specify a @dfn{condition} for a
4591 breakpoint. A condition is just a Boolean expression in your
4592 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4593 a condition evaluates the expression each time your program reaches it,
4594 and your program stops only if the condition is @emph{true}.
4596 This is the converse of using assertions for program validation; in that
4597 situation, you want to stop when the assertion is violated---that is,
4598 when the condition is false. In C, if you want to test an assertion expressed
4599 by the condition @var{assert}, you should set the condition
4600 @samp{! @var{assert}} on the appropriate breakpoint.
4602 Conditions are also accepted for watchpoints; you may not need them,
4603 since a watchpoint is inspecting the value of an expression anyhow---but
4604 it might be simpler, say, to just set a watchpoint on a variable name,
4605 and specify a condition that tests whether the new value is an interesting
4608 Break conditions can have side effects, and may even call functions in
4609 your program. This can be useful, for example, to activate functions
4610 that log program progress, or to use your own print functions to
4611 format special data structures. The effects are completely predictable
4612 unless there is another enabled breakpoint at the same address. (In
4613 that case, @value{GDBN} might see the other breakpoint first and stop your
4614 program without checking the condition of this one.) Note that
4615 breakpoint commands are usually more convenient and flexible than break
4617 purpose of performing side effects when a breakpoint is reached
4618 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4620 Breakpoint conditions can also be evaluated on the target's side if
4621 the target supports it. Instead of evaluating the conditions locally,
4622 @value{GDBN} encodes the expression into an agent expression
4623 (@pxref{Agent Expressions}) suitable for execution on the target,
4624 independently of @value{GDBN}. Global variables become raw memory
4625 locations, locals become stack accesses, and so forth.
4627 In this case, @value{GDBN} will only be notified of a breakpoint trigger
4628 when its condition evaluates to true. This mechanism may provide faster
4629 response times depending on the performance characteristics of the target
4630 since it does not need to keep @value{GDBN} informed about
4631 every breakpoint trigger, even those with false conditions.
4633 Break conditions can be specified when a breakpoint is set, by using
4634 @samp{if} in the arguments to the @code{break} command. @xref{Set
4635 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4636 with the @code{condition} command.
4638 You can also use the @code{if} keyword with the @code{watch} command.
4639 The @code{catch} command does not recognize the @code{if} keyword;
4640 @code{condition} is the only way to impose a further condition on a
4645 @item condition @var{bnum} @var{expression}
4646 Specify @var{expression} as the break condition for breakpoint,
4647 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4648 breakpoint @var{bnum} stops your program only if the value of
4649 @var{expression} is true (nonzero, in C). When you use
4650 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4651 syntactic correctness, and to determine whether symbols in it have
4652 referents in the context of your breakpoint. If @var{expression} uses
4653 symbols not referenced in the context of the breakpoint, @value{GDBN}
4654 prints an error message:
4657 No symbol "foo" in current context.
4662 not actually evaluate @var{expression} at the time the @code{condition}
4663 command (or a command that sets a breakpoint with a condition, like
4664 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4666 @item condition @var{bnum}
4667 Remove the condition from breakpoint number @var{bnum}. It becomes
4668 an ordinary unconditional breakpoint.
4671 @cindex ignore count (of breakpoint)
4672 A special case of a breakpoint condition is to stop only when the
4673 breakpoint has been reached a certain number of times. This is so
4674 useful that there is a special way to do it, using the @dfn{ignore
4675 count} of the breakpoint. Every breakpoint has an ignore count, which
4676 is an integer. Most of the time, the ignore count is zero, and
4677 therefore has no effect. But if your program reaches a breakpoint whose
4678 ignore count is positive, then instead of stopping, it just decrements
4679 the ignore count by one and continues. As a result, if the ignore count
4680 value is @var{n}, the breakpoint does not stop the next @var{n} times
4681 your program reaches it.
4685 @item ignore @var{bnum} @var{count}
4686 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4687 The next @var{count} times the breakpoint is reached, your program's
4688 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4691 To make the breakpoint stop the next time it is reached, specify
4694 When you use @code{continue} to resume execution of your program from a
4695 breakpoint, you can specify an ignore count directly as an argument to
4696 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4697 Stepping,,Continuing and Stepping}.
4699 If a breakpoint has a positive ignore count and a condition, the
4700 condition is not checked. Once the ignore count reaches zero,
4701 @value{GDBN} resumes checking the condition.
4703 You could achieve the effect of the ignore count with a condition such
4704 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4705 is decremented each time. @xref{Convenience Vars, ,Convenience
4709 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4712 @node Break Commands
4713 @subsection Breakpoint Command Lists
4715 @cindex breakpoint commands
4716 You can give any breakpoint (or watchpoint or catchpoint) a series of
4717 commands to execute when your program stops due to that breakpoint. For
4718 example, you might want to print the values of certain expressions, or
4719 enable other breakpoints.
4723 @kindex end@r{ (breakpoint commands)}
4724 @item commands @r{[}@var{range}@dots{}@r{]}
4725 @itemx @dots{} @var{command-list} @dots{}
4727 Specify a list of commands for the given breakpoints. The commands
4728 themselves appear on the following lines. Type a line containing just
4729 @code{end} to terminate the commands.
4731 To remove all commands from a breakpoint, type @code{commands} and
4732 follow it immediately with @code{end}; that is, give no commands.
4734 With no argument, @code{commands} refers to the last breakpoint,
4735 watchpoint, or catchpoint set (not to the breakpoint most recently
4736 encountered). If the most recent breakpoints were set with a single
4737 command, then the @code{commands} will apply to all the breakpoints
4738 set by that command. This applies to breakpoints set by
4739 @code{rbreak}, and also applies when a single @code{break} command
4740 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4744 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4745 disabled within a @var{command-list}.
4747 You can use breakpoint commands to start your program up again. Simply
4748 use the @code{continue} command, or @code{step}, or any other command
4749 that resumes execution.
4751 Any other commands in the command list, after a command that resumes
4752 execution, are ignored. This is because any time you resume execution
4753 (even with a simple @code{next} or @code{step}), you may encounter
4754 another breakpoint---which could have its own command list, leading to
4755 ambiguities about which list to execute.
4758 If the first command you specify in a command list is @code{silent}, the
4759 usual message about stopping at a breakpoint is not printed. This may
4760 be desirable for breakpoints that are to print a specific message and
4761 then continue. If none of the remaining commands print anything, you
4762 see no sign that the breakpoint was reached. @code{silent} is
4763 meaningful only at the beginning of a breakpoint command list.
4765 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4766 print precisely controlled output, and are often useful in silent
4767 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4769 For example, here is how you could use breakpoint commands to print the
4770 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4776 printf "x is %d\n",x
4781 One application for breakpoint commands is to compensate for one bug so
4782 you can test for another. Put a breakpoint just after the erroneous line
4783 of code, give it a condition to detect the case in which something
4784 erroneous has been done, and give it commands to assign correct values
4785 to any variables that need them. End with the @code{continue} command
4786 so that your program does not stop, and start with the @code{silent}
4787 command so that no output is produced. Here is an example:
4798 @node Dynamic Printf
4799 @subsection Dynamic Printf
4801 @cindex dynamic printf
4803 The dynamic printf command @code{dprintf} combines a breakpoint with
4804 formatted printing of your program's data to give you the effect of
4805 inserting @code{printf} calls into your program on-the-fly, without
4806 having to recompile it.
4808 In its most basic form, the output goes to the GDB console. However,
4809 you can set the variable @code{dprintf-style} for alternate handling.
4810 For instance, you can ask to format the output by calling your
4811 program's @code{printf} function. This has the advantage that the
4812 characters go to the program's output device, so they can recorded in
4813 redirects to files and so forth.
4815 If you are doing remote debugging with a stub or agent, you can also
4816 ask to have the printf handled by the remote agent. In addition to
4817 ensuring that the output goes to the remote program's device along
4818 with any other output the program might produce, you can also ask that
4819 the dprintf remain active even after disconnecting from the remote
4820 target. Using the stub/agent is also more efficient, as it can do
4821 everything without needing to communicate with @value{GDBN}.
4825 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
4826 Whenever execution reaches @var{location}, print the values of one or
4827 more @var{expressions} under the control of the string @var{template}.
4828 To print several values, separate them with commas.
4830 @item set dprintf-style @var{style}
4831 Set the dprintf output to be handled in one of several different
4832 styles enumerated below. A change of style affects all existing
4833 dynamic printfs immediately. (If you need individual control over the
4834 print commands, simply define normal breakpoints with
4835 explicitly-supplied command lists.)
4838 @kindex dprintf-style gdb
4839 Handle the output using the @value{GDBN} @code{printf} command.
4842 @kindex dprintf-style call
4843 Handle the output by calling a function in your program (normally
4847 @kindex dprintf-style agent
4848 Have the remote debugging agent (such as @code{gdbserver}) handle
4849 the output itself. This style is only available for agents that
4850 support running commands on the target.
4852 @item set dprintf-function @var{function}
4853 Set the function to call if the dprintf style is @code{call}. By
4854 default its value is @code{printf}. You may set it to any expression.
4855 that @value{GDBN} can evaluate to a function, as per the @code{call}
4858 @item set dprintf-channel @var{channel}
4859 Set a ``channel'' for dprintf. If set to a non-empty value,
4860 @value{GDBN} will evaluate it as an expression and pass the result as
4861 a first argument to the @code{dprintf-function}, in the manner of
4862 @code{fprintf} and similar functions. Otherwise, the dprintf format
4863 string will be the first argument, in the manner of @code{printf}.
4865 As an example, if you wanted @code{dprintf} output to go to a logfile
4866 that is a standard I/O stream assigned to the variable @code{mylog},
4867 you could do the following:
4870 (gdb) set dprintf-style call
4871 (gdb) set dprintf-function fprintf
4872 (gdb) set dprintf-channel mylog
4873 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
4874 Dprintf 1 at 0x123456: file main.c, line 25.
4876 1 dprintf keep y 0x00123456 in main at main.c:25
4877 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
4882 Note that the @code{info break} displays the dynamic printf commands
4883 as normal breakpoint commands; you can thus easily see the effect of
4884 the variable settings.
4886 @item set disconnected-dprintf on
4887 @itemx set disconnected-dprintf off
4888 @kindex set disconnected-dprintf
4889 Choose whether @code{dprintf} commands should continue to run if
4890 @value{GDBN} has disconnected from the target. This only applies
4891 if the @code{dprintf-style} is @code{agent}.
4893 @item show disconnected-dprintf off
4894 @kindex show disconnected-dprintf
4895 Show the current choice for disconnected @code{dprintf}.
4899 @value{GDBN} does not check the validity of function and channel,
4900 relying on you to supply values that are meaningful for the contexts
4901 in which they are being used. For instance, the function and channel
4902 may be the values of local variables, but if that is the case, then
4903 all enabled dynamic prints must be at locations within the scope of
4904 those locals. If evaluation fails, @value{GDBN} will report an error.
4906 @node Save Breakpoints
4907 @subsection How to save breakpoints to a file
4909 To save breakpoint definitions to a file use the @w{@code{save
4910 breakpoints}} command.
4913 @kindex save breakpoints
4914 @cindex save breakpoints to a file for future sessions
4915 @item save breakpoints [@var{filename}]
4916 This command saves all current breakpoint definitions together with
4917 their commands and ignore counts, into a file @file{@var{filename}}
4918 suitable for use in a later debugging session. This includes all
4919 types of breakpoints (breakpoints, watchpoints, catchpoints,
4920 tracepoints). To read the saved breakpoint definitions, use the
4921 @code{source} command (@pxref{Command Files}). Note that watchpoints
4922 with expressions involving local variables may fail to be recreated
4923 because it may not be possible to access the context where the
4924 watchpoint is valid anymore. Because the saved breakpoint definitions
4925 are simply a sequence of @value{GDBN} commands that recreate the
4926 breakpoints, you can edit the file in your favorite editing program,
4927 and remove the breakpoint definitions you're not interested in, or
4928 that can no longer be recreated.
4931 @node Static Probe Points
4932 @subsection Static Probe Points
4934 @cindex static probe point, SystemTap
4935 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
4936 for Statically Defined Tracing, and the probes are designed to have a tiny
4937 runtime code and data footprint, and no dynamic relocations. They are
4938 usable from assembly, C and C@t{++} languages. See
4939 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
4940 for a good reference on how the @acronym{SDT} probes are implemented.
4942 Currently, @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
4943 @acronym{SDT} probes are supported on ELF-compatible systems. See
4944 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
4945 for more information on how to add @code{SystemTap} @acronym{SDT} probes
4946 in your applications.
4948 @cindex semaphores on static probe points
4949 Some probes have an associated semaphore variable; for instance, this
4950 happens automatically if you defined your probe using a DTrace-style
4951 @file{.d} file. If your probe has a semaphore, @value{GDBN} will
4952 automatically enable it when you specify a breakpoint using the
4953 @samp{-probe-stap} notation. But, if you put a breakpoint at a probe's
4954 location by some other method (e.g., @code{break file:line}), then
4955 @value{GDBN} will not automatically set the semaphore.
4957 You can examine the available static static probes using @code{info
4958 probes}, with optional arguments:
4962 @item info probes stap @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
4963 If given, @var{provider} is a regular expression used to match against provider
4964 names when selecting which probes to list. If omitted, probes by all
4965 probes from all providers are listed.
4967 If given, @var{name} is a regular expression to match against probe names
4968 when selecting which probes to list. If omitted, probe names are not
4969 considered when deciding whether to display them.
4971 If given, @var{objfile} is a regular expression used to select which
4972 object files (executable or shared libraries) to examine. If not
4973 given, all object files are considered.
4975 @item info probes all
4976 List the available static probes, from all types.
4979 @vindex $_probe_arg@r{, convenience variable}
4980 A probe may specify up to twelve arguments. These are available at the
4981 point at which the probe is defined---that is, when the current PC is
4982 at the probe's location. The arguments are available using the
4983 convenience variables (@pxref{Convenience Vars})
4984 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. Each probe argument is
4985 an integer of the appropriate size; types are not preserved. The
4986 convenience variable @code{$_probe_argc} holds the number of arguments
4987 at the current probe point.
4989 These variables are always available, but attempts to access them at
4990 any location other than a probe point will cause @value{GDBN} to give
4994 @c @ifclear BARETARGET
4995 @node Error in Breakpoints
4996 @subsection ``Cannot insert breakpoints''
4998 If you request too many active hardware-assisted breakpoints and
4999 watchpoints, you will see this error message:
5001 @c FIXME: the precise wording of this message may change; the relevant
5002 @c source change is not committed yet (Sep 3, 1999).
5004 Stopped; cannot insert breakpoints.
5005 You may have requested too many hardware breakpoints and watchpoints.
5009 This message is printed when you attempt to resume the program, since
5010 only then @value{GDBN} knows exactly how many hardware breakpoints and
5011 watchpoints it needs to insert.
5013 When this message is printed, you need to disable or remove some of the
5014 hardware-assisted breakpoints and watchpoints, and then continue.
5016 @node Breakpoint-related Warnings
5017 @subsection ``Breakpoint address adjusted...''
5018 @cindex breakpoint address adjusted
5020 Some processor architectures place constraints on the addresses at
5021 which breakpoints may be placed. For architectures thus constrained,
5022 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5023 with the constraints dictated by the architecture.
5025 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5026 a VLIW architecture in which a number of RISC-like instructions may be
5027 bundled together for parallel execution. The FR-V architecture
5028 constrains the location of a breakpoint instruction within such a
5029 bundle to the instruction with the lowest address. @value{GDBN}
5030 honors this constraint by adjusting a breakpoint's address to the
5031 first in the bundle.
5033 It is not uncommon for optimized code to have bundles which contain
5034 instructions from different source statements, thus it may happen that
5035 a breakpoint's address will be adjusted from one source statement to
5036 another. Since this adjustment may significantly alter @value{GDBN}'s
5037 breakpoint related behavior from what the user expects, a warning is
5038 printed when the breakpoint is first set and also when the breakpoint
5041 A warning like the one below is printed when setting a breakpoint
5042 that's been subject to address adjustment:
5045 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5048 Such warnings are printed both for user settable and @value{GDBN}'s
5049 internal breakpoints. If you see one of these warnings, you should
5050 verify that a breakpoint set at the adjusted address will have the
5051 desired affect. If not, the breakpoint in question may be removed and
5052 other breakpoints may be set which will have the desired behavior.
5053 E.g., it may be sufficient to place the breakpoint at a later
5054 instruction. A conditional breakpoint may also be useful in some
5055 cases to prevent the breakpoint from triggering too often.
5057 @value{GDBN} will also issue a warning when stopping at one of these
5058 adjusted breakpoints:
5061 warning: Breakpoint 1 address previously adjusted from 0x00010414
5065 When this warning is encountered, it may be too late to take remedial
5066 action except in cases where the breakpoint is hit earlier or more
5067 frequently than expected.
5069 @node Continuing and Stepping
5070 @section Continuing and Stepping
5074 @cindex resuming execution
5075 @dfn{Continuing} means resuming program execution until your program
5076 completes normally. In contrast, @dfn{stepping} means executing just
5077 one more ``step'' of your program, where ``step'' may mean either one
5078 line of source code, or one machine instruction (depending on what
5079 particular command you use). Either when continuing or when stepping,
5080 your program may stop even sooner, due to a breakpoint or a signal. (If
5081 it stops due to a signal, you may want to use @code{handle}, or use
5082 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5083 or you may step into the signal's handler (@pxref{stepping and signal
5088 @kindex c @r{(@code{continue})}
5089 @kindex fg @r{(resume foreground execution)}
5090 @item continue @r{[}@var{ignore-count}@r{]}
5091 @itemx c @r{[}@var{ignore-count}@r{]}
5092 @itemx fg @r{[}@var{ignore-count}@r{]}
5093 Resume program execution, at the address where your program last stopped;
5094 any breakpoints set at that address are bypassed. The optional argument
5095 @var{ignore-count} allows you to specify a further number of times to
5096 ignore a breakpoint at this location; its effect is like that of
5097 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5099 The argument @var{ignore-count} is meaningful only when your program
5100 stopped due to a breakpoint. At other times, the argument to
5101 @code{continue} is ignored.
5103 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5104 debugged program is deemed to be the foreground program) are provided
5105 purely for convenience, and have exactly the same behavior as
5109 To resume execution at a different place, you can use @code{return}
5110 (@pxref{Returning, ,Returning from a Function}) to go back to the
5111 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5112 Different Address}) to go to an arbitrary location in your program.
5114 A typical technique for using stepping is to set a breakpoint
5115 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5116 beginning of the function or the section of your program where a problem
5117 is believed to lie, run your program until it stops at that breakpoint,
5118 and then step through the suspect area, examining the variables that are
5119 interesting, until you see the problem happen.
5123 @kindex s @r{(@code{step})}
5125 Continue running your program until control reaches a different source
5126 line, then stop it and return control to @value{GDBN}. This command is
5127 abbreviated @code{s}.
5130 @c "without debugging information" is imprecise; actually "without line
5131 @c numbers in the debugging information". (gcc -g1 has debugging info but
5132 @c not line numbers). But it seems complex to try to make that
5133 @c distinction here.
5134 @emph{Warning:} If you use the @code{step} command while control is
5135 within a function that was compiled without debugging information,
5136 execution proceeds until control reaches a function that does have
5137 debugging information. Likewise, it will not step into a function which
5138 is compiled without debugging information. To step through functions
5139 without debugging information, use the @code{stepi} command, described
5143 The @code{step} command only stops at the first instruction of a source
5144 line. This prevents the multiple stops that could otherwise occur in
5145 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5146 to stop if a function that has debugging information is called within
5147 the line. In other words, @code{step} @emph{steps inside} any functions
5148 called within the line.
5150 Also, the @code{step} command only enters a function if there is line
5151 number information for the function. Otherwise it acts like the
5152 @code{next} command. This avoids problems when using @code{cc -gl}
5153 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5154 was any debugging information about the routine.
5156 @item step @var{count}
5157 Continue running as in @code{step}, but do so @var{count} times. If a
5158 breakpoint is reached, or a signal not related to stepping occurs before
5159 @var{count} steps, stepping stops right away.
5162 @kindex n @r{(@code{next})}
5163 @item next @r{[}@var{count}@r{]}
5164 Continue to the next source line in the current (innermost) stack frame.
5165 This is similar to @code{step}, but function calls that appear within
5166 the line of code are executed without stopping. Execution stops when
5167 control reaches a different line of code at the original stack level
5168 that was executing when you gave the @code{next} command. This command
5169 is abbreviated @code{n}.
5171 An argument @var{count} is a repeat count, as for @code{step}.
5174 @c FIX ME!! Do we delete this, or is there a way it fits in with
5175 @c the following paragraph? --- Vctoria
5177 @c @code{next} within a function that lacks debugging information acts like
5178 @c @code{step}, but any function calls appearing within the code of the
5179 @c function are executed without stopping.
5181 The @code{next} command only stops at the first instruction of a
5182 source line. This prevents multiple stops that could otherwise occur in
5183 @code{switch} statements, @code{for} loops, etc.
5185 @kindex set step-mode
5187 @cindex functions without line info, and stepping
5188 @cindex stepping into functions with no line info
5189 @itemx set step-mode on
5190 The @code{set step-mode on} command causes the @code{step} command to
5191 stop at the first instruction of a function which contains no debug line
5192 information rather than stepping over it.
5194 This is useful in cases where you may be interested in inspecting the
5195 machine instructions of a function which has no symbolic info and do not
5196 want @value{GDBN} to automatically skip over this function.
5198 @item set step-mode off
5199 Causes the @code{step} command to step over any functions which contains no
5200 debug information. This is the default.
5202 @item show step-mode
5203 Show whether @value{GDBN} will stop in or step over functions without
5204 source line debug information.
5207 @kindex fin @r{(@code{finish})}
5209 Continue running until just after function in the selected stack frame
5210 returns. Print the returned value (if any). This command can be
5211 abbreviated as @code{fin}.
5213 Contrast this with the @code{return} command (@pxref{Returning,
5214 ,Returning from a Function}).
5217 @kindex u @r{(@code{until})}
5218 @cindex run until specified location
5221 Continue running until a source line past the current line, in the
5222 current stack frame, is reached. This command is used to avoid single
5223 stepping through a loop more than once. It is like the @code{next}
5224 command, except that when @code{until} encounters a jump, it
5225 automatically continues execution until the program counter is greater
5226 than the address of the jump.
5228 This means that when you reach the end of a loop after single stepping
5229 though it, @code{until} makes your program continue execution until it
5230 exits the loop. In contrast, a @code{next} command at the end of a loop
5231 simply steps back to the beginning of the loop, which forces you to step
5232 through the next iteration.
5234 @code{until} always stops your program if it attempts to exit the current
5237 @code{until} may produce somewhat counterintuitive results if the order
5238 of machine code does not match the order of the source lines. For
5239 example, in the following excerpt from a debugging session, the @code{f}
5240 (@code{frame}) command shows that execution is stopped at line
5241 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5245 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5247 (@value{GDBP}) until
5248 195 for ( ; argc > 0; NEXTARG) @{
5251 This happened because, for execution efficiency, the compiler had
5252 generated code for the loop closure test at the end, rather than the
5253 start, of the loop---even though the test in a C @code{for}-loop is
5254 written before the body of the loop. The @code{until} command appeared
5255 to step back to the beginning of the loop when it advanced to this
5256 expression; however, it has not really gone to an earlier
5257 statement---not in terms of the actual machine code.
5259 @code{until} with no argument works by means of single
5260 instruction stepping, and hence is slower than @code{until} with an
5263 @item until @var{location}
5264 @itemx u @var{location}
5265 Continue running your program until either the specified @var{location} is
5266 reached, or the current stack frame returns. The location is any of
5267 the forms described in @ref{Specify Location}.
5268 This form of the command uses temporary breakpoints, and
5269 hence is quicker than @code{until} without an argument. The specified
5270 location is actually reached only if it is in the current frame. This
5271 implies that @code{until} can be used to skip over recursive function
5272 invocations. For instance in the code below, if the current location is
5273 line @code{96}, issuing @code{until 99} will execute the program up to
5274 line @code{99} in the same invocation of factorial, i.e., after the inner
5275 invocations have returned.
5278 94 int factorial (int value)
5280 96 if (value > 1) @{
5281 97 value *= factorial (value - 1);
5288 @kindex advance @var{location}
5289 @item advance @var{location}
5290 Continue running the program up to the given @var{location}. An argument is
5291 required, which should be of one of the forms described in
5292 @ref{Specify Location}.
5293 Execution will also stop upon exit from the current stack
5294 frame. This command is similar to @code{until}, but @code{advance} will
5295 not skip over recursive function calls, and the target location doesn't
5296 have to be in the same frame as the current one.
5300 @kindex si @r{(@code{stepi})}
5302 @itemx stepi @var{arg}
5304 Execute one machine instruction, then stop and return to the debugger.
5306 It is often useful to do @samp{display/i $pc} when stepping by machine
5307 instructions. This makes @value{GDBN} automatically display the next
5308 instruction to be executed, each time your program stops. @xref{Auto
5309 Display,, Automatic Display}.
5311 An argument is a repeat count, as in @code{step}.
5315 @kindex ni @r{(@code{nexti})}
5317 @itemx nexti @var{arg}
5319 Execute one machine instruction, but if it is a function call,
5320 proceed until the function returns.
5322 An argument is a repeat count, as in @code{next}.
5326 @anchor{range stepping}
5327 @cindex range stepping
5328 @cindex target-assisted range stepping
5329 By default, and if available, @value{GDBN} makes use of
5330 target-assisted @dfn{range stepping}. In other words, whenever you
5331 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
5332 tells the target to step the corresponding range of instruction
5333 addresses instead of issuing multiple single-steps. This speeds up
5334 line stepping, particularly for remote targets. Ideally, there should
5335 be no reason you would want to turn range stepping off. However, it's
5336 possible that a bug in the debug info, a bug in the remote stub (for
5337 remote targets), or even a bug in @value{GDBN} could make line
5338 stepping behave incorrectly when target-assisted range stepping is
5339 enabled. You can use the following command to turn off range stepping
5343 @kindex set range-stepping
5344 @kindex show range-stepping
5345 @item set range-stepping
5346 @itemx show range-stepping
5347 Control whether range stepping is enabled.
5349 If @code{on}, and the target supports it, @value{GDBN} tells the
5350 target to step a range of addresses itself, instead of issuing
5351 multiple single-steps. If @code{off}, @value{GDBN} always issues
5352 single-steps, even if range stepping is supported by the target. The
5353 default is @code{on}.
5357 @node Skipping Over Functions and Files
5358 @section Skipping Over Functions and Files
5359 @cindex skipping over functions and files
5361 The program you are debugging may contain some functions which are
5362 uninteresting to debug. The @code{skip} comand lets you tell @value{GDBN} to
5363 skip a function or all functions in a file when stepping.
5365 For example, consider the following C function:
5376 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5377 are not interested in stepping through @code{boring}. If you run @code{step}
5378 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5379 step over both @code{foo} and @code{boring}!
5381 One solution is to @code{step} into @code{boring} and use the @code{finish}
5382 command to immediately exit it. But this can become tedious if @code{boring}
5383 is called from many places.
5385 A more flexible solution is to execute @kbd{skip boring}. This instructs
5386 @value{GDBN} never to step into @code{boring}. Now when you execute
5387 @code{step} at line 103, you'll step over @code{boring} and directly into
5390 You can also instruct @value{GDBN} to skip all functions in a file, with, for
5391 example, @code{skip file boring.c}.
5394 @kindex skip function
5395 @item skip @r{[}@var{linespec}@r{]}
5396 @itemx skip function @r{[}@var{linespec}@r{]}
5397 After running this command, the function named by @var{linespec} or the
5398 function containing the line named by @var{linespec} will be skipped over when
5399 stepping. @xref{Specify Location}.
5401 If you do not specify @var{linespec}, the function you're currently debugging
5404 (If you have a function called @code{file} that you want to skip, use
5405 @kbd{skip function file}.)
5408 @item skip file @r{[}@var{filename}@r{]}
5409 After running this command, any function whose source lives in @var{filename}
5410 will be skipped over when stepping.
5412 If you do not specify @var{filename}, functions whose source lives in the file
5413 you're currently debugging will be skipped.
5416 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
5417 These are the commands for managing your list of skips:
5421 @item info skip @r{[}@var{range}@r{]}
5422 Print details about the specified skip(s). If @var{range} is not specified,
5423 print a table with details about all functions and files marked for skipping.
5424 @code{info skip} prints the following information about each skip:
5428 A number identifying this skip.
5430 The type of this skip, either @samp{function} or @samp{file}.
5431 @item Enabled or Disabled
5432 Enabled skips are marked with @samp{y}. Disabled skips are marked with @samp{n}.
5434 For function skips, this column indicates the address in memory of the function
5435 being skipped. If you've set a function skip on a function which has not yet
5436 been loaded, this field will contain @samp{<PENDING>}. Once a shared library
5437 which has the function is loaded, @code{info skip} will show the function's
5440 For file skips, this field contains the filename being skipped. For functions
5441 skips, this field contains the function name and its line number in the file
5442 where it is defined.
5446 @item skip delete @r{[}@var{range}@r{]}
5447 Delete the specified skip(s). If @var{range} is not specified, delete all
5451 @item skip enable @r{[}@var{range}@r{]}
5452 Enable the specified skip(s). If @var{range} is not specified, enable all
5455 @kindex skip disable
5456 @item skip disable @r{[}@var{range}@r{]}
5457 Disable the specified skip(s). If @var{range} is not specified, disable all
5466 A signal is an asynchronous event that can happen in a program. The
5467 operating system defines the possible kinds of signals, and gives each
5468 kind a name and a number. For example, in Unix @code{SIGINT} is the
5469 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
5470 @code{SIGSEGV} is the signal a program gets from referencing a place in
5471 memory far away from all the areas in use; @code{SIGALRM} occurs when
5472 the alarm clock timer goes off (which happens only if your program has
5473 requested an alarm).
5475 @cindex fatal signals
5476 Some signals, including @code{SIGALRM}, are a normal part of the
5477 functioning of your program. Others, such as @code{SIGSEGV}, indicate
5478 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
5479 program has not specified in advance some other way to handle the signal.
5480 @code{SIGINT} does not indicate an error in your program, but it is normally
5481 fatal so it can carry out the purpose of the interrupt: to kill the program.
5483 @value{GDBN} has the ability to detect any occurrence of a signal in your
5484 program. You can tell @value{GDBN} in advance what to do for each kind of
5487 @cindex handling signals
5488 Normally, @value{GDBN} is set up to let the non-erroneous signals like
5489 @code{SIGALRM} be silently passed to your program
5490 (so as not to interfere with their role in the program's functioning)
5491 but to stop your program immediately whenever an error signal happens.
5492 You can change these settings with the @code{handle} command.
5495 @kindex info signals
5499 Print a table of all the kinds of signals and how @value{GDBN} has been told to
5500 handle each one. You can use this to see the signal numbers of all
5501 the defined types of signals.
5503 @item info signals @var{sig}
5504 Similar, but print information only about the specified signal number.
5506 @code{info handle} is an alias for @code{info signals}.
5508 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5509 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
5510 for details about this command.
5513 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
5514 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
5515 can be the number of a signal or its name (with or without the
5516 @samp{SIG} at the beginning); a list of signal numbers of the form
5517 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
5518 known signals. Optional arguments @var{keywords}, described below,
5519 say what change to make.
5523 The keywords allowed by the @code{handle} command can be abbreviated.
5524 Their full names are:
5528 @value{GDBN} should not stop your program when this signal happens. It may
5529 still print a message telling you that the signal has come in.
5532 @value{GDBN} should stop your program when this signal happens. This implies
5533 the @code{print} keyword as well.
5536 @value{GDBN} should print a message when this signal happens.
5539 @value{GDBN} should not mention the occurrence of the signal at all. This
5540 implies the @code{nostop} keyword as well.
5544 @value{GDBN} should allow your program to see this signal; your program
5545 can handle the signal, or else it may terminate if the signal is fatal
5546 and not handled. @code{pass} and @code{noignore} are synonyms.
5550 @value{GDBN} should not allow your program to see this signal.
5551 @code{nopass} and @code{ignore} are synonyms.
5555 When a signal stops your program, the signal is not visible to the
5557 continue. Your program sees the signal then, if @code{pass} is in
5558 effect for the signal in question @emph{at that time}. In other words,
5559 after @value{GDBN} reports a signal, you can use the @code{handle}
5560 command with @code{pass} or @code{nopass} to control whether your
5561 program sees that signal when you continue.
5563 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
5564 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
5565 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
5568 You can also use the @code{signal} command to prevent your program from
5569 seeing a signal, or cause it to see a signal it normally would not see,
5570 or to give it any signal at any time. For example, if your program stopped
5571 due to some sort of memory reference error, you might store correct
5572 values into the erroneous variables and continue, hoping to see more
5573 execution; but your program would probably terminate immediately as
5574 a result of the fatal signal once it saw the signal. To prevent this,
5575 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
5578 @cindex stepping and signal handlers
5579 @anchor{stepping and signal handlers}
5581 @value{GDBN} optimizes for stepping the mainline code. If a signal
5582 that has @code{handle nostop} and @code{handle pass} set arrives while
5583 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
5584 in progress, @value{GDBN} lets the signal handler run and then resumes
5585 stepping the mainline code once the signal handler returns. In other
5586 words, @value{GDBN} steps over the signal handler. This prevents
5587 signals that you've specified as not interesting (with @code{handle
5588 nostop}) from changing the focus of debugging unexpectedly. Note that
5589 the signal handler itself may still hit a breakpoint, stop for another
5590 signal that has @code{handle stop} in effect, or for any other event
5591 that normally results in stopping the stepping command sooner. Also
5592 note that @value{GDBN} still informs you that the program received a
5593 signal if @code{handle print} is set.
5595 @anchor{stepping into signal handlers}
5597 If you set @code{handle pass} for a signal, and your program sets up a
5598 handler for it, then issuing a stepping command, such as @code{step}
5599 or @code{stepi}, when your program is stopped due to the signal will
5600 step @emph{into} the signal handler (if the target supports that).
5602 Likewise, if you use the @code{queue-signal} command to queue a signal
5603 to be delivered to the current thread when execution of the thread
5604 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
5605 stepping command will step into the signal handler.
5607 Here's an example, using @code{stepi} to step to the first instruction
5608 of @code{SIGUSR1}'s handler:
5611 (@value{GDBP}) handle SIGUSR1
5612 Signal Stop Print Pass to program Description
5613 SIGUSR1 Yes Yes Yes User defined signal 1
5617 Program received signal SIGUSR1, User defined signal 1.
5618 main () sigusr1.c:28
5621 sigusr1_handler () at sigusr1.c:9
5625 The same, but using @code{queue-signal} instead of waiting for the
5626 program to receive the signal first:
5631 (@value{GDBP}) queue-signal SIGUSR1
5633 sigusr1_handler () at sigusr1.c:9
5638 @cindex extra signal information
5639 @anchor{extra signal information}
5641 On some targets, @value{GDBN} can inspect extra signal information
5642 associated with the intercepted signal, before it is actually
5643 delivered to the program being debugged. This information is exported
5644 by the convenience variable @code{$_siginfo}, and consists of data
5645 that is passed by the kernel to the signal handler at the time of the
5646 receipt of a signal. The data type of the information itself is
5647 target dependent. You can see the data type using the @code{ptype
5648 $_siginfo} command. On Unix systems, it typically corresponds to the
5649 standard @code{siginfo_t} type, as defined in the @file{signal.h}
5652 Here's an example, on a @sc{gnu}/Linux system, printing the stray
5653 referenced address that raised a segmentation fault.
5657 (@value{GDBP}) continue
5658 Program received signal SIGSEGV, Segmentation fault.
5659 0x0000000000400766 in main ()
5661 (@value{GDBP}) ptype $_siginfo
5668 struct @{...@} _kill;
5669 struct @{...@} _timer;
5671 struct @{...@} _sigchld;
5672 struct @{...@} _sigfault;
5673 struct @{...@} _sigpoll;
5676 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
5680 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
5681 $1 = (void *) 0x7ffff7ff7000
5685 Depending on target support, @code{$_siginfo} may also be writable.
5688 @section Stopping and Starting Multi-thread Programs
5690 @cindex stopped threads
5691 @cindex threads, stopped
5693 @cindex continuing threads
5694 @cindex threads, continuing
5696 @value{GDBN} supports debugging programs with multiple threads
5697 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
5698 are two modes of controlling execution of your program within the
5699 debugger. In the default mode, referred to as @dfn{all-stop mode},
5700 when any thread in your program stops (for example, at a breakpoint
5701 or while being stepped), all other threads in the program are also stopped by
5702 @value{GDBN}. On some targets, @value{GDBN} also supports
5703 @dfn{non-stop mode}, in which other threads can continue to run freely while
5704 you examine the stopped thread in the debugger.
5707 * All-Stop Mode:: All threads stop when GDB takes control
5708 * Non-Stop Mode:: Other threads continue to execute
5709 * Background Execution:: Running your program asynchronously
5710 * Thread-Specific Breakpoints:: Controlling breakpoints
5711 * Interrupted System Calls:: GDB may interfere with system calls
5712 * Observer Mode:: GDB does not alter program behavior
5716 @subsection All-Stop Mode
5718 @cindex all-stop mode
5720 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
5721 @emph{all} threads of execution stop, not just the current thread. This
5722 allows you to examine the overall state of the program, including
5723 switching between threads, without worrying that things may change
5726 Conversely, whenever you restart the program, @emph{all} threads start
5727 executing. @emph{This is true even when single-stepping} with commands
5728 like @code{step} or @code{next}.
5730 In particular, @value{GDBN} cannot single-step all threads in lockstep.
5731 Since thread scheduling is up to your debugging target's operating
5732 system (not controlled by @value{GDBN}), other threads may
5733 execute more than one statement while the current thread completes a
5734 single step. Moreover, in general other threads stop in the middle of a
5735 statement, rather than at a clean statement boundary, when the program
5738 You might even find your program stopped in another thread after
5739 continuing or even single-stepping. This happens whenever some other
5740 thread runs into a breakpoint, a signal, or an exception before the
5741 first thread completes whatever you requested.
5743 @cindex automatic thread selection
5744 @cindex switching threads automatically
5745 @cindex threads, automatic switching
5746 Whenever @value{GDBN} stops your program, due to a breakpoint or a
5747 signal, it automatically selects the thread where that breakpoint or
5748 signal happened. @value{GDBN} alerts you to the context switch with a
5749 message such as @samp{[Switching to Thread @var{n}]} to identify the
5752 On some OSes, you can modify @value{GDBN}'s default behavior by
5753 locking the OS scheduler to allow only a single thread to run.
5756 @item set scheduler-locking @var{mode}
5757 @cindex scheduler locking mode
5758 @cindex lock scheduler
5759 Set the scheduler locking mode. If it is @code{off}, then there is no
5760 locking and any thread may run at any time. If @code{on}, then only the
5761 current thread may run when the inferior is resumed. The @code{step}
5762 mode optimizes for single-stepping; it prevents other threads
5763 from preempting the current thread while you are stepping, so that
5764 the focus of debugging does not change unexpectedly.
5765 Other threads only rarely (or never) get a chance to run
5766 when you step. They are more likely to run when you @samp{next} over a
5767 function call, and they are completely free to run when you use commands
5768 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
5769 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
5770 the current thread away from the thread that you are debugging.
5772 @item show scheduler-locking
5773 Display the current scheduler locking mode.
5776 @cindex resume threads of multiple processes simultaneously
5777 By default, when you issue one of the execution commands such as
5778 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
5779 threads of the current inferior to run. For example, if @value{GDBN}
5780 is attached to two inferiors, each with two threads, the
5781 @code{continue} command resumes only the two threads of the current
5782 inferior. This is useful, for example, when you debug a program that
5783 forks and you want to hold the parent stopped (so that, for instance,
5784 it doesn't run to exit), while you debug the child. In other
5785 situations, you may not be interested in inspecting the current state
5786 of any of the processes @value{GDBN} is attached to, and you may want
5787 to resume them all until some breakpoint is hit. In the latter case,
5788 you can instruct @value{GDBN} to allow all threads of all the
5789 inferiors to run with the @w{@code{set schedule-multiple}} command.
5792 @kindex set schedule-multiple
5793 @item set schedule-multiple
5794 Set the mode for allowing threads of multiple processes to be resumed
5795 when an execution command is issued. When @code{on}, all threads of
5796 all processes are allowed to run. When @code{off}, only the threads
5797 of the current process are resumed. The default is @code{off}. The
5798 @code{scheduler-locking} mode takes precedence when set to @code{on},
5799 or while you are stepping and set to @code{step}.
5801 @item show schedule-multiple
5802 Display the current mode for resuming the execution of threads of
5807 @subsection Non-Stop Mode
5809 @cindex non-stop mode
5811 @c This section is really only a place-holder, and needs to be expanded
5812 @c with more details.
5814 For some multi-threaded targets, @value{GDBN} supports an optional
5815 mode of operation in which you can examine stopped program threads in
5816 the debugger while other threads continue to execute freely. This
5817 minimizes intrusion when debugging live systems, such as programs
5818 where some threads have real-time constraints or must continue to
5819 respond to external events. This is referred to as @dfn{non-stop} mode.
5821 In non-stop mode, when a thread stops to report a debugging event,
5822 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5823 threads as well, in contrast to the all-stop mode behavior. Additionally,
5824 execution commands such as @code{continue} and @code{step} apply by default
5825 only to the current thread in non-stop mode, rather than all threads as
5826 in all-stop mode. This allows you to control threads explicitly in
5827 ways that are not possible in all-stop mode --- for example, stepping
5828 one thread while allowing others to run freely, stepping
5829 one thread while holding all others stopped, or stepping several threads
5830 independently and simultaneously.
5832 To enter non-stop mode, use this sequence of commands before you run
5833 or attach to your program:
5836 # If using the CLI, pagination breaks non-stop.
5839 # Finally, turn it on!
5843 You can use these commands to manipulate the non-stop mode setting:
5846 @kindex set non-stop
5847 @item set non-stop on
5848 Enable selection of non-stop mode.
5849 @item set non-stop off
5850 Disable selection of non-stop mode.
5851 @kindex show non-stop
5853 Show the current non-stop enablement setting.
5856 Note these commands only reflect whether non-stop mode is enabled,
5857 not whether the currently-executing program is being run in non-stop mode.
5858 In particular, the @code{set non-stop} preference is only consulted when
5859 @value{GDBN} starts or connects to the target program, and it is generally
5860 not possible to switch modes once debugging has started. Furthermore,
5861 since not all targets support non-stop mode, even when you have enabled
5862 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
5865 In non-stop mode, all execution commands apply only to the current thread
5866 by default. That is, @code{continue} only continues one thread.
5867 To continue all threads, issue @code{continue -a} or @code{c -a}.
5869 You can use @value{GDBN}'s background execution commands
5870 (@pxref{Background Execution}) to run some threads in the background
5871 while you continue to examine or step others from @value{GDBN}.
5872 The MI execution commands (@pxref{GDB/MI Program Execution}) are
5873 always executed asynchronously in non-stop mode.
5875 Suspending execution is done with the @code{interrupt} command when
5876 running in the background, or @kbd{Ctrl-c} during foreground execution.
5877 In all-stop mode, this stops the whole process;
5878 but in non-stop mode the interrupt applies only to the current thread.
5879 To stop the whole program, use @code{interrupt -a}.
5881 Other execution commands do not currently support the @code{-a} option.
5883 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
5884 that thread current, as it does in all-stop mode. This is because the
5885 thread stop notifications are asynchronous with respect to @value{GDBN}'s
5886 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
5887 changed to a different thread just as you entered a command to operate on the
5888 previously current thread.
5890 @node Background Execution
5891 @subsection Background Execution
5893 @cindex foreground execution
5894 @cindex background execution
5895 @cindex asynchronous execution
5896 @cindex execution, foreground, background and asynchronous
5898 @value{GDBN}'s execution commands have two variants: the normal
5899 foreground (synchronous) behavior, and a background
5900 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
5901 the program to report that some thread has stopped before prompting for
5902 another command. In background execution, @value{GDBN} immediately gives
5903 a command prompt so that you can issue other commands while your program runs.
5905 If the target doesn't support async mode, @value{GDBN} issues an error
5906 message if you attempt to use the background execution commands.
5908 To specify background execution, add a @code{&} to the command. For example,
5909 the background form of the @code{continue} command is @code{continue&}, or
5910 just @code{c&}. The execution commands that accept background execution
5916 @xref{Starting, , Starting your Program}.
5920 @xref{Attach, , Debugging an Already-running Process}.
5924 @xref{Continuing and Stepping, step}.
5928 @xref{Continuing and Stepping, stepi}.
5932 @xref{Continuing and Stepping, next}.
5936 @xref{Continuing and Stepping, nexti}.
5940 @xref{Continuing and Stepping, continue}.
5944 @xref{Continuing and Stepping, finish}.
5948 @xref{Continuing and Stepping, until}.
5952 Background execution is especially useful in conjunction with non-stop
5953 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
5954 However, you can also use these commands in the normal all-stop mode with
5955 the restriction that you cannot issue another execution command until the
5956 previous one finishes. Examples of commands that are valid in all-stop
5957 mode while the program is running include @code{help} and @code{info break}.
5959 You can interrupt your program while it is running in the background by
5960 using the @code{interrupt} command.
5967 Suspend execution of the running program. In all-stop mode,
5968 @code{interrupt} stops the whole process, but in non-stop mode, it stops
5969 only the current thread. To stop the whole program in non-stop mode,
5970 use @code{interrupt -a}.
5973 @node Thread-Specific Breakpoints
5974 @subsection Thread-Specific Breakpoints
5976 When your program has multiple threads (@pxref{Threads,, Debugging
5977 Programs with Multiple Threads}), you can choose whether to set
5978 breakpoints on all threads, or on a particular thread.
5981 @cindex breakpoints and threads
5982 @cindex thread breakpoints
5983 @kindex break @dots{} thread @var{threadno}
5984 @item break @var{linespec} thread @var{threadno}
5985 @itemx break @var{linespec} thread @var{threadno} if @dots{}
5986 @var{linespec} specifies source lines; there are several ways of
5987 writing them (@pxref{Specify Location}), but the effect is always to
5988 specify some source line.
5990 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
5991 to specify that you only want @value{GDBN} to stop the program when a
5992 particular thread reaches this breakpoint. The @var{threadno} specifier
5993 is one of the numeric thread identifiers assigned by @value{GDBN}, shown
5994 in the first column of the @samp{info threads} display.
5996 If you do not specify @samp{thread @var{threadno}} when you set a
5997 breakpoint, the breakpoint applies to @emph{all} threads of your
6000 You can use the @code{thread} qualifier on conditional breakpoints as
6001 well; in this case, place @samp{thread @var{threadno}} before or
6002 after the breakpoint condition, like this:
6005 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6010 Thread-specific breakpoints are automatically deleted when
6011 @value{GDBN} detects the corresponding thread is no longer in the
6012 thread list. For example:
6016 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6019 There are several ways for a thread to disappear, such as a regular
6020 thread exit, but also when you detach from the process with the
6021 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6022 Process}), or if @value{GDBN} loses the remote connection
6023 (@pxref{Remote Debugging}), etc. Note that with some targets,
6024 @value{GDBN} is only able to detect a thread has exited when the user
6025 explictly asks for the thread list with the @code{info threads}
6028 @node Interrupted System Calls
6029 @subsection Interrupted System Calls
6031 @cindex thread breakpoints and system calls
6032 @cindex system calls and thread breakpoints
6033 @cindex premature return from system calls
6034 There is an unfortunate side effect when using @value{GDBN} to debug
6035 multi-threaded programs. If one thread stops for a
6036 breakpoint, or for some other reason, and another thread is blocked in a
6037 system call, then the system call may return prematurely. This is a
6038 consequence of the interaction between multiple threads and the signals
6039 that @value{GDBN} uses to implement breakpoints and other events that
6042 To handle this problem, your program should check the return value of
6043 each system call and react appropriately. This is good programming
6046 For example, do not write code like this:
6052 The call to @code{sleep} will return early if a different thread stops
6053 at a breakpoint or for some other reason.
6055 Instead, write this:
6060 unslept = sleep (unslept);
6063 A system call is allowed to return early, so the system is still
6064 conforming to its specification. But @value{GDBN} does cause your
6065 multi-threaded program to behave differently than it would without
6068 Also, @value{GDBN} uses internal breakpoints in the thread library to
6069 monitor certain events such as thread creation and thread destruction.
6070 When such an event happens, a system call in another thread may return
6071 prematurely, even though your program does not appear to stop.
6074 @subsection Observer Mode
6076 If you want to build on non-stop mode and observe program behavior
6077 without any chance of disruption by @value{GDBN}, you can set
6078 variables to disable all of the debugger's attempts to modify state,
6079 whether by writing memory, inserting breakpoints, etc. These operate
6080 at a low level, intercepting operations from all commands.
6082 When all of these are set to @code{off}, then @value{GDBN} is said to
6083 be @dfn{observer mode}. As a convenience, the variable
6084 @code{observer} can be set to disable these, plus enable non-stop
6087 Note that @value{GDBN} will not prevent you from making nonsensical
6088 combinations of these settings. For instance, if you have enabled
6089 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6090 then breakpoints that work by writing trap instructions into the code
6091 stream will still not be able to be placed.
6096 @item set observer on
6097 @itemx set observer off
6098 When set to @code{on}, this disables all the permission variables
6099 below (except for @code{insert-fast-tracepoints}), plus enables
6100 non-stop debugging. Setting this to @code{off} switches back to
6101 normal debugging, though remaining in non-stop mode.
6104 Show whether observer mode is on or off.
6106 @kindex may-write-registers
6107 @item set may-write-registers on
6108 @itemx set may-write-registers off
6109 This controls whether @value{GDBN} will attempt to alter the values of
6110 registers, such as with assignment expressions in @code{print}, or the
6111 @code{jump} command. It defaults to @code{on}.
6113 @item show may-write-registers
6114 Show the current permission to write registers.
6116 @kindex may-write-memory
6117 @item set may-write-memory on
6118 @itemx set may-write-memory off
6119 This controls whether @value{GDBN} will attempt to alter the contents
6120 of memory, such as with assignment expressions in @code{print}. It
6121 defaults to @code{on}.
6123 @item show may-write-memory
6124 Show the current permission to write memory.
6126 @kindex may-insert-breakpoints
6127 @item set may-insert-breakpoints on
6128 @itemx set may-insert-breakpoints off
6129 This controls whether @value{GDBN} will attempt to insert breakpoints.
6130 This affects all breakpoints, including internal breakpoints defined
6131 by @value{GDBN}. It defaults to @code{on}.
6133 @item show may-insert-breakpoints
6134 Show the current permission to insert breakpoints.
6136 @kindex may-insert-tracepoints
6137 @item set may-insert-tracepoints on
6138 @itemx set may-insert-tracepoints off
6139 This controls whether @value{GDBN} will attempt to insert (regular)
6140 tracepoints at the beginning of a tracing experiment. It affects only
6141 non-fast tracepoints, fast tracepoints being under the control of
6142 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6144 @item show may-insert-tracepoints
6145 Show the current permission to insert tracepoints.
6147 @kindex may-insert-fast-tracepoints
6148 @item set may-insert-fast-tracepoints on
6149 @itemx set may-insert-fast-tracepoints off
6150 This controls whether @value{GDBN} will attempt to insert fast
6151 tracepoints at the beginning of a tracing experiment. It affects only
6152 fast tracepoints, regular (non-fast) tracepoints being under the
6153 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6155 @item show may-insert-fast-tracepoints
6156 Show the current permission to insert fast tracepoints.
6158 @kindex may-interrupt
6159 @item set may-interrupt on
6160 @itemx set may-interrupt off
6161 This controls whether @value{GDBN} will attempt to interrupt or stop
6162 program execution. When this variable is @code{off}, the
6163 @code{interrupt} command will have no effect, nor will
6164 @kbd{Ctrl-c}. It defaults to @code{on}.
6166 @item show may-interrupt
6167 Show the current permission to interrupt or stop the program.
6171 @node Reverse Execution
6172 @chapter Running programs backward
6173 @cindex reverse execution
6174 @cindex running programs backward
6176 When you are debugging a program, it is not unusual to realize that
6177 you have gone too far, and some event of interest has already happened.
6178 If the target environment supports it, @value{GDBN} can allow you to
6179 ``rewind'' the program by running it backward.
6181 A target environment that supports reverse execution should be able
6182 to ``undo'' the changes in machine state that have taken place as the
6183 program was executing normally. Variables, registers etc.@: should
6184 revert to their previous values. Obviously this requires a great
6185 deal of sophistication on the part of the target environment; not
6186 all target environments can support reverse execution.
6188 When a program is executed in reverse, the instructions that
6189 have most recently been executed are ``un-executed'', in reverse
6190 order. The program counter runs backward, following the previous
6191 thread of execution in reverse. As each instruction is ``un-executed'',
6192 the values of memory and/or registers that were changed by that
6193 instruction are reverted to their previous states. After executing
6194 a piece of source code in reverse, all side effects of that code
6195 should be ``undone'', and all variables should be returned to their
6196 prior values@footnote{
6197 Note that some side effects are easier to undo than others. For instance,
6198 memory and registers are relatively easy, but device I/O is hard. Some
6199 targets may be able undo things like device I/O, and some may not.
6201 The contract between @value{GDBN} and the reverse executing target
6202 requires only that the target do something reasonable when
6203 @value{GDBN} tells it to execute backwards, and then report the
6204 results back to @value{GDBN}. Whatever the target reports back to
6205 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6206 assumes that the memory and registers that the target reports are in a
6207 consistant state, but @value{GDBN} accepts whatever it is given.
6210 If you are debugging in a target environment that supports
6211 reverse execution, @value{GDBN} provides the following commands.
6214 @kindex reverse-continue
6215 @kindex rc @r{(@code{reverse-continue})}
6216 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6217 @itemx rc @r{[}@var{ignore-count}@r{]}
6218 Beginning at the point where your program last stopped, start executing
6219 in reverse. Reverse execution will stop for breakpoints and synchronous
6220 exceptions (signals), just like normal execution. Behavior of
6221 asynchronous signals depends on the target environment.
6223 @kindex reverse-step
6224 @kindex rs @r{(@code{step})}
6225 @item reverse-step @r{[}@var{count}@r{]}
6226 Run the program backward until control reaches the start of a
6227 different source line; then stop it, and return control to @value{GDBN}.
6229 Like the @code{step} command, @code{reverse-step} will only stop
6230 at the beginning of a source line. It ``un-executes'' the previously
6231 executed source line. If the previous source line included calls to
6232 debuggable functions, @code{reverse-step} will step (backward) into
6233 the called function, stopping at the beginning of the @emph{last}
6234 statement in the called function (typically a return statement).
6236 Also, as with the @code{step} command, if non-debuggable functions are
6237 called, @code{reverse-step} will run thru them backward without stopping.
6239 @kindex reverse-stepi
6240 @kindex rsi @r{(@code{reverse-stepi})}
6241 @item reverse-stepi @r{[}@var{count}@r{]}
6242 Reverse-execute one machine instruction. Note that the instruction
6243 to be reverse-executed is @emph{not} the one pointed to by the program
6244 counter, but the instruction executed prior to that one. For instance,
6245 if the last instruction was a jump, @code{reverse-stepi} will take you
6246 back from the destination of the jump to the jump instruction itself.
6248 @kindex reverse-next
6249 @kindex rn @r{(@code{reverse-next})}
6250 @item reverse-next @r{[}@var{count}@r{]}
6251 Run backward to the beginning of the previous line executed in
6252 the current (innermost) stack frame. If the line contains function
6253 calls, they will be ``un-executed'' without stopping. Starting from
6254 the first line of a function, @code{reverse-next} will take you back
6255 to the caller of that function, @emph{before} the function was called,
6256 just as the normal @code{next} command would take you from the last
6257 line of a function back to its return to its caller
6258 @footnote{Unless the code is too heavily optimized.}.
6260 @kindex reverse-nexti
6261 @kindex rni @r{(@code{reverse-nexti})}
6262 @item reverse-nexti @r{[}@var{count}@r{]}
6263 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
6264 in reverse, except that called functions are ``un-executed'' atomically.
6265 That is, if the previously executed instruction was a return from
6266 another function, @code{reverse-nexti} will continue to execute
6267 in reverse until the call to that function (from the current stack
6270 @kindex reverse-finish
6271 @item reverse-finish
6272 Just as the @code{finish} command takes you to the point where the
6273 current function returns, @code{reverse-finish} takes you to the point
6274 where it was called. Instead of ending up at the end of the current
6275 function invocation, you end up at the beginning.
6277 @kindex set exec-direction
6278 @item set exec-direction
6279 Set the direction of target execution.
6280 @item set exec-direction reverse
6281 @cindex execute forward or backward in time
6282 @value{GDBN} will perform all execution commands in reverse, until the
6283 exec-direction mode is changed to ``forward''. Affected commands include
6284 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
6285 command cannot be used in reverse mode.
6286 @item set exec-direction forward
6287 @value{GDBN} will perform all execution commands in the normal fashion.
6288 This is the default.
6292 @node Process Record and Replay
6293 @chapter Recording Inferior's Execution and Replaying It
6294 @cindex process record and replay
6295 @cindex recording inferior's execution and replaying it
6297 On some platforms, @value{GDBN} provides a special @dfn{process record
6298 and replay} target that can record a log of the process execution, and
6299 replay it later with both forward and reverse execution commands.
6302 When this target is in use, if the execution log includes the record
6303 for the next instruction, @value{GDBN} will debug in @dfn{replay
6304 mode}. In the replay mode, the inferior does not really execute code
6305 instructions. Instead, all the events that normally happen during
6306 code execution are taken from the execution log. While code is not
6307 really executed in replay mode, the values of registers (including the
6308 program counter register) and the memory of the inferior are still
6309 changed as they normally would. Their contents are taken from the
6313 If the record for the next instruction is not in the execution log,
6314 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6315 inferior executes normally, and @value{GDBN} records the execution log
6318 The process record and replay target supports reverse execution
6319 (@pxref{Reverse Execution}), even if the platform on which the
6320 inferior runs does not. However, the reverse execution is limited in
6321 this case by the range of the instructions recorded in the execution
6322 log. In other words, reverse execution on platforms that don't
6323 support it directly can only be done in the replay mode.
6325 When debugging in the reverse direction, @value{GDBN} will work in
6326 replay mode as long as the execution log includes the record for the
6327 previous instruction; otherwise, it will work in record mode, if the
6328 platform supports reverse execution, or stop if not.
6330 For architecture environments that support process record and replay,
6331 @value{GDBN} provides the following commands:
6334 @kindex target record
6335 @kindex target record-full
6336 @kindex target record-btrace
6339 @kindex record btrace
6343 @item record @var{method}
6344 This command starts the process record and replay target. The
6345 recording method can be specified as parameter. Without a parameter
6346 the command uses the @code{full} recording method. The following
6347 recording methods are available:
6351 Full record/replay recording using @value{GDBN}'s software record and
6352 replay implementation. This method allows replaying and reverse
6356 Hardware-supported instruction recording. This method does not record
6357 data. Further, the data is collected in a ring buffer so old data will
6358 be overwritten when the buffer is full. It allows limited replay and
6361 This recording method may not be available on all processors.
6364 The process record and replay target can only debug a process that is
6365 already running. Therefore, you need first to start the process with
6366 the @kbd{run} or @kbd{start} commands, and then start the recording
6367 with the @kbd{record @var{method}} command.
6369 Both @code{record @var{method}} and @code{rec @var{method}} are
6370 aliases of @code{target record-@var{method}}.
6372 @cindex displaced stepping, and process record and replay
6373 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
6374 will be automatically disabled when process record and replay target
6375 is started. That's because the process record and replay target
6376 doesn't support displaced stepping.
6378 @cindex non-stop mode, and process record and replay
6379 @cindex asynchronous execution, and process record and replay
6380 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
6381 the asynchronous execution mode (@pxref{Background Execution}), not
6382 all recording methods are available. The @code{full} recording method
6383 does not support these two modes.
6388 Stop the process record and replay target. When process record and
6389 replay target stops, the entire execution log will be deleted and the
6390 inferior will either be terminated, or will remain in its final state.
6392 When you stop the process record and replay target in record mode (at
6393 the end of the execution log), the inferior will be stopped at the
6394 next instruction that would have been recorded. In other words, if
6395 you record for a while and then stop recording, the inferior process
6396 will be left in the same state as if the recording never happened.
6398 On the other hand, if the process record and replay target is stopped
6399 while in replay mode (that is, not at the end of the execution log,
6400 but at some earlier point), the inferior process will become ``live''
6401 at that earlier state, and it will then be possible to continue the
6402 usual ``live'' debugging of the process from that state.
6404 When the inferior process exits, or @value{GDBN} detaches from it,
6405 process record and replay target will automatically stop itself.
6409 Go to a specific location in the execution log. There are several
6410 ways to specify the location to go to:
6413 @item record goto begin
6414 @itemx record goto start
6415 Go to the beginning of the execution log.
6417 @item record goto end
6418 Go to the end of the execution log.
6420 @item record goto @var{n}
6421 Go to instruction number @var{n} in the execution log.
6425 @item record save @var{filename}
6426 Save the execution log to a file @file{@var{filename}}.
6427 Default filename is @file{gdb_record.@var{process_id}}, where
6428 @var{process_id} is the process ID of the inferior.
6430 This command may not be available for all recording methods.
6432 @kindex record restore
6433 @item record restore @var{filename}
6434 Restore the execution log from a file @file{@var{filename}}.
6435 File must have been created with @code{record save}.
6437 @kindex set record full
6438 @item set record full insn-number-max @var{limit}
6439 @itemx set record full insn-number-max unlimited
6440 Set the limit of instructions to be recorded for the @code{full}
6441 recording method. Default value is 200000.
6443 If @var{limit} is a positive number, then @value{GDBN} will start
6444 deleting instructions from the log once the number of the record
6445 instructions becomes greater than @var{limit}. For every new recorded
6446 instruction, @value{GDBN} will delete the earliest recorded
6447 instruction to keep the number of recorded instructions at the limit.
6448 (Since deleting recorded instructions loses information, @value{GDBN}
6449 lets you control what happens when the limit is reached, by means of
6450 the @code{stop-at-limit} option, described below.)
6452 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
6453 delete recorded instructions from the execution log. The number of
6454 recorded instructions is limited only by the available memory.
6456 @kindex show record full
6457 @item show record full insn-number-max
6458 Show the limit of instructions to be recorded with the @code{full}
6461 @item set record full stop-at-limit
6462 Control the behavior of the @code{full} recording method when the
6463 number of recorded instructions reaches the limit. If ON (the
6464 default), @value{GDBN} will stop when the limit is reached for the
6465 first time and ask you whether you want to stop the inferior or
6466 continue running it and recording the execution log. If you decide
6467 to continue recording, each new recorded instruction will cause the
6468 oldest one to be deleted.
6470 If this option is OFF, @value{GDBN} will automatically delete the
6471 oldest record to make room for each new one, without asking.
6473 @item show record full stop-at-limit
6474 Show the current setting of @code{stop-at-limit}.
6476 @item set record full memory-query
6477 Control the behavior when @value{GDBN} is unable to record memory
6478 changes caused by an instruction for the @code{full} recording method.
6479 If ON, @value{GDBN} will query whether to stop the inferior in that
6482 If this option is OFF (the default), @value{GDBN} will automatically
6483 ignore the effect of such instructions on memory. Later, when
6484 @value{GDBN} replays this execution log, it will mark the log of this
6485 instruction as not accessible, and it will not affect the replay
6488 @item show record full memory-query
6489 Show the current setting of @code{memory-query}.
6491 @kindex set record btrace
6492 The @code{btrace} record target does not trace data. As a
6493 convenience, when replaying, @value{GDBN} reads read-only memory off
6494 the live program directly, assuming that the addresses of the
6495 read-only areas don't change. This for example makes it possible to
6496 disassemble code while replaying, but not to print variables.
6497 In some cases, being able to inspect variables might be useful.
6498 You can use the following command for that:
6500 @item set record btrace replay-memory-access
6501 Control the behavior of the @code{btrace} recording method when
6502 accessing memory during replay. If @code{read-only} (the default),
6503 @value{GDBN} will only allow accesses to read-only memory.
6504 If @code{read-write}, @value{GDBN} will allow accesses to read-only
6505 and to read-write memory. Beware that the accessed memory corresponds
6506 to the live target and not necessarily to the current replay
6509 @kindex show record btrace
6510 @item show record btrace replay-memory-access
6511 Show the current setting of @code{replay-memory-access}.
6515 Show various statistics about the recording depending on the recording
6520 For the @code{full} recording method, it shows the state of process
6521 record and its in-memory execution log buffer, including:
6525 Whether in record mode or replay mode.
6527 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
6529 Highest recorded instruction number.
6531 Current instruction about to be replayed (if in replay mode).
6533 Number of instructions contained in the execution log.
6535 Maximum number of instructions that may be contained in the execution log.
6539 For the @code{btrace} recording method, it shows the number of
6540 instructions that have been recorded and the number of blocks of
6541 sequential control-flow that is formed by the recorded instructions.
6544 @kindex record delete
6547 When record target runs in replay mode (``in the past''), delete the
6548 subsequent execution log and begin to record a new execution log starting
6549 from the current address. This means you will abandon the previously
6550 recorded ``future'' and begin recording a new ``future''.
6552 @kindex record instruction-history
6553 @kindex rec instruction-history
6554 @item record instruction-history
6555 Disassembles instructions from the recorded execution log. By
6556 default, ten instructions are disassembled. This can be changed using
6557 the @code{set record instruction-history-size} command. Instructions
6558 are printed in execution order. There are several ways to specify
6559 what part of the execution log to disassemble:
6562 @item record instruction-history @var{insn}
6563 Disassembles ten instructions starting from instruction number
6566 @item record instruction-history @var{insn}, +/-@var{n}
6567 Disassembles @var{n} instructions around instruction number
6568 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
6569 @var{n} instructions after instruction number @var{insn}. If
6570 @var{n} is preceded with @code{-}, disassembles @var{n}
6571 instructions before instruction number @var{insn}.
6573 @item record instruction-history
6574 Disassembles ten more instructions after the last disassembly.
6576 @item record instruction-history -
6577 Disassembles ten more instructions before the last disassembly.
6579 @item record instruction-history @var{begin} @var{end}
6580 Disassembles instructions beginning with instruction number
6581 @var{begin} until instruction number @var{end}. The instruction
6582 number @var{end} is included.
6585 This command may not be available for all recording methods.
6588 @item set record instruction-history-size @var{size}
6589 @itemx set record instruction-history-size unlimited
6590 Define how many instructions to disassemble in the @code{record
6591 instruction-history} command. The default value is 10.
6592 A @var{size} of @code{unlimited} means unlimited instructions.
6595 @item show record instruction-history-size
6596 Show how many instructions to disassemble in the @code{record
6597 instruction-history} command.
6599 @kindex record function-call-history
6600 @kindex rec function-call-history
6601 @item record function-call-history
6602 Prints the execution history at function granularity. It prints one
6603 line for each sequence of instructions that belong to the same
6604 function giving the name of that function, the source lines
6605 for this instruction sequence (if the @code{/l} modifier is
6606 specified), and the instructions numbers that form the sequence (if
6607 the @code{/i} modifier is specified). The function names are indented
6608 to reflect the call stack depth if the @code{/c} modifier is
6609 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
6613 (@value{GDBP}) @b{list 1, 10}
6624 (@value{GDBP}) @b{record function-call-history /ilc}
6625 1 bar inst 1,4 at foo.c:6,8
6626 2 foo inst 5,10 at foo.c:2,3
6627 3 bar inst 11,13 at foo.c:9,10
6630 By default, ten lines are printed. This can be changed using the
6631 @code{set record function-call-history-size} command. Functions are
6632 printed in execution order. There are several ways to specify what
6636 @item record function-call-history @var{func}
6637 Prints ten functions starting from function number @var{func}.
6639 @item record function-call-history @var{func}, +/-@var{n}
6640 Prints @var{n} functions around function number @var{func}. If
6641 @var{n} is preceded with @code{+}, prints @var{n} functions after
6642 function number @var{func}. If @var{n} is preceded with @code{-},
6643 prints @var{n} functions before function number @var{func}.
6645 @item record function-call-history
6646 Prints ten more functions after the last ten-line print.
6648 @item record function-call-history -
6649 Prints ten more functions before the last ten-line print.
6651 @item record function-call-history @var{begin} @var{end}
6652 Prints functions beginning with function number @var{begin} until
6653 function number @var{end}. The function number @var{end} is included.
6656 This command may not be available for all recording methods.
6658 @item set record function-call-history-size @var{size}
6659 @itemx set record function-call-history-size unlimited
6660 Define how many lines to print in the
6661 @code{record function-call-history} command. The default value is 10.
6662 A size of @code{unlimited} means unlimited lines.
6664 @item show record function-call-history-size
6665 Show how many lines to print in the
6666 @code{record function-call-history} command.
6671 @chapter Examining the Stack
6673 When your program has stopped, the first thing you need to know is where it
6674 stopped and how it got there.
6677 Each time your program performs a function call, information about the call
6679 That information includes the location of the call in your program,
6680 the arguments of the call,
6681 and the local variables of the function being called.
6682 The information is saved in a block of data called a @dfn{stack frame}.
6683 The stack frames are allocated in a region of memory called the @dfn{call
6686 When your program stops, the @value{GDBN} commands for examining the
6687 stack allow you to see all of this information.
6689 @cindex selected frame
6690 One of the stack frames is @dfn{selected} by @value{GDBN} and many
6691 @value{GDBN} commands refer implicitly to the selected frame. In
6692 particular, whenever you ask @value{GDBN} for the value of a variable in
6693 your program, the value is found in the selected frame. There are
6694 special @value{GDBN} commands to select whichever frame you are
6695 interested in. @xref{Selection, ,Selecting a Frame}.
6697 When your program stops, @value{GDBN} automatically selects the
6698 currently executing frame and describes it briefly, similar to the
6699 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
6702 * Frames:: Stack frames
6703 * Backtrace:: Backtraces
6704 * Frame Filter Management:: Managing frame filters
6705 * Selection:: Selecting a frame
6706 * Frame Info:: Information on a frame
6711 @section Stack Frames
6713 @cindex frame, definition
6715 The call stack is divided up into contiguous pieces called @dfn{stack
6716 frames}, or @dfn{frames} for short; each frame is the data associated
6717 with one call to one function. The frame contains the arguments given
6718 to the function, the function's local variables, and the address at
6719 which the function is executing.
6721 @cindex initial frame
6722 @cindex outermost frame
6723 @cindex innermost frame
6724 When your program is started, the stack has only one frame, that of the
6725 function @code{main}. This is called the @dfn{initial} frame or the
6726 @dfn{outermost} frame. Each time a function is called, a new frame is
6727 made. Each time a function returns, the frame for that function invocation
6728 is eliminated. If a function is recursive, there can be many frames for
6729 the same function. The frame for the function in which execution is
6730 actually occurring is called the @dfn{innermost} frame. This is the most
6731 recently created of all the stack frames that still exist.
6733 @cindex frame pointer
6734 Inside your program, stack frames are identified by their addresses. A
6735 stack frame consists of many bytes, each of which has its own address; each
6736 kind of computer has a convention for choosing one byte whose
6737 address serves as the address of the frame. Usually this address is kept
6738 in a register called the @dfn{frame pointer register}
6739 (@pxref{Registers, $fp}) while execution is going on in that frame.
6741 @cindex frame number
6742 @value{GDBN} assigns numbers to all existing stack frames, starting with
6743 zero for the innermost frame, one for the frame that called it,
6744 and so on upward. These numbers do not really exist in your program;
6745 they are assigned by @value{GDBN} to give you a way of designating stack
6746 frames in @value{GDBN} commands.
6748 @c The -fomit-frame-pointer below perennially causes hbox overflow
6749 @c underflow problems.
6750 @cindex frameless execution
6751 Some compilers provide a way to compile functions so that they operate
6752 without stack frames. (For example, the @value{NGCC} option
6754 @samp{-fomit-frame-pointer}
6756 generates functions without a frame.)
6757 This is occasionally done with heavily used library functions to save
6758 the frame setup time. @value{GDBN} has limited facilities for dealing
6759 with these function invocations. If the innermost function invocation
6760 has no stack frame, @value{GDBN} nevertheless regards it as though
6761 it had a separate frame, which is numbered zero as usual, allowing
6762 correct tracing of the function call chain. However, @value{GDBN} has
6763 no provision for frameless functions elsewhere in the stack.
6766 @kindex frame@r{, command}
6767 @cindex current stack frame
6768 @item frame @r{[}@var{framespec}@r{]}
6769 The @code{frame} command allows you to move from one stack frame to another,
6770 and to print the stack frame you select. The @var{framespec} may be either the
6771 address of the frame or the stack frame number. Without an argument,
6772 @code{frame} prints the current stack frame.
6774 @kindex select-frame
6775 @cindex selecting frame silently
6777 The @code{select-frame} command allows you to move from one stack frame
6778 to another without printing the frame. This is the silent version of
6786 @cindex call stack traces
6787 A backtrace is a summary of how your program got where it is. It shows one
6788 line per frame, for many frames, starting with the currently executing
6789 frame (frame zero), followed by its caller (frame one), and on up the
6792 @anchor{backtrace-command}
6795 @kindex bt @r{(@code{backtrace})}
6798 Print a backtrace of the entire stack: one line per frame for all
6799 frames in the stack.
6801 You can stop the backtrace at any time by typing the system interrupt
6802 character, normally @kbd{Ctrl-c}.
6804 @item backtrace @var{n}
6806 Similar, but print only the innermost @var{n} frames.
6808 @item backtrace -@var{n}
6810 Similar, but print only the outermost @var{n} frames.
6812 @item backtrace full
6814 @itemx bt full @var{n}
6815 @itemx bt full -@var{n}
6816 Print the values of the local variables also. As described above,
6817 @var{n} specifies the number of frames to print.
6819 @item backtrace no-filters
6820 @itemx bt no-filters
6821 @itemx bt no-filters @var{n}
6822 @itemx bt no-filters -@var{n}
6823 @itemx bt no-filters full
6824 @itemx bt no-filters full @var{n}
6825 @itemx bt no-filters full -@var{n}
6826 Do not run Python frame filters on this backtrace. @xref{Frame
6827 Filter API}, for more information. Additionally use @ref{disable
6828 frame-filter all} to turn off all frame filters. This is only
6829 relevant when @value{GDBN} has been configured with @code{Python}
6835 The names @code{where} and @code{info stack} (abbreviated @code{info s})
6836 are additional aliases for @code{backtrace}.
6838 @cindex multiple threads, backtrace
6839 In a multi-threaded program, @value{GDBN} by default shows the
6840 backtrace only for the current thread. To display the backtrace for
6841 several or all of the threads, use the command @code{thread apply}
6842 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
6843 apply all backtrace}, @value{GDBN} will display the backtrace for all
6844 the threads; this is handy when you debug a core dump of a
6845 multi-threaded program.
6847 Each line in the backtrace shows the frame number and the function name.
6848 The program counter value is also shown---unless you use @code{set
6849 print address off}. The backtrace also shows the source file name and
6850 line number, as well as the arguments to the function. The program
6851 counter value is omitted if it is at the beginning of the code for that
6854 Here is an example of a backtrace. It was made with the command
6855 @samp{bt 3}, so it shows the innermost three frames.
6859 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6861 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
6862 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
6864 (More stack frames follow...)
6869 The display for frame zero does not begin with a program counter
6870 value, indicating that your program has stopped at the beginning of the
6871 code for line @code{993} of @code{builtin.c}.
6874 The value of parameter @code{data} in frame 1 has been replaced by
6875 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
6876 only if it is a scalar (integer, pointer, enumeration, etc). See command
6877 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
6878 on how to configure the way function parameter values are printed.
6880 @cindex optimized out, in backtrace
6881 @cindex function call arguments, optimized out
6882 If your program was compiled with optimizations, some compilers will
6883 optimize away arguments passed to functions if those arguments are
6884 never used after the call. Such optimizations generate code that
6885 passes arguments through registers, but doesn't store those arguments
6886 in the stack frame. @value{GDBN} has no way of displaying such
6887 arguments in stack frames other than the innermost one. Here's what
6888 such a backtrace might look like:
6892 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6894 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
6895 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
6897 (More stack frames follow...)
6902 The values of arguments that were not saved in their stack frames are
6903 shown as @samp{<optimized out>}.
6905 If you need to display the values of such optimized-out arguments,
6906 either deduce that from other variables whose values depend on the one
6907 you are interested in, or recompile without optimizations.
6909 @cindex backtrace beyond @code{main} function
6910 @cindex program entry point
6911 @cindex startup code, and backtrace
6912 Most programs have a standard user entry point---a place where system
6913 libraries and startup code transition into user code. For C this is
6914 @code{main}@footnote{
6915 Note that embedded programs (the so-called ``free-standing''
6916 environment) are not required to have a @code{main} function as the
6917 entry point. They could even have multiple entry points.}.
6918 When @value{GDBN} finds the entry function in a backtrace
6919 it will terminate the backtrace, to avoid tracing into highly
6920 system-specific (and generally uninteresting) code.
6922 If you need to examine the startup code, or limit the number of levels
6923 in a backtrace, you can change this behavior:
6926 @item set backtrace past-main
6927 @itemx set backtrace past-main on
6928 @kindex set backtrace
6929 Backtraces will continue past the user entry point.
6931 @item set backtrace past-main off
6932 Backtraces will stop when they encounter the user entry point. This is the
6935 @item show backtrace past-main
6936 @kindex show backtrace
6937 Display the current user entry point backtrace policy.
6939 @item set backtrace past-entry
6940 @itemx set backtrace past-entry on
6941 Backtraces will continue past the internal entry point of an application.
6942 This entry point is encoded by the linker when the application is built,
6943 and is likely before the user entry point @code{main} (or equivalent) is called.
6945 @item set backtrace past-entry off
6946 Backtraces will stop when they encounter the internal entry point of an
6947 application. This is the default.
6949 @item show backtrace past-entry
6950 Display the current internal entry point backtrace policy.
6952 @item set backtrace limit @var{n}
6953 @itemx set backtrace limit 0
6954 @itemx set backtrace limit unlimited
6955 @cindex backtrace limit
6956 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
6957 or zero means unlimited levels.
6959 @item show backtrace limit
6960 Display the current limit on backtrace levels.
6963 You can control how file names are displayed.
6966 @item set filename-display
6967 @itemx set filename-display relative
6968 @cindex filename-display
6969 Display file names relative to the compilation directory. This is the default.
6971 @item set filename-display basename
6972 Display only basename of a filename.
6974 @item set filename-display absolute
6975 Display an absolute filename.
6977 @item show filename-display
6978 Show the current way to display filenames.
6981 @node Frame Filter Management
6982 @section Management of Frame Filters.
6983 @cindex managing frame filters
6985 Frame filters are Python based utilities to manage and decorate the
6986 output of frames. @xref{Frame Filter API}, for further information.
6988 Managing frame filters is performed by several commands available
6989 within @value{GDBN}, detailed here.
6992 @kindex info frame-filter
6993 @item info frame-filter
6994 Print a list of installed frame filters from all dictionaries, showing
6995 their name, priority and enabled status.
6997 @kindex disable frame-filter
6998 @anchor{disable frame-filter all}
6999 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
7000 Disable a frame filter in the dictionary matching
7001 @var{filter-dictionary} and @var{filter-name}. The
7002 @var{filter-dictionary} may be @code{all}, @code{global},
7003 @code{progspace}, or the name of the object file where the frame filter
7004 dictionary resides. When @code{all} is specified, all frame filters
7005 across all dictionaries are disabled. The @var{filter-name} is the name
7006 of the frame filter and is used when @code{all} is not the option for
7007 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
7008 may be enabled again later.
7010 @kindex enable frame-filter
7011 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
7012 Enable a frame filter in the dictionary matching
7013 @var{filter-dictionary} and @var{filter-name}. The
7014 @var{filter-dictionary} may be @code{all}, @code{global},
7015 @code{progspace} or the name of the object file where the frame filter
7016 dictionary resides. When @code{all} is specified, all frame filters across
7017 all dictionaries are enabled. The @var{filter-name} is the name of the frame
7018 filter and is used when @code{all} is not the option for
7019 @var{filter-dictionary}.
7024 (gdb) info frame-filter
7026 global frame-filters:
7027 Priority Enabled Name
7028 1000 No PrimaryFunctionFilter
7031 progspace /build/test frame-filters:
7032 Priority Enabled Name
7033 100 Yes ProgspaceFilter
7035 objfile /build/test frame-filters:
7036 Priority Enabled Name
7037 999 Yes BuildProgra Filter
7039 (gdb) disable frame-filter /build/test BuildProgramFilter
7040 (gdb) info frame-filter
7042 global frame-filters:
7043 Priority Enabled Name
7044 1000 No PrimaryFunctionFilter
7047 progspace /build/test frame-filters:
7048 Priority Enabled Name
7049 100 Yes ProgspaceFilter
7051 objfile /build/test frame-filters:
7052 Priority Enabled Name
7053 999 No BuildProgramFilter
7055 (gdb) enable frame-filter global PrimaryFunctionFilter
7056 (gdb) info frame-filter
7058 global frame-filters:
7059 Priority Enabled Name
7060 1000 Yes PrimaryFunctionFilter
7063 progspace /build/test frame-filters:
7064 Priority Enabled Name
7065 100 Yes ProgspaceFilter
7067 objfile /build/test frame-filters:
7068 Priority Enabled Name
7069 999 No BuildProgramFilter
7072 @kindex set frame-filter priority
7073 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
7074 Set the @var{priority} of a frame filter in the dictionary matching
7075 @var{filter-dictionary}, and the frame filter name matching
7076 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
7077 @code{progspace} or the name of the object file where the frame filter
7078 dictionary resides. The @var{priority} is an integer.
7080 @kindex show frame-filter priority
7081 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
7082 Show the @var{priority} of a frame filter in the dictionary matching
7083 @var{filter-dictionary}, and the frame filter name matching
7084 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
7085 @code{progspace} or the name of the object file where the frame filter
7091 (gdb) info frame-filter
7093 global frame-filters:
7094 Priority Enabled Name
7095 1000 Yes PrimaryFunctionFilter
7098 progspace /build/test frame-filters:
7099 Priority Enabled Name
7100 100 Yes ProgspaceFilter
7102 objfile /build/test frame-filters:
7103 Priority Enabled Name
7104 999 No BuildProgramFilter
7106 (gdb) set frame-filter priority global Reverse 50
7107 (gdb) info frame-filter
7109 global frame-filters:
7110 Priority Enabled Name
7111 1000 Yes PrimaryFunctionFilter
7114 progspace /build/test frame-filters:
7115 Priority Enabled Name
7116 100 Yes ProgspaceFilter
7118 objfile /build/test frame-filters:
7119 Priority Enabled Name
7120 999 No BuildProgramFilter
7125 @section Selecting a Frame
7127 Most commands for examining the stack and other data in your program work on
7128 whichever stack frame is selected at the moment. Here are the commands for
7129 selecting a stack frame; all of them finish by printing a brief description
7130 of the stack frame just selected.
7133 @kindex frame@r{, selecting}
7134 @kindex f @r{(@code{frame})}
7137 Select frame number @var{n}. Recall that frame zero is the innermost
7138 (currently executing) frame, frame one is the frame that called the
7139 innermost one, and so on. The highest-numbered frame is the one for
7142 @item frame @var{addr}
7144 Select the frame at address @var{addr}. This is useful mainly if the
7145 chaining of stack frames has been damaged by a bug, making it
7146 impossible for @value{GDBN} to assign numbers properly to all frames. In
7147 addition, this can be useful when your program has multiple stacks and
7148 switches between them.
7150 On the SPARC architecture, @code{frame} needs two addresses to
7151 select an arbitrary frame: a frame pointer and a stack pointer.
7153 On the @acronym{MIPS} and Alpha architecture, it needs two addresses: a stack
7154 pointer and a program counter.
7156 On the 29k architecture, it needs three addresses: a register stack
7157 pointer, a program counter, and a memory stack pointer.
7161 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
7162 numbers @var{n}, this advances toward the outermost frame, to higher
7163 frame numbers, to frames that have existed longer.
7166 @kindex do @r{(@code{down})}
7168 Move @var{n} frames down the stack; @var{n} defaults to 1. For
7169 positive numbers @var{n}, this advances toward the innermost frame, to
7170 lower frame numbers, to frames that were created more recently.
7171 You may abbreviate @code{down} as @code{do}.
7174 All of these commands end by printing two lines of output describing the
7175 frame. The first line shows the frame number, the function name, the
7176 arguments, and the source file and line number of execution in that
7177 frame. The second line shows the text of that source line.
7185 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
7187 10 read_input_file (argv[i]);
7191 After such a printout, the @code{list} command with no arguments
7192 prints ten lines centered on the point of execution in the frame.
7193 You can also edit the program at the point of execution with your favorite
7194 editing program by typing @code{edit}.
7195 @xref{List, ,Printing Source Lines},
7199 @kindex down-silently
7201 @item up-silently @var{n}
7202 @itemx down-silently @var{n}
7203 These two commands are variants of @code{up} and @code{down},
7204 respectively; they differ in that they do their work silently, without
7205 causing display of the new frame. They are intended primarily for use
7206 in @value{GDBN} command scripts, where the output might be unnecessary and
7211 @section Information About a Frame
7213 There are several other commands to print information about the selected
7219 When used without any argument, this command does not change which
7220 frame is selected, but prints a brief description of the currently
7221 selected stack frame. It can be abbreviated @code{f}. With an
7222 argument, this command is used to select a stack frame.
7223 @xref{Selection, ,Selecting a Frame}.
7226 @kindex info f @r{(@code{info frame})}
7229 This command prints a verbose description of the selected stack frame,
7234 the address of the frame
7236 the address of the next frame down (called by this frame)
7238 the address of the next frame up (caller of this frame)
7240 the language in which the source code corresponding to this frame is written
7242 the address of the frame's arguments
7244 the address of the frame's local variables
7246 the program counter saved in it (the address of execution in the caller frame)
7248 which registers were saved in the frame
7251 @noindent The verbose description is useful when
7252 something has gone wrong that has made the stack format fail to fit
7253 the usual conventions.
7255 @item info frame @var{addr}
7256 @itemx info f @var{addr}
7257 Print a verbose description of the frame at address @var{addr}, without
7258 selecting that frame. The selected frame remains unchanged by this
7259 command. This requires the same kind of address (more than one for some
7260 architectures) that you specify in the @code{frame} command.
7261 @xref{Selection, ,Selecting a Frame}.
7265 Print the arguments of the selected frame, each on a separate line.
7269 Print the local variables of the selected frame, each on a separate
7270 line. These are all variables (declared either static or automatic)
7271 accessible at the point of execution of the selected frame.
7277 @chapter Examining Source Files
7279 @value{GDBN} can print parts of your program's source, since the debugging
7280 information recorded in the program tells @value{GDBN} what source files were
7281 used to build it. When your program stops, @value{GDBN} spontaneously prints
7282 the line where it stopped. Likewise, when you select a stack frame
7283 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
7284 execution in that frame has stopped. You can print other portions of
7285 source files by explicit command.
7287 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
7288 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
7289 @value{GDBN} under @sc{gnu} Emacs}.
7292 * List:: Printing source lines
7293 * Specify Location:: How to specify code locations
7294 * Edit:: Editing source files
7295 * Search:: Searching source files
7296 * Source Path:: Specifying source directories
7297 * Machine Code:: Source and machine code
7301 @section Printing Source Lines
7304 @kindex l @r{(@code{list})}
7305 To print lines from a source file, use the @code{list} command
7306 (abbreviated @code{l}). By default, ten lines are printed.
7307 There are several ways to specify what part of the file you want to
7308 print; see @ref{Specify Location}, for the full list.
7310 Here are the forms of the @code{list} command most commonly used:
7313 @item list @var{linenum}
7314 Print lines centered around line number @var{linenum} in the
7315 current source file.
7317 @item list @var{function}
7318 Print lines centered around the beginning of function
7322 Print more lines. If the last lines printed were printed with a
7323 @code{list} command, this prints lines following the last lines
7324 printed; however, if the last line printed was a solitary line printed
7325 as part of displaying a stack frame (@pxref{Stack, ,Examining the
7326 Stack}), this prints lines centered around that line.
7329 Print lines just before the lines last printed.
7332 @cindex @code{list}, how many lines to display
7333 By default, @value{GDBN} prints ten source lines with any of these forms of
7334 the @code{list} command. You can change this using @code{set listsize}:
7337 @kindex set listsize
7338 @item set listsize @var{count}
7339 @itemx set listsize unlimited
7340 Make the @code{list} command display @var{count} source lines (unless
7341 the @code{list} argument explicitly specifies some other number).
7342 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
7344 @kindex show listsize
7346 Display the number of lines that @code{list} prints.
7349 Repeating a @code{list} command with @key{RET} discards the argument,
7350 so it is equivalent to typing just @code{list}. This is more useful
7351 than listing the same lines again. An exception is made for an
7352 argument of @samp{-}; that argument is preserved in repetition so that
7353 each repetition moves up in the source file.
7355 In general, the @code{list} command expects you to supply zero, one or two
7356 @dfn{linespecs}. Linespecs specify source lines; there are several ways
7357 of writing them (@pxref{Specify Location}), but the effect is always
7358 to specify some source line.
7360 Here is a complete description of the possible arguments for @code{list}:
7363 @item list @var{linespec}
7364 Print lines centered around the line specified by @var{linespec}.
7366 @item list @var{first},@var{last}
7367 Print lines from @var{first} to @var{last}. Both arguments are
7368 linespecs. When a @code{list} command has two linespecs, and the
7369 source file of the second linespec is omitted, this refers to
7370 the same source file as the first linespec.
7372 @item list ,@var{last}
7373 Print lines ending with @var{last}.
7375 @item list @var{first},
7376 Print lines starting with @var{first}.
7379 Print lines just after the lines last printed.
7382 Print lines just before the lines last printed.
7385 As described in the preceding table.
7388 @node Specify Location
7389 @section Specifying a Location
7390 @cindex specifying location
7393 Several @value{GDBN} commands accept arguments that specify a location
7394 of your program's code. Since @value{GDBN} is a source-level
7395 debugger, a location usually specifies some line in the source code;
7396 for that reason, locations are also known as @dfn{linespecs}.
7398 Here are all the different ways of specifying a code location that
7399 @value{GDBN} understands:
7403 Specifies the line number @var{linenum} of the current source file.
7406 @itemx +@var{offset}
7407 Specifies the line @var{offset} lines before or after the @dfn{current
7408 line}. For the @code{list} command, the current line is the last one
7409 printed; for the breakpoint commands, this is the line at which
7410 execution stopped in the currently selected @dfn{stack frame}
7411 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
7412 used as the second of the two linespecs in a @code{list} command,
7413 this specifies the line @var{offset} lines up or down from the first
7416 @item @var{filename}:@var{linenum}
7417 Specifies the line @var{linenum} in the source file @var{filename}.
7418 If @var{filename} is a relative file name, then it will match any
7419 source file name with the same trailing components. For example, if
7420 @var{filename} is @samp{gcc/expr.c}, then it will match source file
7421 name of @file{/build/trunk/gcc/expr.c}, but not
7422 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
7424 @item @var{function}
7425 Specifies the line that begins the body of the function @var{function}.
7426 For example, in C, this is the line with the open brace.
7428 @item @var{function}:@var{label}
7429 Specifies the line where @var{label} appears in @var{function}.
7431 @item @var{filename}:@var{function}
7432 Specifies the line that begins the body of the function @var{function}
7433 in the file @var{filename}. You only need the file name with a
7434 function name to avoid ambiguity when there are identically named
7435 functions in different source files.
7438 Specifies the line at which the label named @var{label} appears.
7439 @value{GDBN} searches for the label in the function corresponding to
7440 the currently selected stack frame. If there is no current selected
7441 stack frame (for instance, if the inferior is not running), then
7442 @value{GDBN} will not search for a label.
7444 @item *@var{address}
7445 Specifies the program address @var{address}. For line-oriented
7446 commands, such as @code{list} and @code{edit}, this specifies a source
7447 line that contains @var{address}. For @code{break} and other
7448 breakpoint oriented commands, this can be used to set breakpoints in
7449 parts of your program which do not have debugging information or
7452 Here @var{address} may be any expression valid in the current working
7453 language (@pxref{Languages, working language}) that specifies a code
7454 address. In addition, as a convenience, @value{GDBN} extends the
7455 semantics of expressions used in locations to cover the situations
7456 that frequently happen during debugging. Here are the various forms
7460 @item @var{expression}
7461 Any expression valid in the current working language.
7463 @item @var{funcaddr}
7464 An address of a function or procedure derived from its name. In C,
7465 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
7466 simply the function's name @var{function} (and actually a special case
7467 of a valid expression). In Pascal and Modula-2, this is
7468 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
7469 (although the Pascal form also works).
7471 This form specifies the address of the function's first instruction,
7472 before the stack frame and arguments have been set up.
7474 @item '@var{filename}'::@var{funcaddr}
7475 Like @var{funcaddr} above, but also specifies the name of the source
7476 file explicitly. This is useful if the name of the function does not
7477 specify the function unambiguously, e.g., if there are several
7478 functions with identical names in different source files.
7481 @cindex breakpoint at static probe point
7482 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
7483 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
7484 applications to embed static probes. @xref{Static Probe Points}, for more
7485 information on finding and using static probes. This form of linespec
7486 specifies the location of such a static probe.
7488 If @var{objfile} is given, only probes coming from that shared library
7489 or executable matching @var{objfile} as a regular expression are considered.
7490 If @var{provider} is given, then only probes from that provider are considered.
7491 If several probes match the spec, @value{GDBN} will insert a breakpoint at
7492 each one of those probes.
7498 @section Editing Source Files
7499 @cindex editing source files
7502 @kindex e @r{(@code{edit})}
7503 To edit the lines in a source file, use the @code{edit} command.
7504 The editing program of your choice
7505 is invoked with the current line set to
7506 the active line in the program.
7507 Alternatively, there are several ways to specify what part of the file you
7508 want to print if you want to see other parts of the program:
7511 @item edit @var{location}
7512 Edit the source file specified by @code{location}. Editing starts at
7513 that @var{location}, e.g., at the specified source line of the
7514 specified file. @xref{Specify Location}, for all the possible forms
7515 of the @var{location} argument; here are the forms of the @code{edit}
7516 command most commonly used:
7519 @item edit @var{number}
7520 Edit the current source file with @var{number} as the active line number.
7522 @item edit @var{function}
7523 Edit the file containing @var{function} at the beginning of its definition.
7528 @subsection Choosing your Editor
7529 You can customize @value{GDBN} to use any editor you want
7531 The only restriction is that your editor (say @code{ex}), recognizes the
7532 following command-line syntax:
7534 ex +@var{number} file
7536 The optional numeric value +@var{number} specifies the number of the line in
7537 the file where to start editing.}.
7538 By default, it is @file{@value{EDITOR}}, but you can change this
7539 by setting the environment variable @code{EDITOR} before using
7540 @value{GDBN}. For example, to configure @value{GDBN} to use the
7541 @code{vi} editor, you could use these commands with the @code{sh} shell:
7547 or in the @code{csh} shell,
7549 setenv EDITOR /usr/bin/vi
7554 @section Searching Source Files
7555 @cindex searching source files
7557 There are two commands for searching through the current source file for a
7562 @kindex forward-search
7563 @kindex fo @r{(@code{forward-search})}
7564 @item forward-search @var{regexp}
7565 @itemx search @var{regexp}
7566 The command @samp{forward-search @var{regexp}} checks each line,
7567 starting with the one following the last line listed, for a match for
7568 @var{regexp}. It lists the line that is found. You can use the
7569 synonym @samp{search @var{regexp}} or abbreviate the command name as
7572 @kindex reverse-search
7573 @item reverse-search @var{regexp}
7574 The command @samp{reverse-search @var{regexp}} checks each line, starting
7575 with the one before the last line listed and going backward, for a match
7576 for @var{regexp}. It lists the line that is found. You can abbreviate
7577 this command as @code{rev}.
7581 @section Specifying Source Directories
7584 @cindex directories for source files
7585 Executable programs sometimes do not record the directories of the source
7586 files from which they were compiled, just the names. Even when they do,
7587 the directories could be moved between the compilation and your debugging
7588 session. @value{GDBN} has a list of directories to search for source files;
7589 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
7590 it tries all the directories in the list, in the order they are present
7591 in the list, until it finds a file with the desired name.
7593 For example, suppose an executable references the file
7594 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
7595 @file{/mnt/cross}. The file is first looked up literally; if this
7596 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
7597 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
7598 message is printed. @value{GDBN} does not look up the parts of the
7599 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
7600 Likewise, the subdirectories of the source path are not searched: if
7601 the source path is @file{/mnt/cross}, and the binary refers to
7602 @file{foo.c}, @value{GDBN} would not find it under
7603 @file{/mnt/cross/usr/src/foo-1.0/lib}.
7605 Plain file names, relative file names with leading directories, file
7606 names containing dots, etc.@: are all treated as described above; for
7607 instance, if the source path is @file{/mnt/cross}, and the source file
7608 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
7609 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
7610 that---@file{/mnt/cross/foo.c}.
7612 Note that the executable search path is @emph{not} used to locate the
7615 Whenever you reset or rearrange the source path, @value{GDBN} clears out
7616 any information it has cached about where source files are found and where
7617 each line is in the file.
7621 When you start @value{GDBN}, its source path includes only @samp{cdir}
7622 and @samp{cwd}, in that order.
7623 To add other directories, use the @code{directory} command.
7625 The search path is used to find both program source files and @value{GDBN}
7626 script files (read using the @samp{-command} option and @samp{source} command).
7628 In addition to the source path, @value{GDBN} provides a set of commands
7629 that manage a list of source path substitution rules. A @dfn{substitution
7630 rule} specifies how to rewrite source directories stored in the program's
7631 debug information in case the sources were moved to a different
7632 directory between compilation and debugging. A rule is made of
7633 two strings, the first specifying what needs to be rewritten in
7634 the path, and the second specifying how it should be rewritten.
7635 In @ref{set substitute-path}, we name these two parts @var{from} and
7636 @var{to} respectively. @value{GDBN} does a simple string replacement
7637 of @var{from} with @var{to} at the start of the directory part of the
7638 source file name, and uses that result instead of the original file
7639 name to look up the sources.
7641 Using the previous example, suppose the @file{foo-1.0} tree has been
7642 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
7643 @value{GDBN} to replace @file{/usr/src} in all source path names with
7644 @file{/mnt/cross}. The first lookup will then be
7645 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
7646 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
7647 substitution rule, use the @code{set substitute-path} command
7648 (@pxref{set substitute-path}).
7650 To avoid unexpected substitution results, a rule is applied only if the
7651 @var{from} part of the directory name ends at a directory separator.
7652 For instance, a rule substituting @file{/usr/source} into
7653 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
7654 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
7655 is applied only at the beginning of the directory name, this rule will
7656 not be applied to @file{/root/usr/source/baz.c} either.
7658 In many cases, you can achieve the same result using the @code{directory}
7659 command. However, @code{set substitute-path} can be more efficient in
7660 the case where the sources are organized in a complex tree with multiple
7661 subdirectories. With the @code{directory} command, you need to add each
7662 subdirectory of your project. If you moved the entire tree while
7663 preserving its internal organization, then @code{set substitute-path}
7664 allows you to direct the debugger to all the sources with one single
7667 @code{set substitute-path} is also more than just a shortcut command.
7668 The source path is only used if the file at the original location no
7669 longer exists. On the other hand, @code{set substitute-path} modifies
7670 the debugger behavior to look at the rewritten location instead. So, if
7671 for any reason a source file that is not relevant to your executable is
7672 located at the original location, a substitution rule is the only
7673 method available to point @value{GDBN} at the new location.
7675 @cindex @samp{--with-relocated-sources}
7676 @cindex default source path substitution
7677 You can configure a default source path substitution rule by
7678 configuring @value{GDBN} with the
7679 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
7680 should be the name of a directory under @value{GDBN}'s configured
7681 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
7682 directory names in debug information under @var{dir} will be adjusted
7683 automatically if the installed @value{GDBN} is moved to a new
7684 location. This is useful if @value{GDBN}, libraries or executables
7685 with debug information and corresponding source code are being moved
7689 @item directory @var{dirname} @dots{}
7690 @item dir @var{dirname} @dots{}
7691 Add directory @var{dirname} to the front of the source path. Several
7692 directory names may be given to this command, separated by @samp{:}
7693 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
7694 part of absolute file names) or
7695 whitespace. You may specify a directory that is already in the source
7696 path; this moves it forward, so @value{GDBN} searches it sooner.
7700 @vindex $cdir@r{, convenience variable}
7701 @vindex $cwd@r{, convenience variable}
7702 @cindex compilation directory
7703 @cindex current directory
7704 @cindex working directory
7705 @cindex directory, current
7706 @cindex directory, compilation
7707 You can use the string @samp{$cdir} to refer to the compilation
7708 directory (if one is recorded), and @samp{$cwd} to refer to the current
7709 working directory. @samp{$cwd} is not the same as @samp{.}---the former
7710 tracks the current working directory as it changes during your @value{GDBN}
7711 session, while the latter is immediately expanded to the current
7712 directory at the time you add an entry to the source path.
7715 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
7717 @c RET-repeat for @code{directory} is explicitly disabled, but since
7718 @c repeating it would be a no-op we do not say that. (thanks to RMS)
7720 @item set directories @var{path-list}
7721 @kindex set directories
7722 Set the source path to @var{path-list}.
7723 @samp{$cdir:$cwd} are added if missing.
7725 @item show directories
7726 @kindex show directories
7727 Print the source path: show which directories it contains.
7729 @anchor{set substitute-path}
7730 @item set substitute-path @var{from} @var{to}
7731 @kindex set substitute-path
7732 Define a source path substitution rule, and add it at the end of the
7733 current list of existing substitution rules. If a rule with the same
7734 @var{from} was already defined, then the old rule is also deleted.
7736 For example, if the file @file{/foo/bar/baz.c} was moved to
7737 @file{/mnt/cross/baz.c}, then the command
7740 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
7744 will tell @value{GDBN} to replace @samp{/usr/src} with
7745 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
7746 @file{baz.c} even though it was moved.
7748 In the case when more than one substitution rule have been defined,
7749 the rules are evaluated one by one in the order where they have been
7750 defined. The first one matching, if any, is selected to perform
7753 For instance, if we had entered the following commands:
7756 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
7757 (@value{GDBP}) set substitute-path /usr/src /mnt/src
7761 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
7762 @file{/mnt/include/defs.h} by using the first rule. However, it would
7763 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
7764 @file{/mnt/src/lib/foo.c}.
7767 @item unset substitute-path [path]
7768 @kindex unset substitute-path
7769 If a path is specified, search the current list of substitution rules
7770 for a rule that would rewrite that path. Delete that rule if found.
7771 A warning is emitted by the debugger if no rule could be found.
7773 If no path is specified, then all substitution rules are deleted.
7775 @item show substitute-path [path]
7776 @kindex show substitute-path
7777 If a path is specified, then print the source path substitution rule
7778 which would rewrite that path, if any.
7780 If no path is specified, then print all existing source path substitution
7785 If your source path is cluttered with directories that are no longer of
7786 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
7787 versions of source. You can correct the situation as follows:
7791 Use @code{directory} with no argument to reset the source path to its default value.
7794 Use @code{directory} with suitable arguments to reinstall the
7795 directories you want in the source path. You can add all the
7796 directories in one command.
7800 @section Source and Machine Code
7801 @cindex source line and its code address
7803 You can use the command @code{info line} to map source lines to program
7804 addresses (and vice versa), and the command @code{disassemble} to display
7805 a range of addresses as machine instructions. You can use the command
7806 @code{set disassemble-next-line} to set whether to disassemble next
7807 source line when execution stops. When run under @sc{gnu} Emacs
7808 mode, the @code{info line} command causes the arrow to point to the
7809 line specified. Also, @code{info line} prints addresses in symbolic form as
7814 @item info line @var{linespec}
7815 Print the starting and ending addresses of the compiled code for
7816 source line @var{linespec}. You can specify source lines in any of
7817 the ways documented in @ref{Specify Location}.
7820 For example, we can use @code{info line} to discover the location of
7821 the object code for the first line of function
7822 @code{m4_changequote}:
7824 @c FIXME: I think this example should also show the addresses in
7825 @c symbolic form, as they usually would be displayed.
7827 (@value{GDBP}) info line m4_changequote
7828 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
7832 @cindex code address and its source line
7833 We can also inquire (using @code{*@var{addr}} as the form for
7834 @var{linespec}) what source line covers a particular address:
7836 (@value{GDBP}) info line *0x63ff
7837 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
7840 @cindex @code{$_} and @code{info line}
7841 @cindex @code{x} command, default address
7842 @kindex x@r{(examine), and} info line
7843 After @code{info line}, the default address for the @code{x} command
7844 is changed to the starting address of the line, so that @samp{x/i} is
7845 sufficient to begin examining the machine code (@pxref{Memory,
7846 ,Examining Memory}). Also, this address is saved as the value of the
7847 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
7852 @cindex assembly instructions
7853 @cindex instructions, assembly
7854 @cindex machine instructions
7855 @cindex listing machine instructions
7857 @itemx disassemble /m
7858 @itemx disassemble /r
7859 This specialized command dumps a range of memory as machine
7860 instructions. It can also print mixed source+disassembly by specifying
7861 the @code{/m} modifier and print the raw instructions in hex as well as
7862 in symbolic form by specifying the @code{/r}.
7863 The default memory range is the function surrounding the
7864 program counter of the selected frame. A single argument to this
7865 command is a program counter value; @value{GDBN} dumps the function
7866 surrounding this value. When two arguments are given, they should
7867 be separated by a comma, possibly surrounded by whitespace. The
7868 arguments specify a range of addresses to dump, in one of two forms:
7871 @item @var{start},@var{end}
7872 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
7873 @item @var{start},+@var{length}
7874 the addresses from @var{start} (inclusive) to
7875 @code{@var{start}+@var{length}} (exclusive).
7879 When 2 arguments are specified, the name of the function is also
7880 printed (since there could be several functions in the given range).
7882 The argument(s) can be any expression yielding a numeric value, such as
7883 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
7885 If the range of memory being disassembled contains current program counter,
7886 the instruction at that location is shown with a @code{=>} marker.
7889 The following example shows the disassembly of a range of addresses of
7890 HP PA-RISC 2.0 code:
7893 (@value{GDBP}) disas 0x32c4, 0x32e4
7894 Dump of assembler code from 0x32c4 to 0x32e4:
7895 0x32c4 <main+204>: addil 0,dp
7896 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
7897 0x32cc <main+212>: ldil 0x3000,r31
7898 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
7899 0x32d4 <main+220>: ldo 0(r31),rp
7900 0x32d8 <main+224>: addil -0x800,dp
7901 0x32dc <main+228>: ldo 0x588(r1),r26
7902 0x32e0 <main+232>: ldil 0x3000,r31
7903 End of assembler dump.
7906 Here is an example showing mixed source+assembly for Intel x86, when the
7907 program is stopped just after function prologue:
7910 (@value{GDBP}) disas /m main
7911 Dump of assembler code for function main:
7913 0x08048330 <+0>: push %ebp
7914 0x08048331 <+1>: mov %esp,%ebp
7915 0x08048333 <+3>: sub $0x8,%esp
7916 0x08048336 <+6>: and $0xfffffff0,%esp
7917 0x08048339 <+9>: sub $0x10,%esp
7919 6 printf ("Hello.\n");
7920 => 0x0804833c <+12>: movl $0x8048440,(%esp)
7921 0x08048343 <+19>: call 0x8048284 <puts@@plt>
7925 0x08048348 <+24>: mov $0x0,%eax
7926 0x0804834d <+29>: leave
7927 0x0804834e <+30>: ret
7929 End of assembler dump.
7932 Here is another example showing raw instructions in hex for AMD x86-64,
7935 (gdb) disas /r 0x400281,+10
7936 Dump of assembler code from 0x400281 to 0x40028b:
7937 0x0000000000400281: 38 36 cmp %dh,(%rsi)
7938 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
7939 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
7940 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
7941 End of assembler dump.
7944 Addresses cannot be specified as a linespec (@pxref{Specify Location}).
7945 So, for example, if you want to disassemble function @code{bar}
7946 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
7947 and not @samp{disassemble foo.c:bar}.
7949 Some architectures have more than one commonly-used set of instruction
7950 mnemonics or other syntax.
7952 For programs that were dynamically linked and use shared libraries,
7953 instructions that call functions or branch to locations in the shared
7954 libraries might show a seemingly bogus location---it's actually a
7955 location of the relocation table. On some architectures, @value{GDBN}
7956 might be able to resolve these to actual function names.
7959 @kindex set disassembly-flavor
7960 @cindex Intel disassembly flavor
7961 @cindex AT&T disassembly flavor
7962 @item set disassembly-flavor @var{instruction-set}
7963 Select the instruction set to use when disassembling the
7964 program via the @code{disassemble} or @code{x/i} commands.
7966 Currently this command is only defined for the Intel x86 family. You
7967 can set @var{instruction-set} to either @code{intel} or @code{att}.
7968 The default is @code{att}, the AT&T flavor used by default by Unix
7969 assemblers for x86-based targets.
7971 @kindex show disassembly-flavor
7972 @item show disassembly-flavor
7973 Show the current setting of the disassembly flavor.
7977 @kindex set disassemble-next-line
7978 @kindex show disassemble-next-line
7979 @item set disassemble-next-line
7980 @itemx show disassemble-next-line
7981 Control whether or not @value{GDBN} will disassemble the next source
7982 line or instruction when execution stops. If ON, @value{GDBN} will
7983 display disassembly of the next source line when execution of the
7984 program being debugged stops. This is @emph{in addition} to
7985 displaying the source line itself, which @value{GDBN} always does if
7986 possible. If the next source line cannot be displayed for some reason
7987 (e.g., if @value{GDBN} cannot find the source file, or there's no line
7988 info in the debug info), @value{GDBN} will display disassembly of the
7989 next @emph{instruction} instead of showing the next source line. If
7990 AUTO, @value{GDBN} will display disassembly of next instruction only
7991 if the source line cannot be displayed. This setting causes
7992 @value{GDBN} to display some feedback when you step through a function
7993 with no line info or whose source file is unavailable. The default is
7994 OFF, which means never display the disassembly of the next line or
8000 @chapter Examining Data
8002 @cindex printing data
8003 @cindex examining data
8006 The usual way to examine data in your program is with the @code{print}
8007 command (abbreviated @code{p}), or its synonym @code{inspect}. It
8008 evaluates and prints the value of an expression of the language your
8009 program is written in (@pxref{Languages, ,Using @value{GDBN} with
8010 Different Languages}). It may also print the expression using a
8011 Python-based pretty-printer (@pxref{Pretty Printing}).
8014 @item print @var{expr}
8015 @itemx print /@var{f} @var{expr}
8016 @var{expr} is an expression (in the source language). By default the
8017 value of @var{expr} is printed in a format appropriate to its data type;
8018 you can choose a different format by specifying @samp{/@var{f}}, where
8019 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
8023 @itemx print /@var{f}
8024 @cindex reprint the last value
8025 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
8026 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
8027 conveniently inspect the same value in an alternative format.
8030 A more low-level way of examining data is with the @code{x} command.
8031 It examines data in memory at a specified address and prints it in a
8032 specified format. @xref{Memory, ,Examining Memory}.
8034 If you are interested in information about types, or about how the
8035 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
8036 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
8039 @cindex exploring hierarchical data structures
8041 Another way of examining values of expressions and type information is
8042 through the Python extension command @code{explore} (available only if
8043 the @value{GDBN} build is configured with @code{--with-python}). It
8044 offers an interactive way to start at the highest level (or, the most
8045 abstract level) of the data type of an expression (or, the data type
8046 itself) and explore all the way down to leaf scalar values/fields
8047 embedded in the higher level data types.
8050 @item explore @var{arg}
8051 @var{arg} is either an expression (in the source language), or a type
8052 visible in the current context of the program being debugged.
8055 The working of the @code{explore} command can be illustrated with an
8056 example. If a data type @code{struct ComplexStruct} is defined in your
8066 struct ComplexStruct
8068 struct SimpleStruct *ss_p;
8074 followed by variable declarations as
8077 struct SimpleStruct ss = @{ 10, 1.11 @};
8078 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
8082 then, the value of the variable @code{cs} can be explored using the
8083 @code{explore} command as follows.
8087 The value of `cs' is a struct/class of type `struct ComplexStruct' with
8088 the following fields:
8090 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
8091 arr = <Enter 1 to explore this field of type `int [10]'>
8093 Enter the field number of choice:
8097 Since the fields of @code{cs} are not scalar values, you are being
8098 prompted to chose the field you want to explore. Let's say you choose
8099 the field @code{ss_p} by entering @code{0}. Then, since this field is a
8100 pointer, you will be asked if it is pointing to a single value. From
8101 the declaration of @code{cs} above, it is indeed pointing to a single
8102 value, hence you enter @code{y}. If you enter @code{n}, then you will
8103 be asked if it were pointing to an array of values, in which case this
8104 field will be explored as if it were an array.
8107 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
8108 Continue exploring it as a pointer to a single value [y/n]: y
8109 The value of `*(cs.ss_p)' is a struct/class of type `struct
8110 SimpleStruct' with the following fields:
8112 i = 10 .. (Value of type `int')
8113 d = 1.1100000000000001 .. (Value of type `double')
8115 Press enter to return to parent value:
8119 If the field @code{arr} of @code{cs} was chosen for exploration by
8120 entering @code{1} earlier, then since it is as array, you will be
8121 prompted to enter the index of the element in the array that you want
8125 `cs.arr' is an array of `int'.
8126 Enter the index of the element you want to explore in `cs.arr': 5
8128 `(cs.arr)[5]' is a scalar value of type `int'.
8132 Press enter to return to parent value:
8135 In general, at any stage of exploration, you can go deeper towards the
8136 leaf values by responding to the prompts appropriately, or hit the
8137 return key to return to the enclosing data structure (the @i{higher}
8138 level data structure).
8140 Similar to exploring values, you can use the @code{explore} command to
8141 explore types. Instead of specifying a value (which is typically a
8142 variable name or an expression valid in the current context of the
8143 program being debugged), you specify a type name. If you consider the
8144 same example as above, your can explore the type
8145 @code{struct ComplexStruct} by passing the argument
8146 @code{struct ComplexStruct} to the @code{explore} command.
8149 (gdb) explore struct ComplexStruct
8153 By responding to the prompts appropriately in the subsequent interactive
8154 session, you can explore the type @code{struct ComplexStruct} in a
8155 manner similar to how the value @code{cs} was explored in the above
8158 The @code{explore} command also has two sub-commands,
8159 @code{explore value} and @code{explore type}. The former sub-command is
8160 a way to explicitly specify that value exploration of the argument is
8161 being invoked, while the latter is a way to explicitly specify that type
8162 exploration of the argument is being invoked.
8165 @item explore value @var{expr}
8166 @cindex explore value
8167 This sub-command of @code{explore} explores the value of the
8168 expression @var{expr} (if @var{expr} is an expression valid in the
8169 current context of the program being debugged). The behavior of this
8170 command is identical to that of the behavior of the @code{explore}
8171 command being passed the argument @var{expr}.
8173 @item explore type @var{arg}
8174 @cindex explore type
8175 This sub-command of @code{explore} explores the type of @var{arg} (if
8176 @var{arg} is a type visible in the current context of program being
8177 debugged), or the type of the value/expression @var{arg} (if @var{arg}
8178 is an expression valid in the current context of the program being
8179 debugged). If @var{arg} is a type, then the behavior of this command is
8180 identical to that of the @code{explore} command being passed the
8181 argument @var{arg}. If @var{arg} is an expression, then the behavior of
8182 this command will be identical to that of the @code{explore} command
8183 being passed the type of @var{arg} as the argument.
8187 * Expressions:: Expressions
8188 * Ambiguous Expressions:: Ambiguous Expressions
8189 * Variables:: Program variables
8190 * Arrays:: Artificial arrays
8191 * Output Formats:: Output formats
8192 * Memory:: Examining memory
8193 * Auto Display:: Automatic display
8194 * Print Settings:: Print settings
8195 * Pretty Printing:: Python pretty printing
8196 * Value History:: Value history
8197 * Convenience Vars:: Convenience variables
8198 * Convenience Funs:: Convenience functions
8199 * Registers:: Registers
8200 * Floating Point Hardware:: Floating point hardware
8201 * Vector Unit:: Vector Unit
8202 * OS Information:: Auxiliary data provided by operating system
8203 * Memory Region Attributes:: Memory region attributes
8204 * Dump/Restore Files:: Copy between memory and a file
8205 * Core File Generation:: Cause a program dump its core
8206 * Character Sets:: Debugging programs that use a different
8207 character set than GDB does
8208 * Caching Target Data:: Data caching for targets
8209 * Searching Memory:: Searching memory for a sequence of bytes
8213 @section Expressions
8216 @code{print} and many other @value{GDBN} commands accept an expression and
8217 compute its value. Any kind of constant, variable or operator defined
8218 by the programming language you are using is valid in an expression in
8219 @value{GDBN}. This includes conditional expressions, function calls,
8220 casts, and string constants. It also includes preprocessor macros, if
8221 you compiled your program to include this information; see
8224 @cindex arrays in expressions
8225 @value{GDBN} supports array constants in expressions input by
8226 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
8227 you can use the command @code{print @{1, 2, 3@}} to create an array
8228 of three integers. If you pass an array to a function or assign it
8229 to a program variable, @value{GDBN} copies the array to memory that
8230 is @code{malloc}ed in the target program.
8232 Because C is so widespread, most of the expressions shown in examples in
8233 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
8234 Languages}, for information on how to use expressions in other
8237 In this section, we discuss operators that you can use in @value{GDBN}
8238 expressions regardless of your programming language.
8240 @cindex casts, in expressions
8241 Casts are supported in all languages, not just in C, because it is so
8242 useful to cast a number into a pointer in order to examine a structure
8243 at that address in memory.
8244 @c FIXME: casts supported---Mod2 true?
8246 @value{GDBN} supports these operators, in addition to those common
8247 to programming languages:
8251 @samp{@@} is a binary operator for treating parts of memory as arrays.
8252 @xref{Arrays, ,Artificial Arrays}, for more information.
8255 @samp{::} allows you to specify a variable in terms of the file or
8256 function where it is defined. @xref{Variables, ,Program Variables}.
8258 @cindex @{@var{type}@}
8259 @cindex type casting memory
8260 @cindex memory, viewing as typed object
8261 @cindex casts, to view memory
8262 @item @{@var{type}@} @var{addr}
8263 Refers to an object of type @var{type} stored at address @var{addr} in
8264 memory. The address @var{addr} may be any expression whose value is
8265 an integer or pointer (but parentheses are required around binary
8266 operators, just as in a cast). This construct is allowed regardless
8267 of what kind of data is normally supposed to reside at @var{addr}.
8270 @node Ambiguous Expressions
8271 @section Ambiguous Expressions
8272 @cindex ambiguous expressions
8274 Expressions can sometimes contain some ambiguous elements. For instance,
8275 some programming languages (notably Ada, C@t{++} and Objective-C) permit
8276 a single function name to be defined several times, for application in
8277 different contexts. This is called @dfn{overloading}. Another example
8278 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
8279 templates and is typically instantiated several times, resulting in
8280 the same function name being defined in different contexts.
8282 In some cases and depending on the language, it is possible to adjust
8283 the expression to remove the ambiguity. For instance in C@t{++}, you
8284 can specify the signature of the function you want to break on, as in
8285 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
8286 qualified name of your function often makes the expression unambiguous
8289 When an ambiguity that needs to be resolved is detected, the debugger
8290 has the capability to display a menu of numbered choices for each
8291 possibility, and then waits for the selection with the prompt @samp{>}.
8292 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
8293 aborts the current command. If the command in which the expression was
8294 used allows more than one choice to be selected, the next option in the
8295 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
8298 For example, the following session excerpt shows an attempt to set a
8299 breakpoint at the overloaded symbol @code{String::after}.
8300 We choose three particular definitions of that function name:
8302 @c FIXME! This is likely to change to show arg type lists, at least
8305 (@value{GDBP}) b String::after
8308 [2] file:String.cc; line number:867
8309 [3] file:String.cc; line number:860
8310 [4] file:String.cc; line number:875
8311 [5] file:String.cc; line number:853
8312 [6] file:String.cc; line number:846
8313 [7] file:String.cc; line number:735
8315 Breakpoint 1 at 0xb26c: file String.cc, line 867.
8316 Breakpoint 2 at 0xb344: file String.cc, line 875.
8317 Breakpoint 3 at 0xafcc: file String.cc, line 846.
8318 Multiple breakpoints were set.
8319 Use the "delete" command to delete unwanted
8326 @kindex set multiple-symbols
8327 @item set multiple-symbols @var{mode}
8328 @cindex multiple-symbols menu
8330 This option allows you to adjust the debugger behavior when an expression
8333 By default, @var{mode} is set to @code{all}. If the command with which
8334 the expression is used allows more than one choice, then @value{GDBN}
8335 automatically selects all possible choices. For instance, inserting
8336 a breakpoint on a function using an ambiguous name results in a breakpoint
8337 inserted on each possible match. However, if a unique choice must be made,
8338 then @value{GDBN} uses the menu to help you disambiguate the expression.
8339 For instance, printing the address of an overloaded function will result
8340 in the use of the menu.
8342 When @var{mode} is set to @code{ask}, the debugger always uses the menu
8343 when an ambiguity is detected.
8345 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
8346 an error due to the ambiguity and the command is aborted.
8348 @kindex show multiple-symbols
8349 @item show multiple-symbols
8350 Show the current value of the @code{multiple-symbols} setting.
8354 @section Program Variables
8356 The most common kind of expression to use is the name of a variable
8359 Variables in expressions are understood in the selected stack frame
8360 (@pxref{Selection, ,Selecting a Frame}); they must be either:
8364 global (or file-static)
8371 visible according to the scope rules of the
8372 programming language from the point of execution in that frame
8375 @noindent This means that in the function
8390 you can examine and use the variable @code{a} whenever your program is
8391 executing within the function @code{foo}, but you can only use or
8392 examine the variable @code{b} while your program is executing inside
8393 the block where @code{b} is declared.
8395 @cindex variable name conflict
8396 There is an exception: you can refer to a variable or function whose
8397 scope is a single source file even if the current execution point is not
8398 in this file. But it is possible to have more than one such variable or
8399 function with the same name (in different source files). If that
8400 happens, referring to that name has unpredictable effects. If you wish,
8401 you can specify a static variable in a particular function or file by
8402 using the colon-colon (@code{::}) notation:
8404 @cindex colon-colon, context for variables/functions
8406 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
8407 @cindex @code{::}, context for variables/functions
8410 @var{file}::@var{variable}
8411 @var{function}::@var{variable}
8415 Here @var{file} or @var{function} is the name of the context for the
8416 static @var{variable}. In the case of file names, you can use quotes to
8417 make sure @value{GDBN} parses the file name as a single word---for example,
8418 to print a global value of @code{x} defined in @file{f2.c}:
8421 (@value{GDBP}) p 'f2.c'::x
8424 The @code{::} notation is normally used for referring to
8425 static variables, since you typically disambiguate uses of local variables
8426 in functions by selecting the appropriate frame and using the
8427 simple name of the variable. However, you may also use this notation
8428 to refer to local variables in frames enclosing the selected frame:
8437 process (a); /* Stop here */
8448 For example, if there is a breakpoint at the commented line,
8449 here is what you might see
8450 when the program stops after executing the call @code{bar(0)}:
8455 (@value{GDBP}) p bar::a
8458 #2 0x080483d0 in foo (a=5) at foobar.c:12
8461 (@value{GDBP}) p bar::a
8465 @cindex C@t{++} scope resolution
8466 These uses of @samp{::} are very rarely in conflict with the very
8467 similar use of the same notation in C@t{++}. When they are in
8468 conflict, the C@t{++} meaning takes precedence; however, this can be
8469 overridden by quoting the file or function name with single quotes.
8471 For example, suppose the program is stopped in a method of a class
8472 that has a field named @code{includefile}, and there is also an
8473 include file named @file{includefile} that defines a variable,
8477 (@value{GDBP}) p includefile
8479 (@value{GDBP}) p includefile::some_global
8480 A syntax error in expression, near `'.
8481 (@value{GDBP}) p 'includefile'::some_global
8485 @cindex wrong values
8486 @cindex variable values, wrong
8487 @cindex function entry/exit, wrong values of variables
8488 @cindex optimized code, wrong values of variables
8490 @emph{Warning:} Occasionally, a local variable may appear to have the
8491 wrong value at certain points in a function---just after entry to a new
8492 scope, and just before exit.
8494 You may see this problem when you are stepping by machine instructions.
8495 This is because, on most machines, it takes more than one instruction to
8496 set up a stack frame (including local variable definitions); if you are
8497 stepping by machine instructions, variables may appear to have the wrong
8498 values until the stack frame is completely built. On exit, it usually
8499 also takes more than one machine instruction to destroy a stack frame;
8500 after you begin stepping through that group of instructions, local
8501 variable definitions may be gone.
8503 This may also happen when the compiler does significant optimizations.
8504 To be sure of always seeing accurate values, turn off all optimization
8507 @cindex ``No symbol "foo" in current context''
8508 Another possible effect of compiler optimizations is to optimize
8509 unused variables out of existence, or assign variables to registers (as
8510 opposed to memory addresses). Depending on the support for such cases
8511 offered by the debug info format used by the compiler, @value{GDBN}
8512 might not be able to display values for such local variables. If that
8513 happens, @value{GDBN} will print a message like this:
8516 No symbol "foo" in current context.
8519 To solve such problems, either recompile without optimizations, or use a
8520 different debug info format, if the compiler supports several such
8521 formats. @xref{Compilation}, for more information on choosing compiler
8522 options. @xref{C, ,C and C@t{++}}, for more information about debug
8523 info formats that are best suited to C@t{++} programs.
8525 If you ask to print an object whose contents are unknown to
8526 @value{GDBN}, e.g., because its data type is not completely specified
8527 by the debug information, @value{GDBN} will say @samp{<incomplete
8528 type>}. @xref{Symbols, incomplete type}, for more about this.
8530 If you append @kbd{@@entry} string to a function parameter name you get its
8531 value at the time the function got called. If the value is not available an
8532 error message is printed. Entry values are available only with some compilers.
8533 Entry values are normally also printed at the function parameter list according
8534 to @ref{set print entry-values}.
8537 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
8543 (gdb) print i@@entry
8547 Strings are identified as arrays of @code{char} values without specified
8548 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
8549 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
8550 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
8551 defines literal string type @code{"char"} as @code{char} without a sign.
8556 signed char var1[] = "A";
8559 You get during debugging
8564 $2 = @{65 'A', 0 '\0'@}
8568 @section Artificial Arrays
8570 @cindex artificial array
8572 @kindex @@@r{, referencing memory as an array}
8573 It is often useful to print out several successive objects of the
8574 same type in memory; a section of an array, or an array of
8575 dynamically determined size for which only a pointer exists in the
8578 You can do this by referring to a contiguous span of memory as an
8579 @dfn{artificial array}, using the binary operator @samp{@@}. The left
8580 operand of @samp{@@} should be the first element of the desired array
8581 and be an individual object. The right operand should be the desired length
8582 of the array. The result is an array value whose elements are all of
8583 the type of the left argument. The first element is actually the left
8584 argument; the second element comes from bytes of memory immediately
8585 following those that hold the first element, and so on. Here is an
8586 example. If a program says
8589 int *array = (int *) malloc (len * sizeof (int));
8593 you can print the contents of @code{array} with
8599 The left operand of @samp{@@} must reside in memory. Array values made
8600 with @samp{@@} in this way behave just like other arrays in terms of
8601 subscripting, and are coerced to pointers when used in expressions.
8602 Artificial arrays most often appear in expressions via the value history
8603 (@pxref{Value History, ,Value History}), after printing one out.
8605 Another way to create an artificial array is to use a cast.
8606 This re-interprets a value as if it were an array.
8607 The value need not be in memory:
8609 (@value{GDBP}) p/x (short[2])0x12345678
8610 $1 = @{0x1234, 0x5678@}
8613 As a convenience, if you leave the array length out (as in
8614 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
8615 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
8617 (@value{GDBP}) p/x (short[])0x12345678
8618 $2 = @{0x1234, 0x5678@}
8621 Sometimes the artificial array mechanism is not quite enough; in
8622 moderately complex data structures, the elements of interest may not
8623 actually be adjacent---for example, if you are interested in the values
8624 of pointers in an array. One useful work-around in this situation is
8625 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
8626 Variables}) as a counter in an expression that prints the first
8627 interesting value, and then repeat that expression via @key{RET}. For
8628 instance, suppose you have an array @code{dtab} of pointers to
8629 structures, and you are interested in the values of a field @code{fv}
8630 in each structure. Here is an example of what you might type:
8640 @node Output Formats
8641 @section Output Formats
8643 @cindex formatted output
8644 @cindex output formats
8645 By default, @value{GDBN} prints a value according to its data type. Sometimes
8646 this is not what you want. For example, you might want to print a number
8647 in hex, or a pointer in decimal. Or you might want to view data in memory
8648 at a certain address as a character string or as an instruction. To do
8649 these things, specify an @dfn{output format} when you print a value.
8651 The simplest use of output formats is to say how to print a value
8652 already computed. This is done by starting the arguments of the
8653 @code{print} command with a slash and a format letter. The format
8654 letters supported are:
8658 Regard the bits of the value as an integer, and print the integer in
8662 Print as integer in signed decimal.
8665 Print as integer in unsigned decimal.
8668 Print as integer in octal.
8671 Print as integer in binary. The letter @samp{t} stands for ``two''.
8672 @footnote{@samp{b} cannot be used because these format letters are also
8673 used with the @code{x} command, where @samp{b} stands for ``byte'';
8674 see @ref{Memory,,Examining Memory}.}
8677 @cindex unknown address, locating
8678 @cindex locate address
8679 Print as an address, both absolute in hexadecimal and as an offset from
8680 the nearest preceding symbol. You can use this format used to discover
8681 where (in what function) an unknown address is located:
8684 (@value{GDBP}) p/a 0x54320
8685 $3 = 0x54320 <_initialize_vx+396>
8689 The command @code{info symbol 0x54320} yields similar results.
8690 @xref{Symbols, info symbol}.
8693 Regard as an integer and print it as a character constant. This
8694 prints both the numerical value and its character representation. The
8695 character representation is replaced with the octal escape @samp{\nnn}
8696 for characters outside the 7-bit @sc{ascii} range.
8698 Without this format, @value{GDBN} displays @code{char},
8699 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
8700 constants. Single-byte members of vectors are displayed as integer
8704 Regard the bits of the value as a floating point number and print
8705 using typical floating point syntax.
8708 @cindex printing strings
8709 @cindex printing byte arrays
8710 Regard as a string, if possible. With this format, pointers to single-byte
8711 data are displayed as null-terminated strings and arrays of single-byte data
8712 are displayed as fixed-length strings. Other values are displayed in their
8715 Without this format, @value{GDBN} displays pointers to and arrays of
8716 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
8717 strings. Single-byte members of a vector are displayed as an integer
8721 Like @samp{x} formatting, the value is treated as an integer and
8722 printed as hexadecimal, but leading zeros are printed to pad the value
8723 to the size of the integer type.
8726 @cindex raw printing
8727 Print using the @samp{raw} formatting. By default, @value{GDBN} will
8728 use a Python-based pretty-printer, if one is available (@pxref{Pretty
8729 Printing}). This typically results in a higher-level display of the
8730 value's contents. The @samp{r} format bypasses any Python
8731 pretty-printer which might exist.
8734 For example, to print the program counter in hex (@pxref{Registers}), type
8741 Note that no space is required before the slash; this is because command
8742 names in @value{GDBN} cannot contain a slash.
8744 To reprint the last value in the value history with a different format,
8745 you can use the @code{print} command with just a format and no
8746 expression. For example, @samp{p/x} reprints the last value in hex.
8749 @section Examining Memory
8751 You can use the command @code{x} (for ``examine'') to examine memory in
8752 any of several formats, independently of your program's data types.
8754 @cindex examining memory
8756 @kindex x @r{(examine memory)}
8757 @item x/@var{nfu} @var{addr}
8760 Use the @code{x} command to examine memory.
8763 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
8764 much memory to display and how to format it; @var{addr} is an
8765 expression giving the address where you want to start displaying memory.
8766 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
8767 Several commands set convenient defaults for @var{addr}.
8770 @item @var{n}, the repeat count
8771 The repeat count is a decimal integer; the default is 1. It specifies
8772 how much memory (counting by units @var{u}) to display.
8773 @c This really is **decimal**; unaffected by 'set radix' as of GDB
8776 @item @var{f}, the display format
8777 The display format is one of the formats used by @code{print}
8778 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
8779 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
8780 The default is @samp{x} (hexadecimal) initially. The default changes
8781 each time you use either @code{x} or @code{print}.
8783 @item @var{u}, the unit size
8784 The unit size is any of
8790 Halfwords (two bytes).
8792 Words (four bytes). This is the initial default.
8794 Giant words (eight bytes).
8797 Each time you specify a unit size with @code{x}, that size becomes the
8798 default unit the next time you use @code{x}. For the @samp{i} format,
8799 the unit size is ignored and is normally not written. For the @samp{s} format,
8800 the unit size defaults to @samp{b}, unless it is explicitly given.
8801 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
8802 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
8803 Note that the results depend on the programming language of the
8804 current compilation unit. If the language is C, the @samp{s}
8805 modifier will use the UTF-16 encoding while @samp{w} will use
8806 UTF-32. The encoding is set by the programming language and cannot
8809 @item @var{addr}, starting display address
8810 @var{addr} is the address where you want @value{GDBN} to begin displaying
8811 memory. The expression need not have a pointer value (though it may);
8812 it is always interpreted as an integer address of a byte of memory.
8813 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
8814 @var{addr} is usually just after the last address examined---but several
8815 other commands also set the default address: @code{info breakpoints} (to
8816 the address of the last breakpoint listed), @code{info line} (to the
8817 starting address of a line), and @code{print} (if you use it to display
8818 a value from memory).
8821 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
8822 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
8823 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
8824 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
8825 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
8827 Since the letters indicating unit sizes are all distinct from the
8828 letters specifying output formats, you do not have to remember whether
8829 unit size or format comes first; either order works. The output
8830 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
8831 (However, the count @var{n} must come first; @samp{wx4} does not work.)
8833 Even though the unit size @var{u} is ignored for the formats @samp{s}
8834 and @samp{i}, you might still want to use a count @var{n}; for example,
8835 @samp{3i} specifies that you want to see three machine instructions,
8836 including any operands. For convenience, especially when used with
8837 the @code{display} command, the @samp{i} format also prints branch delay
8838 slot instructions, if any, beyond the count specified, which immediately
8839 follow the last instruction that is within the count. The command
8840 @code{disassemble} gives an alternative way of inspecting machine
8841 instructions; see @ref{Machine Code,,Source and Machine Code}.
8843 All the defaults for the arguments to @code{x} are designed to make it
8844 easy to continue scanning memory with minimal specifications each time
8845 you use @code{x}. For example, after you have inspected three machine
8846 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
8847 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
8848 the repeat count @var{n} is used again; the other arguments default as
8849 for successive uses of @code{x}.
8851 When examining machine instructions, the instruction at current program
8852 counter is shown with a @code{=>} marker. For example:
8855 (@value{GDBP}) x/5i $pc-6
8856 0x804837f <main+11>: mov %esp,%ebp
8857 0x8048381 <main+13>: push %ecx
8858 0x8048382 <main+14>: sub $0x4,%esp
8859 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
8860 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
8863 @cindex @code{$_}, @code{$__}, and value history
8864 The addresses and contents printed by the @code{x} command are not saved
8865 in the value history because there is often too much of them and they
8866 would get in the way. Instead, @value{GDBN} makes these values available for
8867 subsequent use in expressions as values of the convenience variables
8868 @code{$_} and @code{$__}. After an @code{x} command, the last address
8869 examined is available for use in expressions in the convenience variable
8870 @code{$_}. The contents of that address, as examined, are available in
8871 the convenience variable @code{$__}.
8873 If the @code{x} command has a repeat count, the address and contents saved
8874 are from the last memory unit printed; this is not the same as the last
8875 address printed if several units were printed on the last line of output.
8877 @cindex remote memory comparison
8878 @cindex target memory comparison
8879 @cindex verify remote memory image
8880 @cindex verify target memory image
8881 When you are debugging a program running on a remote target machine
8882 (@pxref{Remote Debugging}), you may wish to verify the program's image
8883 in the remote machine's memory against the executable file you
8884 downloaded to the target. Or, on any target, you may want to check
8885 whether the program has corrupted its own read-only sections. The
8886 @code{compare-sections} command is provided for such situations.
8889 @kindex compare-sections
8890 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
8891 Compare the data of a loadable section @var{section-name} in the
8892 executable file of the program being debugged with the same section in
8893 the target machine's memory, and report any mismatches. With no
8894 arguments, compares all loadable sections. With an argument of
8895 @code{-r}, compares all loadable read-only sections.
8897 Note: for remote targets, this command can be accelerated if the
8898 target supports computing the CRC checksum of a block of memory
8899 (@pxref{qCRC packet}).
8903 @section Automatic Display
8904 @cindex automatic display
8905 @cindex display of expressions
8907 If you find that you want to print the value of an expression frequently
8908 (to see how it changes), you might want to add it to the @dfn{automatic
8909 display list} so that @value{GDBN} prints its value each time your program stops.
8910 Each expression added to the list is given a number to identify it;
8911 to remove an expression from the list, you specify that number.
8912 The automatic display looks like this:
8916 3: bar[5] = (struct hack *) 0x3804
8920 This display shows item numbers, expressions and their current values. As with
8921 displays you request manually using @code{x} or @code{print}, you can
8922 specify the output format you prefer; in fact, @code{display} decides
8923 whether to use @code{print} or @code{x} depending your format
8924 specification---it uses @code{x} if you specify either the @samp{i}
8925 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
8929 @item display @var{expr}
8930 Add the expression @var{expr} to the list of expressions to display
8931 each time your program stops. @xref{Expressions, ,Expressions}.
8933 @code{display} does not repeat if you press @key{RET} again after using it.
8935 @item display/@var{fmt} @var{expr}
8936 For @var{fmt} specifying only a display format and not a size or
8937 count, add the expression @var{expr} to the auto-display list but
8938 arrange to display it each time in the specified format @var{fmt}.
8939 @xref{Output Formats,,Output Formats}.
8941 @item display/@var{fmt} @var{addr}
8942 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
8943 number of units, add the expression @var{addr} as a memory address to
8944 be examined each time your program stops. Examining means in effect
8945 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
8948 For example, @samp{display/i $pc} can be helpful, to see the machine
8949 instruction about to be executed each time execution stops (@samp{$pc}
8950 is a common name for the program counter; @pxref{Registers, ,Registers}).
8953 @kindex delete display
8955 @item undisplay @var{dnums}@dots{}
8956 @itemx delete display @var{dnums}@dots{}
8957 Remove items from the list of expressions to display. Specify the
8958 numbers of the displays that you want affected with the command
8959 argument @var{dnums}. It can be a single display number, one of the
8960 numbers shown in the first field of the @samp{info display} display;
8961 or it could be a range of display numbers, as in @code{2-4}.
8963 @code{undisplay} does not repeat if you press @key{RET} after using it.
8964 (Otherwise you would just get the error @samp{No display number @dots{}}.)
8966 @kindex disable display
8967 @item disable display @var{dnums}@dots{}
8968 Disable the display of item numbers @var{dnums}. A disabled display
8969 item is not printed automatically, but is not forgotten. It may be
8970 enabled again later. Specify the numbers of the displays that you
8971 want affected with the command argument @var{dnums}. It can be a
8972 single display number, one of the numbers shown in the first field of
8973 the @samp{info display} display; or it could be a range of display
8974 numbers, as in @code{2-4}.
8976 @kindex enable display
8977 @item enable display @var{dnums}@dots{}
8978 Enable display of item numbers @var{dnums}. It becomes effective once
8979 again in auto display of its expression, until you specify otherwise.
8980 Specify the numbers of the displays that you want affected with the
8981 command argument @var{dnums}. It can be a single display number, one
8982 of the numbers shown in the first field of the @samp{info display}
8983 display; or it could be a range of display numbers, as in @code{2-4}.
8986 Display the current values of the expressions on the list, just as is
8987 done when your program stops.
8989 @kindex info display
8991 Print the list of expressions previously set up to display
8992 automatically, each one with its item number, but without showing the
8993 values. This includes disabled expressions, which are marked as such.
8994 It also includes expressions which would not be displayed right now
8995 because they refer to automatic variables not currently available.
8998 @cindex display disabled out of scope
8999 If a display expression refers to local variables, then it does not make
9000 sense outside the lexical context for which it was set up. Such an
9001 expression is disabled when execution enters a context where one of its
9002 variables is not defined. For example, if you give the command
9003 @code{display last_char} while inside a function with an argument
9004 @code{last_char}, @value{GDBN} displays this argument while your program
9005 continues to stop inside that function. When it stops elsewhere---where
9006 there is no variable @code{last_char}---the display is disabled
9007 automatically. The next time your program stops where @code{last_char}
9008 is meaningful, you can enable the display expression once again.
9010 @node Print Settings
9011 @section Print Settings
9013 @cindex format options
9014 @cindex print settings
9015 @value{GDBN} provides the following ways to control how arrays, structures,
9016 and symbols are printed.
9019 These settings are useful for debugging programs in any language:
9023 @item set print address
9024 @itemx set print address on
9025 @cindex print/don't print memory addresses
9026 @value{GDBN} prints memory addresses showing the location of stack
9027 traces, structure values, pointer values, breakpoints, and so forth,
9028 even when it also displays the contents of those addresses. The default
9029 is @code{on}. For example, this is what a stack frame display looks like with
9030 @code{set print address on}:
9035 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
9037 530 if (lquote != def_lquote)
9041 @item set print address off
9042 Do not print addresses when displaying their contents. For example,
9043 this is the same stack frame displayed with @code{set print address off}:
9047 (@value{GDBP}) set print addr off
9049 #0 set_quotes (lq="<<", rq=">>") at input.c:530
9050 530 if (lquote != def_lquote)
9054 You can use @samp{set print address off} to eliminate all machine
9055 dependent displays from the @value{GDBN} interface. For example, with
9056 @code{print address off}, you should get the same text for backtraces on
9057 all machines---whether or not they involve pointer arguments.
9060 @item show print address
9061 Show whether or not addresses are to be printed.
9064 When @value{GDBN} prints a symbolic address, it normally prints the
9065 closest earlier symbol plus an offset. If that symbol does not uniquely
9066 identify the address (for example, it is a name whose scope is a single
9067 source file), you may need to clarify. One way to do this is with
9068 @code{info line}, for example @samp{info line *0x4537}. Alternately,
9069 you can set @value{GDBN} to print the source file and line number when
9070 it prints a symbolic address:
9073 @item set print symbol-filename on
9074 @cindex source file and line of a symbol
9075 @cindex symbol, source file and line
9076 Tell @value{GDBN} to print the source file name and line number of a
9077 symbol in the symbolic form of an address.
9079 @item set print symbol-filename off
9080 Do not print source file name and line number of a symbol. This is the
9083 @item show print symbol-filename
9084 Show whether or not @value{GDBN} will print the source file name and
9085 line number of a symbol in the symbolic form of an address.
9088 Another situation where it is helpful to show symbol filenames and line
9089 numbers is when disassembling code; @value{GDBN} shows you the line
9090 number and source file that corresponds to each instruction.
9092 Also, you may wish to see the symbolic form only if the address being
9093 printed is reasonably close to the closest earlier symbol:
9096 @item set print max-symbolic-offset @var{max-offset}
9097 @itemx set print max-symbolic-offset unlimited
9098 @cindex maximum value for offset of closest symbol
9099 Tell @value{GDBN} to only display the symbolic form of an address if the
9100 offset between the closest earlier symbol and the address is less than
9101 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
9102 to always print the symbolic form of an address if any symbol precedes
9103 it. Zero is equivalent to @code{unlimited}.
9105 @item show print max-symbolic-offset
9106 Ask how large the maximum offset is that @value{GDBN} prints in a
9110 @cindex wild pointer, interpreting
9111 @cindex pointer, finding referent
9112 If you have a pointer and you are not sure where it points, try
9113 @samp{set print symbol-filename on}. Then you can determine the name
9114 and source file location of the variable where it points, using
9115 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
9116 For example, here @value{GDBN} shows that a variable @code{ptt} points
9117 at another variable @code{t}, defined in @file{hi2.c}:
9120 (@value{GDBP}) set print symbol-filename on
9121 (@value{GDBP}) p/a ptt
9122 $4 = 0xe008 <t in hi2.c>
9126 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
9127 does not show the symbol name and filename of the referent, even with
9128 the appropriate @code{set print} options turned on.
9131 You can also enable @samp{/a}-like formatting all the time using
9132 @samp{set print symbol on}:
9135 @item set print symbol on
9136 Tell @value{GDBN} to print the symbol corresponding to an address, if
9139 @item set print symbol off
9140 Tell @value{GDBN} not to print the symbol corresponding to an
9141 address. In this mode, @value{GDBN} will still print the symbol
9142 corresponding to pointers to functions. This is the default.
9144 @item show print symbol
9145 Show whether @value{GDBN} will display the symbol corresponding to an
9149 Other settings control how different kinds of objects are printed:
9152 @item set print array
9153 @itemx set print array on
9154 @cindex pretty print arrays
9155 Pretty print arrays. This format is more convenient to read,
9156 but uses more space. The default is off.
9158 @item set print array off
9159 Return to compressed format for arrays.
9161 @item show print array
9162 Show whether compressed or pretty format is selected for displaying
9165 @cindex print array indexes
9166 @item set print array-indexes
9167 @itemx set print array-indexes on
9168 Print the index of each element when displaying arrays. May be more
9169 convenient to locate a given element in the array or quickly find the
9170 index of a given element in that printed array. The default is off.
9172 @item set print array-indexes off
9173 Stop printing element indexes when displaying arrays.
9175 @item show print array-indexes
9176 Show whether the index of each element is printed when displaying
9179 @item set print elements @var{number-of-elements}
9180 @itemx set print elements unlimited
9181 @cindex number of array elements to print
9182 @cindex limit on number of printed array elements
9183 Set a limit on how many elements of an array @value{GDBN} will print.
9184 If @value{GDBN} is printing a large array, it stops printing after it has
9185 printed the number of elements set by the @code{set print elements} command.
9186 This limit also applies to the display of strings.
9187 When @value{GDBN} starts, this limit is set to 200.
9188 Setting @var{number-of-elements} to @code{unlimited} or zero means
9189 that the number of elements to print is unlimited.
9191 @item show print elements
9192 Display the number of elements of a large array that @value{GDBN} will print.
9193 If the number is 0, then the printing is unlimited.
9195 @item set print frame-arguments @var{value}
9196 @kindex set print frame-arguments
9197 @cindex printing frame argument values
9198 @cindex print all frame argument values
9199 @cindex print frame argument values for scalars only
9200 @cindex do not print frame argument values
9201 This command allows to control how the values of arguments are printed
9202 when the debugger prints a frame (@pxref{Frames}). The possible
9207 The values of all arguments are printed.
9210 Print the value of an argument only if it is a scalar. The value of more
9211 complex arguments such as arrays, structures, unions, etc, is replaced
9212 by @code{@dots{}}. This is the default. Here is an example where
9213 only scalar arguments are shown:
9216 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
9221 None of the argument values are printed. Instead, the value of each argument
9222 is replaced by @code{@dots{}}. In this case, the example above now becomes:
9225 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
9230 By default, only scalar arguments are printed. This command can be used
9231 to configure the debugger to print the value of all arguments, regardless
9232 of their type. However, it is often advantageous to not print the value
9233 of more complex parameters. For instance, it reduces the amount of
9234 information printed in each frame, making the backtrace more readable.
9235 Also, it improves performance when displaying Ada frames, because
9236 the computation of large arguments can sometimes be CPU-intensive,
9237 especially in large applications. Setting @code{print frame-arguments}
9238 to @code{scalars} (the default) or @code{none} avoids this computation,
9239 thus speeding up the display of each Ada frame.
9241 @item show print frame-arguments
9242 Show how the value of arguments should be displayed when printing a frame.
9244 @item set print raw frame-arguments on
9245 Print frame arguments in raw, non pretty-printed, form.
9247 @item set print raw frame-arguments off
9248 Print frame arguments in pretty-printed form, if there is a pretty-printer
9249 for the value (@pxref{Pretty Printing}),
9250 otherwise print the value in raw form.
9251 This is the default.
9253 @item show print raw frame-arguments
9254 Show whether to print frame arguments in raw form.
9256 @anchor{set print entry-values}
9257 @item set print entry-values @var{value}
9258 @kindex set print entry-values
9259 Set printing of frame argument values at function entry. In some cases
9260 @value{GDBN} can determine the value of function argument which was passed by
9261 the function caller, even if the value was modified inside the called function
9262 and therefore is different. With optimized code, the current value could be
9263 unavailable, but the entry value may still be known.
9265 The default value is @code{default} (see below for its description). Older
9266 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
9267 this feature will behave in the @code{default} setting the same way as with the
9270 This functionality is currently supported only by DWARF 2 debugging format and
9271 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
9272 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
9275 The @var{value} parameter can be one of the following:
9279 Print only actual parameter values, never print values from function entry
9283 #0 different (val=6)
9284 #0 lost (val=<optimized out>)
9286 #0 invalid (val=<optimized out>)
9290 Print only parameter values from function entry point. The actual parameter
9291 values are never printed.
9293 #0 equal (val@@entry=5)
9294 #0 different (val@@entry=5)
9295 #0 lost (val@@entry=5)
9296 #0 born (val@@entry=<optimized out>)
9297 #0 invalid (val@@entry=<optimized out>)
9301 Print only parameter values from function entry point. If value from function
9302 entry point is not known while the actual value is known, print the actual
9303 value for such parameter.
9305 #0 equal (val@@entry=5)
9306 #0 different (val@@entry=5)
9307 #0 lost (val@@entry=5)
9309 #0 invalid (val@@entry=<optimized out>)
9313 Print actual parameter values. If actual parameter value is not known while
9314 value from function entry point is known, print the entry point value for such
9318 #0 different (val=6)
9319 #0 lost (val@@entry=5)
9321 #0 invalid (val=<optimized out>)
9325 Always print both the actual parameter value and its value from function entry
9326 point, even if values of one or both are not available due to compiler
9329 #0 equal (val=5, val@@entry=5)
9330 #0 different (val=6, val@@entry=5)
9331 #0 lost (val=<optimized out>, val@@entry=5)
9332 #0 born (val=10, val@@entry=<optimized out>)
9333 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
9337 Print the actual parameter value if it is known and also its value from
9338 function entry point if it is known. If neither is known, print for the actual
9339 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
9340 values are known and identical, print the shortened
9341 @code{param=param@@entry=VALUE} notation.
9343 #0 equal (val=val@@entry=5)
9344 #0 different (val=6, val@@entry=5)
9345 #0 lost (val@@entry=5)
9347 #0 invalid (val=<optimized out>)
9351 Always print the actual parameter value. Print also its value from function
9352 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
9353 if both values are known and identical, print the shortened
9354 @code{param=param@@entry=VALUE} notation.
9356 #0 equal (val=val@@entry=5)
9357 #0 different (val=6, val@@entry=5)
9358 #0 lost (val=<optimized out>, val@@entry=5)
9360 #0 invalid (val=<optimized out>)
9364 For analysis messages on possible failures of frame argument values at function
9365 entry resolution see @ref{set debug entry-values}.
9367 @item show print entry-values
9368 Show the method being used for printing of frame argument values at function
9371 @item set print repeats @var{number-of-repeats}
9372 @itemx set print repeats unlimited
9373 @cindex repeated array elements
9374 Set the threshold for suppressing display of repeated array
9375 elements. When the number of consecutive identical elements of an
9376 array exceeds the threshold, @value{GDBN} prints the string
9377 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
9378 identical repetitions, instead of displaying the identical elements
9379 themselves. Setting the threshold to @code{unlimited} or zero will
9380 cause all elements to be individually printed. The default threshold
9383 @item show print repeats
9384 Display the current threshold for printing repeated identical
9387 @item set print null-stop
9388 @cindex @sc{null} elements in arrays
9389 Cause @value{GDBN} to stop printing the characters of an array when the first
9390 @sc{null} is encountered. This is useful when large arrays actually
9391 contain only short strings.
9394 @item show print null-stop
9395 Show whether @value{GDBN} stops printing an array on the first
9396 @sc{null} character.
9398 @item set print pretty on
9399 @cindex print structures in indented form
9400 @cindex indentation in structure display
9401 Cause @value{GDBN} to print structures in an indented format with one member
9402 per line, like this:
9417 @item set print pretty off
9418 Cause @value{GDBN} to print structures in a compact format, like this:
9422 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
9423 meat = 0x54 "Pork"@}
9428 This is the default format.
9430 @item show print pretty
9431 Show which format @value{GDBN} is using to print structures.
9433 @item set print sevenbit-strings on
9434 @cindex eight-bit characters in strings
9435 @cindex octal escapes in strings
9436 Print using only seven-bit characters; if this option is set,
9437 @value{GDBN} displays any eight-bit characters (in strings or
9438 character values) using the notation @code{\}@var{nnn}. This setting is
9439 best if you are working in English (@sc{ascii}) and you use the
9440 high-order bit of characters as a marker or ``meta'' bit.
9442 @item set print sevenbit-strings off
9443 Print full eight-bit characters. This allows the use of more
9444 international character sets, and is the default.
9446 @item show print sevenbit-strings
9447 Show whether or not @value{GDBN} is printing only seven-bit characters.
9449 @item set print union on
9450 @cindex unions in structures, printing
9451 Tell @value{GDBN} to print unions which are contained in structures
9452 and other unions. This is the default setting.
9454 @item set print union off
9455 Tell @value{GDBN} not to print unions which are contained in
9456 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
9459 @item show print union
9460 Ask @value{GDBN} whether or not it will print unions which are contained in
9461 structures and other unions.
9463 For example, given the declarations
9466 typedef enum @{Tree, Bug@} Species;
9467 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
9468 typedef enum @{Caterpillar, Cocoon, Butterfly@}
9479 struct thing foo = @{Tree, @{Acorn@}@};
9483 with @code{set print union on} in effect @samp{p foo} would print
9486 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
9490 and with @code{set print union off} in effect it would print
9493 $1 = @{it = Tree, form = @{...@}@}
9497 @code{set print union} affects programs written in C-like languages
9503 These settings are of interest when debugging C@t{++} programs:
9506 @cindex demangling C@t{++} names
9507 @item set print demangle
9508 @itemx set print demangle on
9509 Print C@t{++} names in their source form rather than in the encoded
9510 (``mangled'') form passed to the assembler and linker for type-safe
9511 linkage. The default is on.
9513 @item show print demangle
9514 Show whether C@t{++} names are printed in mangled or demangled form.
9516 @item set print asm-demangle
9517 @itemx set print asm-demangle on
9518 Print C@t{++} names in their source form rather than their mangled form, even
9519 in assembler code printouts such as instruction disassemblies.
9522 @item show print asm-demangle
9523 Show whether C@t{++} names in assembly listings are printed in mangled
9526 @cindex C@t{++} symbol decoding style
9527 @cindex symbol decoding style, C@t{++}
9528 @kindex set demangle-style
9529 @item set demangle-style @var{style}
9530 Choose among several encoding schemes used by different compilers to
9531 represent C@t{++} names. The choices for @var{style} are currently:
9535 Allow @value{GDBN} to choose a decoding style by inspecting your program.
9536 This is the default.
9539 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
9542 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
9545 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
9548 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
9549 @strong{Warning:} this setting alone is not sufficient to allow
9550 debugging @code{cfront}-generated executables. @value{GDBN} would
9551 require further enhancement to permit that.
9554 If you omit @var{style}, you will see a list of possible formats.
9556 @item show demangle-style
9557 Display the encoding style currently in use for decoding C@t{++} symbols.
9559 @item set print object
9560 @itemx set print object on
9561 @cindex derived type of an object, printing
9562 @cindex display derived types
9563 When displaying a pointer to an object, identify the @emph{actual}
9564 (derived) type of the object rather than the @emph{declared} type, using
9565 the virtual function table. Note that the virtual function table is
9566 required---this feature can only work for objects that have run-time
9567 type identification; a single virtual method in the object's declared
9568 type is sufficient. Note that this setting is also taken into account when
9569 working with variable objects via MI (@pxref{GDB/MI}).
9571 @item set print object off
9572 Display only the declared type of objects, without reference to the
9573 virtual function table. This is the default setting.
9575 @item show print object
9576 Show whether actual, or declared, object types are displayed.
9578 @item set print static-members
9579 @itemx set print static-members on
9580 @cindex static members of C@t{++} objects
9581 Print static members when displaying a C@t{++} object. The default is on.
9583 @item set print static-members off
9584 Do not print static members when displaying a C@t{++} object.
9586 @item show print static-members
9587 Show whether C@t{++} static members are printed or not.
9589 @item set print pascal_static-members
9590 @itemx set print pascal_static-members on
9591 @cindex static members of Pascal objects
9592 @cindex Pascal objects, static members display
9593 Print static members when displaying a Pascal object. The default is on.
9595 @item set print pascal_static-members off
9596 Do not print static members when displaying a Pascal object.
9598 @item show print pascal_static-members
9599 Show whether Pascal static members are printed or not.
9601 @c These don't work with HP ANSI C++ yet.
9602 @item set print vtbl
9603 @itemx set print vtbl on
9604 @cindex pretty print C@t{++} virtual function tables
9605 @cindex virtual functions (C@t{++}) display
9606 @cindex VTBL display
9607 Pretty print C@t{++} virtual function tables. The default is off.
9608 (The @code{vtbl} commands do not work on programs compiled with the HP
9609 ANSI C@t{++} compiler (@code{aCC}).)
9611 @item set print vtbl off
9612 Do not pretty print C@t{++} virtual function tables.
9614 @item show print vtbl
9615 Show whether C@t{++} virtual function tables are pretty printed, or not.
9618 @node Pretty Printing
9619 @section Pretty Printing
9621 @value{GDBN} provides a mechanism to allow pretty-printing of values using
9622 Python code. It greatly simplifies the display of complex objects. This
9623 mechanism works for both MI and the CLI.
9626 * Pretty-Printer Introduction:: Introduction to pretty-printers
9627 * Pretty-Printer Example:: An example pretty-printer
9628 * Pretty-Printer Commands:: Pretty-printer commands
9631 @node Pretty-Printer Introduction
9632 @subsection Pretty-Printer Introduction
9634 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
9635 registered for the value. If there is then @value{GDBN} invokes the
9636 pretty-printer to print the value. Otherwise the value is printed normally.
9638 Pretty-printers are normally named. This makes them easy to manage.
9639 The @samp{info pretty-printer} command will list all the installed
9640 pretty-printers with their names.
9641 If a pretty-printer can handle multiple data types, then its
9642 @dfn{subprinters} are the printers for the individual data types.
9643 Each such subprinter has its own name.
9644 The format of the name is @var{printer-name};@var{subprinter-name}.
9646 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
9647 Typically they are automatically loaded and registered when the corresponding
9648 debug information is loaded, thus making them available without having to
9649 do anything special.
9651 There are three places where a pretty-printer can be registered.
9655 Pretty-printers registered globally are available when debugging
9659 Pretty-printers registered with a program space are available only
9660 when debugging that program.
9661 @xref{Progspaces In Python}, for more details on program spaces in Python.
9664 Pretty-printers registered with an objfile are loaded and unloaded
9665 with the corresponding objfile (e.g., shared library).
9666 @xref{Objfiles In Python}, for more details on objfiles in Python.
9669 @xref{Selecting Pretty-Printers}, for further information on how
9670 pretty-printers are selected,
9672 @xref{Writing a Pretty-Printer}, for implementing pretty printers
9675 @node Pretty-Printer Example
9676 @subsection Pretty-Printer Example
9678 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
9681 (@value{GDBP}) print s
9683 static npos = 4294967295,
9685 <std::allocator<char>> = @{
9686 <__gnu_cxx::new_allocator<char>> = @{
9687 <No data fields>@}, <No data fields>
9689 members of std::basic_string<char, std::char_traits<char>,
9690 std::allocator<char> >::_Alloc_hider:
9691 _M_p = 0x804a014 "abcd"
9696 With a pretty-printer for @code{std::string} only the contents are printed:
9699 (@value{GDBP}) print s
9703 @node Pretty-Printer Commands
9704 @subsection Pretty-Printer Commands
9705 @cindex pretty-printer commands
9708 @kindex info pretty-printer
9709 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9710 Print the list of installed pretty-printers.
9711 This includes disabled pretty-printers, which are marked as such.
9713 @var{object-regexp} is a regular expression matching the objects
9714 whose pretty-printers to list.
9715 Objects can be @code{global}, the program space's file
9716 (@pxref{Progspaces In Python}),
9717 and the object files within that program space (@pxref{Objfiles In Python}).
9718 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
9719 looks up a printer from these three objects.
9721 @var{name-regexp} is a regular expression matching the name of the printers
9724 @kindex disable pretty-printer
9725 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9726 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
9727 A disabled pretty-printer is not forgotten, it may be enabled again later.
9729 @kindex enable pretty-printer
9730 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9731 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
9736 Suppose we have three pretty-printers installed: one from library1.so
9737 named @code{foo} that prints objects of type @code{foo}, and
9738 another from library2.so named @code{bar} that prints two types of objects,
9739 @code{bar1} and @code{bar2}.
9742 (gdb) info pretty-printer
9749 (gdb) info pretty-printer library2
9754 (gdb) disable pretty-printer library1
9756 2 of 3 printers enabled
9757 (gdb) info pretty-printer
9764 (gdb) disable pretty-printer library2 bar:bar1
9766 1 of 3 printers enabled
9767 (gdb) info pretty-printer library2
9774 (gdb) disable pretty-printer library2 bar
9776 0 of 3 printers enabled
9777 (gdb) info pretty-printer library2
9786 Note that for @code{bar} the entire printer can be disabled,
9787 as can each individual subprinter.
9790 @section Value History
9792 @cindex value history
9793 @cindex history of values printed by @value{GDBN}
9794 Values printed by the @code{print} command are saved in the @value{GDBN}
9795 @dfn{value history}. This allows you to refer to them in other expressions.
9796 Values are kept until the symbol table is re-read or discarded
9797 (for example with the @code{file} or @code{symbol-file} commands).
9798 When the symbol table changes, the value history is discarded,
9799 since the values may contain pointers back to the types defined in the
9804 @cindex history number
9805 The values printed are given @dfn{history numbers} by which you can
9806 refer to them. These are successive integers starting with one.
9807 @code{print} shows you the history number assigned to a value by
9808 printing @samp{$@var{num} = } before the value; here @var{num} is the
9811 To refer to any previous value, use @samp{$} followed by the value's
9812 history number. The way @code{print} labels its output is designed to
9813 remind you of this. Just @code{$} refers to the most recent value in
9814 the history, and @code{$$} refers to the value before that.
9815 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
9816 is the value just prior to @code{$$}, @code{$$1} is equivalent to
9817 @code{$$}, and @code{$$0} is equivalent to @code{$}.
9819 For example, suppose you have just printed a pointer to a structure and
9820 want to see the contents of the structure. It suffices to type
9826 If you have a chain of structures where the component @code{next} points
9827 to the next one, you can print the contents of the next one with this:
9834 You can print successive links in the chain by repeating this
9835 command---which you can do by just typing @key{RET}.
9837 Note that the history records values, not expressions. If the value of
9838 @code{x} is 4 and you type these commands:
9846 then the value recorded in the value history by the @code{print} command
9847 remains 4 even though the value of @code{x} has changed.
9852 Print the last ten values in the value history, with their item numbers.
9853 This is like @samp{p@ $$9} repeated ten times, except that @code{show
9854 values} does not change the history.
9856 @item show values @var{n}
9857 Print ten history values centered on history item number @var{n}.
9860 Print ten history values just after the values last printed. If no more
9861 values are available, @code{show values +} produces no display.
9864 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
9865 same effect as @samp{show values +}.
9867 @node Convenience Vars
9868 @section Convenience Variables
9870 @cindex convenience variables
9871 @cindex user-defined variables
9872 @value{GDBN} provides @dfn{convenience variables} that you can use within
9873 @value{GDBN} to hold on to a value and refer to it later. These variables
9874 exist entirely within @value{GDBN}; they are not part of your program, and
9875 setting a convenience variable has no direct effect on further execution
9876 of your program. That is why you can use them freely.
9878 Convenience variables are prefixed with @samp{$}. Any name preceded by
9879 @samp{$} can be used for a convenience variable, unless it is one of
9880 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
9881 (Value history references, in contrast, are @emph{numbers} preceded
9882 by @samp{$}. @xref{Value History, ,Value History}.)
9884 You can save a value in a convenience variable with an assignment
9885 expression, just as you would set a variable in your program.
9889 set $foo = *object_ptr
9893 would save in @code{$foo} the value contained in the object pointed to by
9896 Using a convenience variable for the first time creates it, but its
9897 value is @code{void} until you assign a new value. You can alter the
9898 value with another assignment at any time.
9900 Convenience variables have no fixed types. You can assign a convenience
9901 variable any type of value, including structures and arrays, even if
9902 that variable already has a value of a different type. The convenience
9903 variable, when used as an expression, has the type of its current value.
9906 @kindex show convenience
9907 @cindex show all user variables and functions
9908 @item show convenience
9909 Print a list of convenience variables used so far, and their values,
9910 as well as a list of the convenience functions.
9911 Abbreviated @code{show conv}.
9913 @kindex init-if-undefined
9914 @cindex convenience variables, initializing
9915 @item init-if-undefined $@var{variable} = @var{expression}
9916 Set a convenience variable if it has not already been set. This is useful
9917 for user-defined commands that keep some state. It is similar, in concept,
9918 to using local static variables with initializers in C (except that
9919 convenience variables are global). It can also be used to allow users to
9920 override default values used in a command script.
9922 If the variable is already defined then the expression is not evaluated so
9923 any side-effects do not occur.
9926 One of the ways to use a convenience variable is as a counter to be
9927 incremented or a pointer to be advanced. For example, to print
9928 a field from successive elements of an array of structures:
9932 print bar[$i++]->contents
9936 Repeat that command by typing @key{RET}.
9938 Some convenience variables are created automatically by @value{GDBN} and given
9939 values likely to be useful.
9942 @vindex $_@r{, convenience variable}
9944 The variable @code{$_} is automatically set by the @code{x} command to
9945 the last address examined (@pxref{Memory, ,Examining Memory}). Other
9946 commands which provide a default address for @code{x} to examine also
9947 set @code{$_} to that address; these commands include @code{info line}
9948 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
9949 except when set by the @code{x} command, in which case it is a pointer
9950 to the type of @code{$__}.
9952 @vindex $__@r{, convenience variable}
9954 The variable @code{$__} is automatically set by the @code{x} command
9955 to the value found in the last address examined. Its type is chosen
9956 to match the format in which the data was printed.
9959 @vindex $_exitcode@r{, convenience variable}
9960 When the program being debugged terminates normally, @value{GDBN}
9961 automatically sets this variable to the exit code of the program, and
9962 resets @code{$_exitsignal} to @code{void}.
9965 @vindex $_exitsignal@r{, convenience variable}
9966 When the program being debugged dies due to an uncaught signal,
9967 @value{GDBN} automatically sets this variable to that signal's number,
9968 and resets @code{$_exitcode} to @code{void}.
9970 To distinguish between whether the program being debugged has exited
9971 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
9972 @code{$_exitsignal} is not @code{void}), the convenience function
9973 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
9974 Functions}). For example, considering the following source code:
9980 main (int argc, char *argv[])
9987 A valid way of telling whether the program being debugged has exited
9988 or signalled would be:
9991 (@value{GDBP}) define has_exited_or_signalled
9992 Type commands for definition of ``has_exited_or_signalled''.
9993 End with a line saying just ``end''.
9994 >if $_isvoid ($_exitsignal)
9995 >echo The program has exited\n
9997 >echo The program has signalled\n
10003 Program terminated with signal SIGALRM, Alarm clock.
10004 The program no longer exists.
10005 (@value{GDBP}) has_exited_or_signalled
10006 The program has signalled
10009 As can be seen, @value{GDBN} correctly informs that the program being
10010 debugged has signalled, since it calls @code{raise} and raises a
10011 @code{SIGALRM} signal. If the program being debugged had not called
10012 @code{raise}, then @value{GDBN} would report a normal exit:
10015 (@value{GDBP}) has_exited_or_signalled
10016 The program has exited
10020 The variable @code{$_exception} is set to the exception object being
10021 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
10024 @itemx $_probe_arg0@dots{}$_probe_arg11
10025 Arguments to a static probe. @xref{Static Probe Points}.
10028 @vindex $_sdata@r{, inspect, convenience variable}
10029 The variable @code{$_sdata} contains extra collected static tracepoint
10030 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
10031 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
10032 if extra static tracepoint data has not been collected.
10035 @vindex $_siginfo@r{, convenience variable}
10036 The variable @code{$_siginfo} contains extra signal information
10037 (@pxref{extra signal information}). Note that @code{$_siginfo}
10038 could be empty, if the application has not yet received any signals.
10039 For example, it will be empty before you execute the @code{run} command.
10042 @vindex $_tlb@r{, convenience variable}
10043 The variable @code{$_tlb} is automatically set when debugging
10044 applications running on MS-Windows in native mode or connected to
10045 gdbserver that supports the @code{qGetTIBAddr} request.
10046 @xref{General Query Packets}.
10047 This variable contains the address of the thread information block.
10051 On HP-UX systems, if you refer to a function or variable name that
10052 begins with a dollar sign, @value{GDBN} searches for a user or system
10053 name first, before it searches for a convenience variable.
10055 @node Convenience Funs
10056 @section Convenience Functions
10058 @cindex convenience functions
10059 @value{GDBN} also supplies some @dfn{convenience functions}. These
10060 have a syntax similar to convenience variables. A convenience
10061 function can be used in an expression just like an ordinary function;
10062 however, a convenience function is implemented internally to
10065 These functions do not require @value{GDBN} to be configured with
10066 @code{Python} support, which means that they are always available.
10070 @item $_isvoid (@var{expr})
10071 @findex $_isvoid@r{, convenience function}
10072 Return one if the expression @var{expr} is @code{void}. Otherwise it
10075 A @code{void} expression is an expression where the type of the result
10076 is @code{void}. For example, you can examine a convenience variable
10077 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
10081 (@value{GDBP}) print $_exitcode
10083 (@value{GDBP}) print $_isvoid ($_exitcode)
10086 Starting program: ./a.out
10087 [Inferior 1 (process 29572) exited normally]
10088 (@value{GDBP}) print $_exitcode
10090 (@value{GDBP}) print $_isvoid ($_exitcode)
10094 In the example above, we used @code{$_isvoid} to check whether
10095 @code{$_exitcode} is @code{void} before and after the execution of the
10096 program being debugged. Before the execution there is no exit code to
10097 be examined, therefore @code{$_exitcode} is @code{void}. After the
10098 execution the program being debugged returned zero, therefore
10099 @code{$_exitcode} is zero, which means that it is not @code{void}
10102 The @code{void} expression can also be a call of a function from the
10103 program being debugged. For example, given the following function:
10112 The result of calling it inside @value{GDBN} is @code{void}:
10115 (@value{GDBP}) print foo ()
10117 (@value{GDBP}) print $_isvoid (foo ())
10119 (@value{GDBP}) set $v = foo ()
10120 (@value{GDBP}) print $v
10122 (@value{GDBP}) print $_isvoid ($v)
10128 These functions require @value{GDBN} to be configured with
10129 @code{Python} support.
10133 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
10134 @findex $_memeq@r{, convenience function}
10135 Returns one if the @var{length} bytes at the addresses given by
10136 @var{buf1} and @var{buf2} are equal.
10137 Otherwise it returns zero.
10139 @item $_regex(@var{str}, @var{regex})
10140 @findex $_regex@r{, convenience function}
10141 Returns one if the string @var{str} matches the regular expression
10142 @var{regex}. Otherwise it returns zero.
10143 The syntax of the regular expression is that specified by @code{Python}'s
10144 regular expression support.
10146 @item $_streq(@var{str1}, @var{str2})
10147 @findex $_streq@r{, convenience function}
10148 Returns one if the strings @var{str1} and @var{str2} are equal.
10149 Otherwise it returns zero.
10151 @item $_strlen(@var{str})
10152 @findex $_strlen@r{, convenience function}
10153 Returns the length of string @var{str}.
10155 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
10156 @findex $_caller_is@r{, convenience function}
10157 Returns one if the calling function's name is equal to @var{name}.
10158 Otherwise it returns zero.
10160 If the optional argument @var{number_of_frames} is provided,
10161 it is the number of frames up in the stack to look.
10169 at testsuite/gdb.python/py-caller-is.c:21
10170 #1 0x00000000004005a0 in middle_func ()
10171 at testsuite/gdb.python/py-caller-is.c:27
10172 #2 0x00000000004005ab in top_func ()
10173 at testsuite/gdb.python/py-caller-is.c:33
10174 #3 0x00000000004005b6 in main ()
10175 at testsuite/gdb.python/py-caller-is.c:39
10176 (gdb) print $_caller_is ("middle_func")
10178 (gdb) print $_caller_is ("top_func", 2)
10182 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
10183 @findex $_caller_matches@r{, convenience function}
10184 Returns one if the calling function's name matches the regular expression
10185 @var{regexp}. Otherwise it returns zero.
10187 If the optional argument @var{number_of_frames} is provided,
10188 it is the number of frames up in the stack to look.
10191 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
10192 @findex $_any_caller_is@r{, convenience function}
10193 Returns one if any calling function's name is equal to @var{name}.
10194 Otherwise it returns zero.
10196 If the optional argument @var{number_of_frames} is provided,
10197 it is the number of frames up in the stack to look.
10200 This function differs from @code{$_caller_is} in that this function
10201 checks all stack frames from the immediate caller to the frame specified
10202 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
10203 frame specified by @var{number_of_frames}.
10205 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
10206 @findex $_any_caller_matches@r{, convenience function}
10207 Returns one if any calling function's name matches the regular expression
10208 @var{regexp}. Otherwise it returns zero.
10210 If the optional argument @var{number_of_frames} is provided,
10211 it is the number of frames up in the stack to look.
10214 This function differs from @code{$_caller_matches} in that this function
10215 checks all stack frames from the immediate caller to the frame specified
10216 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
10217 frame specified by @var{number_of_frames}.
10221 @value{GDBN} provides the ability to list and get help on
10222 convenience functions.
10225 @item help function
10226 @kindex help function
10227 @cindex show all convenience functions
10228 Print a list of all convenience functions.
10235 You can refer to machine register contents, in expressions, as variables
10236 with names starting with @samp{$}. The names of registers are different
10237 for each machine; use @code{info registers} to see the names used on
10241 @kindex info registers
10242 @item info registers
10243 Print the names and values of all registers except floating-point
10244 and vector registers (in the selected stack frame).
10246 @kindex info all-registers
10247 @cindex floating point registers
10248 @item info all-registers
10249 Print the names and values of all registers, including floating-point
10250 and vector registers (in the selected stack frame).
10252 @item info registers @var{regname} @dots{}
10253 Print the @dfn{relativized} value of each specified register @var{regname}.
10254 As discussed in detail below, register values are normally relative to
10255 the selected stack frame. The @var{regname} may be any register name valid on
10256 the machine you are using, with or without the initial @samp{$}.
10259 @anchor{standard registers}
10260 @cindex stack pointer register
10261 @cindex program counter register
10262 @cindex process status register
10263 @cindex frame pointer register
10264 @cindex standard registers
10265 @value{GDBN} has four ``standard'' register names that are available (in
10266 expressions) on most machines---whenever they do not conflict with an
10267 architecture's canonical mnemonics for registers. The register names
10268 @code{$pc} and @code{$sp} are used for the program counter register and
10269 the stack pointer. @code{$fp} is used for a register that contains a
10270 pointer to the current stack frame, and @code{$ps} is used for a
10271 register that contains the processor status. For example,
10272 you could print the program counter in hex with
10279 or print the instruction to be executed next with
10286 or add four to the stack pointer@footnote{This is a way of removing
10287 one word from the stack, on machines where stacks grow downward in
10288 memory (most machines, nowadays). This assumes that the innermost
10289 stack frame is selected; setting @code{$sp} is not allowed when other
10290 stack frames are selected. To pop entire frames off the stack,
10291 regardless of machine architecture, use @code{return};
10292 see @ref{Returning, ,Returning from a Function}.} with
10298 Whenever possible, these four standard register names are available on
10299 your machine even though the machine has different canonical mnemonics,
10300 so long as there is no conflict. The @code{info registers} command
10301 shows the canonical names. For example, on the SPARC, @code{info
10302 registers} displays the processor status register as @code{$psr} but you
10303 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
10304 is an alias for the @sc{eflags} register.
10306 @value{GDBN} always considers the contents of an ordinary register as an
10307 integer when the register is examined in this way. Some machines have
10308 special registers which can hold nothing but floating point; these
10309 registers are considered to have floating point values. There is no way
10310 to refer to the contents of an ordinary register as floating point value
10311 (although you can @emph{print} it as a floating point value with
10312 @samp{print/f $@var{regname}}).
10314 Some registers have distinct ``raw'' and ``virtual'' data formats. This
10315 means that the data format in which the register contents are saved by
10316 the operating system is not the same one that your program normally
10317 sees. For example, the registers of the 68881 floating point
10318 coprocessor are always saved in ``extended'' (raw) format, but all C
10319 programs expect to work with ``double'' (virtual) format. In such
10320 cases, @value{GDBN} normally works with the virtual format only (the format
10321 that makes sense for your program), but the @code{info registers} command
10322 prints the data in both formats.
10324 @cindex SSE registers (x86)
10325 @cindex MMX registers (x86)
10326 Some machines have special registers whose contents can be interpreted
10327 in several different ways. For example, modern x86-based machines
10328 have SSE and MMX registers that can hold several values packed
10329 together in several different formats. @value{GDBN} refers to such
10330 registers in @code{struct} notation:
10333 (@value{GDBP}) print $xmm1
10335 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
10336 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
10337 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
10338 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
10339 v4_int32 = @{0, 20657912, 11, 13@},
10340 v2_int64 = @{88725056443645952, 55834574859@},
10341 uint128 = 0x0000000d0000000b013b36f800000000
10346 To set values of such registers, you need to tell @value{GDBN} which
10347 view of the register you wish to change, as if you were assigning
10348 value to a @code{struct} member:
10351 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
10354 Normally, register values are relative to the selected stack frame
10355 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
10356 value that the register would contain if all stack frames farther in
10357 were exited and their saved registers restored. In order to see the
10358 true contents of hardware registers, you must select the innermost
10359 frame (with @samp{frame 0}).
10361 @cindex caller-saved registers
10362 @cindex call-clobbered registers
10363 @cindex volatile registers
10364 @cindex <not saved> values
10365 Usually ABIs reserve some registers as not needed to be saved by the
10366 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
10367 registers). It may therefore not be possible for @value{GDBN} to know
10368 the value a register had before the call (in other words, in the outer
10369 frame), if the register value has since been changed by the callee.
10370 @value{GDBN} tries to deduce where the inner frame saved
10371 (``callee-saved'') registers, from the debug info, unwind info, or the
10372 machine code generated by your compiler. If some register is not
10373 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
10374 its own knowledge of the ABI, or because the debug/unwind info
10375 explicitly says the register's value is undefined), @value{GDBN}
10376 displays @w{@samp{<not saved>}} as the register's value. With targets
10377 that @value{GDBN} has no knowledge of the register saving convention,
10378 if a register was not saved by the callee, then its value and location
10379 in the outer frame are assumed to be the same of the inner frame.
10380 This is usually harmless, because if the register is call-clobbered,
10381 the caller either does not care what is in the register after the
10382 call, or has code to restore the value that it does care about. Note,
10383 however, that if you change such a register in the outer frame, you
10384 may also be affecting the inner frame. Also, the more ``outer'' the
10385 frame is you're looking at, the more likely a call-clobbered
10386 register's value is to be wrong, in the sense that it doesn't actually
10387 represent the value the register had just before the call.
10389 @node Floating Point Hardware
10390 @section Floating Point Hardware
10391 @cindex floating point
10393 Depending on the configuration, @value{GDBN} may be able to give
10394 you more information about the status of the floating point hardware.
10399 Display hardware-dependent information about the floating
10400 point unit. The exact contents and layout vary depending on the
10401 floating point chip. Currently, @samp{info float} is supported on
10402 the ARM and x86 machines.
10406 @section Vector Unit
10407 @cindex vector unit
10409 Depending on the configuration, @value{GDBN} may be able to give you
10410 more information about the status of the vector unit.
10413 @kindex info vector
10415 Display information about the vector unit. The exact contents and
10416 layout vary depending on the hardware.
10419 @node OS Information
10420 @section Operating System Auxiliary Information
10421 @cindex OS information
10423 @value{GDBN} provides interfaces to useful OS facilities that can help
10424 you debug your program.
10426 @cindex auxiliary vector
10427 @cindex vector, auxiliary
10428 Some operating systems supply an @dfn{auxiliary vector} to programs at
10429 startup. This is akin to the arguments and environment that you
10430 specify for a program, but contains a system-dependent variety of
10431 binary values that tell system libraries important details about the
10432 hardware, operating system, and process. Each value's purpose is
10433 identified by an integer tag; the meanings are well-known but system-specific.
10434 Depending on the configuration and operating system facilities,
10435 @value{GDBN} may be able to show you this information. For remote
10436 targets, this functionality may further depend on the remote stub's
10437 support of the @samp{qXfer:auxv:read} packet, see
10438 @ref{qXfer auxiliary vector read}.
10443 Display the auxiliary vector of the inferior, which can be either a
10444 live process or a core dump file. @value{GDBN} prints each tag value
10445 numerically, and also shows names and text descriptions for recognized
10446 tags. Some values in the vector are numbers, some bit masks, and some
10447 pointers to strings or other data. @value{GDBN} displays each value in the
10448 most appropriate form for a recognized tag, and in hexadecimal for
10449 an unrecognized tag.
10452 On some targets, @value{GDBN} can access operating system-specific
10453 information and show it to you. The types of information available
10454 will differ depending on the type of operating system running on the
10455 target. The mechanism used to fetch the data is described in
10456 @ref{Operating System Information}. For remote targets, this
10457 functionality depends on the remote stub's support of the
10458 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
10462 @item info os @var{infotype}
10464 Display OS information of the requested type.
10466 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
10468 @anchor{linux info os infotypes}
10470 @kindex info os processes
10472 Display the list of processes on the target. For each process,
10473 @value{GDBN} prints the process identifier, the name of the user, the
10474 command corresponding to the process, and the list of processor cores
10475 that the process is currently running on. (To understand what these
10476 properties mean, for this and the following info types, please consult
10477 the general @sc{gnu}/Linux documentation.)
10479 @kindex info os procgroups
10481 Display the list of process groups on the target. For each process,
10482 @value{GDBN} prints the identifier of the process group that it belongs
10483 to, the command corresponding to the process group leader, the process
10484 identifier, and the command line of the process. The list is sorted
10485 first by the process group identifier, then by the process identifier,
10486 so that processes belonging to the same process group are grouped together
10487 and the process group leader is listed first.
10489 @kindex info os threads
10491 Display the list of threads running on the target. For each thread,
10492 @value{GDBN} prints the identifier of the process that the thread
10493 belongs to, the command of the process, the thread identifier, and the
10494 processor core that it is currently running on. The main thread of a
10495 process is not listed.
10497 @kindex info os files
10499 Display the list of open file descriptors on the target. For each
10500 file descriptor, @value{GDBN} prints the identifier of the process
10501 owning the descriptor, the command of the owning process, the value
10502 of the descriptor, and the target of the descriptor.
10504 @kindex info os sockets
10506 Display the list of Internet-domain sockets on the target. For each
10507 socket, @value{GDBN} prints the address and port of the local and
10508 remote endpoints, the current state of the connection, the creator of
10509 the socket, the IP address family of the socket, and the type of the
10512 @kindex info os shm
10514 Display the list of all System V shared-memory regions on the target.
10515 For each shared-memory region, @value{GDBN} prints the region key,
10516 the shared-memory identifier, the access permissions, the size of the
10517 region, the process that created the region, the process that last
10518 attached to or detached from the region, the current number of live
10519 attaches to the region, and the times at which the region was last
10520 attached to, detach from, and changed.
10522 @kindex info os semaphores
10524 Display the list of all System V semaphore sets on the target. For each
10525 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
10526 set identifier, the access permissions, the number of semaphores in the
10527 set, the user and group of the owner and creator of the semaphore set,
10528 and the times at which the semaphore set was operated upon and changed.
10530 @kindex info os msg
10532 Display the list of all System V message queues on the target. For each
10533 message queue, @value{GDBN} prints the message queue key, the message
10534 queue identifier, the access permissions, the current number of bytes
10535 on the queue, the current number of messages on the queue, the processes
10536 that last sent and received a message on the queue, the user and group
10537 of the owner and creator of the message queue, the times at which a
10538 message was last sent and received on the queue, and the time at which
10539 the message queue was last changed.
10541 @kindex info os modules
10543 Display the list of all loaded kernel modules on the target. For each
10544 module, @value{GDBN} prints the module name, the size of the module in
10545 bytes, the number of times the module is used, the dependencies of the
10546 module, the status of the module, and the address of the loaded module
10551 If @var{infotype} is omitted, then list the possible values for
10552 @var{infotype} and the kind of OS information available for each
10553 @var{infotype}. If the target does not return a list of possible
10554 types, this command will report an error.
10557 @node Memory Region Attributes
10558 @section Memory Region Attributes
10559 @cindex memory region attributes
10561 @dfn{Memory region attributes} allow you to describe special handling
10562 required by regions of your target's memory. @value{GDBN} uses
10563 attributes to determine whether to allow certain types of memory
10564 accesses; whether to use specific width accesses; and whether to cache
10565 target memory. By default the description of memory regions is
10566 fetched from the target (if the current target supports this), but the
10567 user can override the fetched regions.
10569 Defined memory regions can be individually enabled and disabled. When a
10570 memory region is disabled, @value{GDBN} uses the default attributes when
10571 accessing memory in that region. Similarly, if no memory regions have
10572 been defined, @value{GDBN} uses the default attributes when accessing
10575 When a memory region is defined, it is given a number to identify it;
10576 to enable, disable, or remove a memory region, you specify that number.
10580 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
10581 Define a memory region bounded by @var{lower} and @var{upper} with
10582 attributes @var{attributes}@dots{}, and add it to the list of regions
10583 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
10584 case: it is treated as the target's maximum memory address.
10585 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
10588 Discard any user changes to the memory regions and use target-supplied
10589 regions, if available, or no regions if the target does not support.
10592 @item delete mem @var{nums}@dots{}
10593 Remove memory regions @var{nums}@dots{} from the list of regions
10594 monitored by @value{GDBN}.
10596 @kindex disable mem
10597 @item disable mem @var{nums}@dots{}
10598 Disable monitoring of memory regions @var{nums}@dots{}.
10599 A disabled memory region is not forgotten.
10600 It may be enabled again later.
10603 @item enable mem @var{nums}@dots{}
10604 Enable monitoring of memory regions @var{nums}@dots{}.
10608 Print a table of all defined memory regions, with the following columns
10612 @item Memory Region Number
10613 @item Enabled or Disabled.
10614 Enabled memory regions are marked with @samp{y}.
10615 Disabled memory regions are marked with @samp{n}.
10618 The address defining the inclusive lower bound of the memory region.
10621 The address defining the exclusive upper bound of the memory region.
10624 The list of attributes set for this memory region.
10629 @subsection Attributes
10631 @subsubsection Memory Access Mode
10632 The access mode attributes set whether @value{GDBN} may make read or
10633 write accesses to a memory region.
10635 While these attributes prevent @value{GDBN} from performing invalid
10636 memory accesses, they do nothing to prevent the target system, I/O DMA,
10637 etc.@: from accessing memory.
10641 Memory is read only.
10643 Memory is write only.
10645 Memory is read/write. This is the default.
10648 @subsubsection Memory Access Size
10649 The access size attribute tells @value{GDBN} to use specific sized
10650 accesses in the memory region. Often memory mapped device registers
10651 require specific sized accesses. If no access size attribute is
10652 specified, @value{GDBN} may use accesses of any size.
10656 Use 8 bit memory accesses.
10658 Use 16 bit memory accesses.
10660 Use 32 bit memory accesses.
10662 Use 64 bit memory accesses.
10665 @c @subsubsection Hardware/Software Breakpoints
10666 @c The hardware/software breakpoint attributes set whether @value{GDBN}
10667 @c will use hardware or software breakpoints for the internal breakpoints
10668 @c used by the step, next, finish, until, etc. commands.
10672 @c Always use hardware breakpoints
10673 @c @item swbreak (default)
10676 @subsubsection Data Cache
10677 The data cache attributes set whether @value{GDBN} will cache target
10678 memory. While this generally improves performance by reducing debug
10679 protocol overhead, it can lead to incorrect results because @value{GDBN}
10680 does not know about volatile variables or memory mapped device
10685 Enable @value{GDBN} to cache target memory.
10687 Disable @value{GDBN} from caching target memory. This is the default.
10690 @subsection Memory Access Checking
10691 @value{GDBN} can be instructed to refuse accesses to memory that is
10692 not explicitly described. This can be useful if accessing such
10693 regions has undesired effects for a specific target, or to provide
10694 better error checking. The following commands control this behaviour.
10697 @kindex set mem inaccessible-by-default
10698 @item set mem inaccessible-by-default [on|off]
10699 If @code{on} is specified, make @value{GDBN} treat memory not
10700 explicitly described by the memory ranges as non-existent and refuse accesses
10701 to such memory. The checks are only performed if there's at least one
10702 memory range defined. If @code{off} is specified, make @value{GDBN}
10703 treat the memory not explicitly described by the memory ranges as RAM.
10704 The default value is @code{on}.
10705 @kindex show mem inaccessible-by-default
10706 @item show mem inaccessible-by-default
10707 Show the current handling of accesses to unknown memory.
10711 @c @subsubsection Memory Write Verification
10712 @c The memory write verification attributes set whether @value{GDBN}
10713 @c will re-reads data after each write to verify the write was successful.
10717 @c @item noverify (default)
10720 @node Dump/Restore Files
10721 @section Copy Between Memory and a File
10722 @cindex dump/restore files
10723 @cindex append data to a file
10724 @cindex dump data to a file
10725 @cindex restore data from a file
10727 You can use the commands @code{dump}, @code{append}, and
10728 @code{restore} to copy data between target memory and a file. The
10729 @code{dump} and @code{append} commands write data to a file, and the
10730 @code{restore} command reads data from a file back into the inferior's
10731 memory. Files may be in binary, Motorola S-record, Intel hex, or
10732 Tektronix Hex format; however, @value{GDBN} can only append to binary
10738 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
10739 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
10740 Dump the contents of memory from @var{start_addr} to @var{end_addr},
10741 or the value of @var{expr}, to @var{filename} in the given format.
10743 The @var{format} parameter may be any one of:
10750 Motorola S-record format.
10752 Tektronix Hex format.
10755 @value{GDBN} uses the same definitions of these formats as the
10756 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
10757 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
10761 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
10762 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
10763 Append the contents of memory from @var{start_addr} to @var{end_addr},
10764 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
10765 (@value{GDBN} can only append data to files in raw binary form.)
10768 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
10769 Restore the contents of file @var{filename} into memory. The
10770 @code{restore} command can automatically recognize any known @sc{bfd}
10771 file format, except for raw binary. To restore a raw binary file you
10772 must specify the optional keyword @code{binary} after the filename.
10774 If @var{bias} is non-zero, its value will be added to the addresses
10775 contained in the file. Binary files always start at address zero, so
10776 they will be restored at address @var{bias}. Other bfd files have
10777 a built-in location; they will be restored at offset @var{bias}
10778 from that location.
10780 If @var{start} and/or @var{end} are non-zero, then only data between
10781 file offset @var{start} and file offset @var{end} will be restored.
10782 These offsets are relative to the addresses in the file, before
10783 the @var{bias} argument is applied.
10787 @node Core File Generation
10788 @section How to Produce a Core File from Your Program
10789 @cindex dump core from inferior
10791 A @dfn{core file} or @dfn{core dump} is a file that records the memory
10792 image of a running process and its process status (register values
10793 etc.). Its primary use is post-mortem debugging of a program that
10794 crashed while it ran outside a debugger. A program that crashes
10795 automatically produces a core file, unless this feature is disabled by
10796 the user. @xref{Files}, for information on invoking @value{GDBN} in
10797 the post-mortem debugging mode.
10799 Occasionally, you may wish to produce a core file of the program you
10800 are debugging in order to preserve a snapshot of its state.
10801 @value{GDBN} has a special command for that.
10805 @kindex generate-core-file
10806 @item generate-core-file [@var{file}]
10807 @itemx gcore [@var{file}]
10808 Produce a core dump of the inferior process. The optional argument
10809 @var{file} specifies the file name where to put the core dump. If not
10810 specified, the file name defaults to @file{core.@var{pid}}, where
10811 @var{pid} is the inferior process ID.
10813 Note that this command is implemented only for some systems (as of
10814 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
10817 @node Character Sets
10818 @section Character Sets
10819 @cindex character sets
10821 @cindex translating between character sets
10822 @cindex host character set
10823 @cindex target character set
10825 If the program you are debugging uses a different character set to
10826 represent characters and strings than the one @value{GDBN} uses itself,
10827 @value{GDBN} can automatically translate between the character sets for
10828 you. The character set @value{GDBN} uses we call the @dfn{host
10829 character set}; the one the inferior program uses we call the
10830 @dfn{target character set}.
10832 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
10833 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
10834 remote protocol (@pxref{Remote Debugging}) to debug a program
10835 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
10836 then the host character set is Latin-1, and the target character set is
10837 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
10838 target-charset EBCDIC-US}, then @value{GDBN} translates between
10839 @sc{ebcdic} and Latin 1 as you print character or string values, or use
10840 character and string literals in expressions.
10842 @value{GDBN} has no way to automatically recognize which character set
10843 the inferior program uses; you must tell it, using the @code{set
10844 target-charset} command, described below.
10846 Here are the commands for controlling @value{GDBN}'s character set
10850 @item set target-charset @var{charset}
10851 @kindex set target-charset
10852 Set the current target character set to @var{charset}. To display the
10853 list of supported target character sets, type
10854 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
10856 @item set host-charset @var{charset}
10857 @kindex set host-charset
10858 Set the current host character set to @var{charset}.
10860 By default, @value{GDBN} uses a host character set appropriate to the
10861 system it is running on; you can override that default using the
10862 @code{set host-charset} command. On some systems, @value{GDBN} cannot
10863 automatically determine the appropriate host character set. In this
10864 case, @value{GDBN} uses @samp{UTF-8}.
10866 @value{GDBN} can only use certain character sets as its host character
10867 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
10868 @value{GDBN} will list the host character sets it supports.
10870 @item set charset @var{charset}
10871 @kindex set charset
10872 Set the current host and target character sets to @var{charset}. As
10873 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
10874 @value{GDBN} will list the names of the character sets that can be used
10875 for both host and target.
10878 @kindex show charset
10879 Show the names of the current host and target character sets.
10881 @item show host-charset
10882 @kindex show host-charset
10883 Show the name of the current host character set.
10885 @item show target-charset
10886 @kindex show target-charset
10887 Show the name of the current target character set.
10889 @item set target-wide-charset @var{charset}
10890 @kindex set target-wide-charset
10891 Set the current target's wide character set to @var{charset}. This is
10892 the character set used by the target's @code{wchar_t} type. To
10893 display the list of supported wide character sets, type
10894 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
10896 @item show target-wide-charset
10897 @kindex show target-wide-charset
10898 Show the name of the current target's wide character set.
10901 Here is an example of @value{GDBN}'s character set support in action.
10902 Assume that the following source code has been placed in the file
10903 @file{charset-test.c}:
10909 = @{72, 101, 108, 108, 111, 44, 32, 119,
10910 111, 114, 108, 100, 33, 10, 0@};
10911 char ibm1047_hello[]
10912 = @{200, 133, 147, 147, 150, 107, 64, 166,
10913 150, 153, 147, 132, 90, 37, 0@};
10917 printf ("Hello, world!\n");
10921 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
10922 containing the string @samp{Hello, world!} followed by a newline,
10923 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
10925 We compile the program, and invoke the debugger on it:
10928 $ gcc -g charset-test.c -o charset-test
10929 $ gdb -nw charset-test
10930 GNU gdb 2001-12-19-cvs
10931 Copyright 2001 Free Software Foundation, Inc.
10936 We can use the @code{show charset} command to see what character sets
10937 @value{GDBN} is currently using to interpret and display characters and
10941 (@value{GDBP}) show charset
10942 The current host and target character set is `ISO-8859-1'.
10946 For the sake of printing this manual, let's use @sc{ascii} as our
10947 initial character set:
10949 (@value{GDBP}) set charset ASCII
10950 (@value{GDBP}) show charset
10951 The current host and target character set is `ASCII'.
10955 Let's assume that @sc{ascii} is indeed the correct character set for our
10956 host system --- in other words, let's assume that if @value{GDBN} prints
10957 characters using the @sc{ascii} character set, our terminal will display
10958 them properly. Since our current target character set is also
10959 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
10962 (@value{GDBP}) print ascii_hello
10963 $1 = 0x401698 "Hello, world!\n"
10964 (@value{GDBP}) print ascii_hello[0]
10969 @value{GDBN} uses the target character set for character and string
10970 literals you use in expressions:
10973 (@value{GDBP}) print '+'
10978 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
10981 @value{GDBN} relies on the user to tell it which character set the
10982 target program uses. If we print @code{ibm1047_hello} while our target
10983 character set is still @sc{ascii}, we get jibberish:
10986 (@value{GDBP}) print ibm1047_hello
10987 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
10988 (@value{GDBP}) print ibm1047_hello[0]
10993 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
10994 @value{GDBN} tells us the character sets it supports:
10997 (@value{GDBP}) set target-charset
10998 ASCII EBCDIC-US IBM1047 ISO-8859-1
10999 (@value{GDBP}) set target-charset
11002 We can select @sc{ibm1047} as our target character set, and examine the
11003 program's strings again. Now the @sc{ascii} string is wrong, but
11004 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
11005 target character set, @sc{ibm1047}, to the host character set,
11006 @sc{ascii}, and they display correctly:
11009 (@value{GDBP}) set target-charset IBM1047
11010 (@value{GDBP}) show charset
11011 The current host character set is `ASCII'.
11012 The current target character set is `IBM1047'.
11013 (@value{GDBP}) print ascii_hello
11014 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
11015 (@value{GDBP}) print ascii_hello[0]
11017 (@value{GDBP}) print ibm1047_hello
11018 $8 = 0x4016a8 "Hello, world!\n"
11019 (@value{GDBP}) print ibm1047_hello[0]
11024 As above, @value{GDBN} uses the target character set for character and
11025 string literals you use in expressions:
11028 (@value{GDBP}) print '+'
11033 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
11036 @node Caching Target Data
11037 @section Caching Data of Targets
11038 @cindex caching data of targets
11040 @value{GDBN} caches data exchanged between the debugger and a target.
11041 Each cache is associated with the address space of the inferior.
11042 @xref{Inferiors and Programs}, about inferior and address space.
11043 Such caching generally improves performance in remote debugging
11044 (@pxref{Remote Debugging}), because it reduces the overhead of the
11045 remote protocol by bundling memory reads and writes into large chunks.
11046 Unfortunately, simply caching everything would lead to incorrect results,
11047 since @value{GDBN} does not necessarily know anything about volatile
11048 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
11049 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
11051 Therefore, by default, @value{GDBN} only caches data
11052 known to be on the stack@footnote{In non-stop mode, it is moderately
11053 rare for a running thread to modify the stack of a stopped thread
11054 in a way that would interfere with a backtrace, and caching of
11055 stack reads provides a significant speed up of remote backtraces.} or
11056 in the code segment.
11057 Other regions of memory can be explicitly marked as
11058 cacheable; @pxref{Memory Region Attributes}.
11061 @kindex set remotecache
11062 @item set remotecache on
11063 @itemx set remotecache off
11064 This option no longer does anything; it exists for compatibility
11067 @kindex show remotecache
11068 @item show remotecache
11069 Show the current state of the obsolete remotecache flag.
11071 @kindex set stack-cache
11072 @item set stack-cache on
11073 @itemx set stack-cache off
11074 Enable or disable caching of stack accesses. When @code{on}, use
11075 caching. By default, this option is @code{on}.
11077 @kindex show stack-cache
11078 @item show stack-cache
11079 Show the current state of data caching for memory accesses.
11081 @kindex set code-cache
11082 @item set code-cache on
11083 @itemx set code-cache off
11084 Enable or disable caching of code segment accesses. When @code{on},
11085 use caching. By default, this option is @code{on}. This improves
11086 performance of disassembly in remote debugging.
11088 @kindex show code-cache
11089 @item show code-cache
11090 Show the current state of target memory cache for code segment
11093 @kindex info dcache
11094 @item info dcache @r{[}line@r{]}
11095 Print the information about the performance of data cache of the
11096 current inferior's address space. The information displayed
11097 includes the dcache width and depth, and for each cache line, its
11098 number, address, and how many times it was referenced. This
11099 command is useful for debugging the data cache operation.
11101 If a line number is specified, the contents of that line will be
11104 @item set dcache size @var{size}
11105 @cindex dcache size
11106 @kindex set dcache size
11107 Set maximum number of entries in dcache (dcache depth above).
11109 @item set dcache line-size @var{line-size}
11110 @cindex dcache line-size
11111 @kindex set dcache line-size
11112 Set number of bytes each dcache entry caches (dcache width above).
11113 Must be a power of 2.
11115 @item show dcache size
11116 @kindex show dcache size
11117 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
11119 @item show dcache line-size
11120 @kindex show dcache line-size
11121 Show default size of dcache lines.
11125 @node Searching Memory
11126 @section Search Memory
11127 @cindex searching memory
11129 Memory can be searched for a particular sequence of bytes with the
11130 @code{find} command.
11134 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
11135 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
11136 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
11137 etc. The search begins at address @var{start_addr} and continues for either
11138 @var{len} bytes or through to @var{end_addr} inclusive.
11141 @var{s} and @var{n} are optional parameters.
11142 They may be specified in either order, apart or together.
11145 @item @var{s}, search query size
11146 The size of each search query value.
11152 halfwords (two bytes)
11156 giant words (eight bytes)
11159 All values are interpreted in the current language.
11160 This means, for example, that if the current source language is C/C@t{++}
11161 then searching for the string ``hello'' includes the trailing '\0'.
11163 If the value size is not specified, it is taken from the
11164 value's type in the current language.
11165 This is useful when one wants to specify the search
11166 pattern as a mixture of types.
11167 Note that this means, for example, that in the case of C-like languages
11168 a search for an untyped 0x42 will search for @samp{(int) 0x42}
11169 which is typically four bytes.
11171 @item @var{n}, maximum number of finds
11172 The maximum number of matches to print. The default is to print all finds.
11175 You can use strings as search values. Quote them with double-quotes
11177 The string value is copied into the search pattern byte by byte,
11178 regardless of the endianness of the target and the size specification.
11180 The address of each match found is printed as well as a count of the
11181 number of matches found.
11183 The address of the last value found is stored in convenience variable
11185 A count of the number of matches is stored in @samp{$numfound}.
11187 For example, if stopped at the @code{printf} in this function:
11193 static char hello[] = "hello-hello";
11194 static struct @{ char c; short s; int i; @}
11195 __attribute__ ((packed)) mixed
11196 = @{ 'c', 0x1234, 0x87654321 @};
11197 printf ("%s\n", hello);
11202 you get during debugging:
11205 (gdb) find &hello[0], +sizeof(hello), "hello"
11206 0x804956d <hello.1620+6>
11208 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
11209 0x8049567 <hello.1620>
11210 0x804956d <hello.1620+6>
11212 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
11213 0x8049567 <hello.1620>
11215 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
11216 0x8049560 <mixed.1625>
11218 (gdb) print $numfound
11221 $2 = (void *) 0x8049560
11224 @node Optimized Code
11225 @chapter Debugging Optimized Code
11226 @cindex optimized code, debugging
11227 @cindex debugging optimized code
11229 Almost all compilers support optimization. With optimization
11230 disabled, the compiler generates assembly code that corresponds
11231 directly to your source code, in a simplistic way. As the compiler
11232 applies more powerful optimizations, the generated assembly code
11233 diverges from your original source code. With help from debugging
11234 information generated by the compiler, @value{GDBN} can map from
11235 the running program back to constructs from your original source.
11237 @value{GDBN} is more accurate with optimization disabled. If you
11238 can recompile without optimization, it is easier to follow the
11239 progress of your program during debugging. But, there are many cases
11240 where you may need to debug an optimized version.
11242 When you debug a program compiled with @samp{-g -O}, remember that the
11243 optimizer has rearranged your code; the debugger shows you what is
11244 really there. Do not be too surprised when the execution path does not
11245 exactly match your source file! An extreme example: if you define a
11246 variable, but never use it, @value{GDBN} never sees that
11247 variable---because the compiler optimizes it out of existence.
11249 Some things do not work as well with @samp{-g -O} as with just
11250 @samp{-g}, particularly on machines with instruction scheduling. If in
11251 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
11252 please report it to us as a bug (including a test case!).
11253 @xref{Variables}, for more information about debugging optimized code.
11256 * Inline Functions:: How @value{GDBN} presents inlining
11257 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
11260 @node Inline Functions
11261 @section Inline Functions
11262 @cindex inline functions, debugging
11264 @dfn{Inlining} is an optimization that inserts a copy of the function
11265 body directly at each call site, instead of jumping to a shared
11266 routine. @value{GDBN} displays inlined functions just like
11267 non-inlined functions. They appear in backtraces. You can view their
11268 arguments and local variables, step into them with @code{step}, skip
11269 them with @code{next}, and escape from them with @code{finish}.
11270 You can check whether a function was inlined by using the
11271 @code{info frame} command.
11273 For @value{GDBN} to support inlined functions, the compiler must
11274 record information about inlining in the debug information ---
11275 @value{NGCC} using the @sc{dwarf 2} format does this, and several
11276 other compilers do also. @value{GDBN} only supports inlined functions
11277 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
11278 do not emit two required attributes (@samp{DW_AT_call_file} and
11279 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
11280 function calls with earlier versions of @value{NGCC}. It instead
11281 displays the arguments and local variables of inlined functions as
11282 local variables in the caller.
11284 The body of an inlined function is directly included at its call site;
11285 unlike a non-inlined function, there are no instructions devoted to
11286 the call. @value{GDBN} still pretends that the call site and the
11287 start of the inlined function are different instructions. Stepping to
11288 the call site shows the call site, and then stepping again shows
11289 the first line of the inlined function, even though no additional
11290 instructions are executed.
11292 This makes source-level debugging much clearer; you can see both the
11293 context of the call and then the effect of the call. Only stepping by
11294 a single instruction using @code{stepi} or @code{nexti} does not do
11295 this; single instruction steps always show the inlined body.
11297 There are some ways that @value{GDBN} does not pretend that inlined
11298 function calls are the same as normal calls:
11302 Setting breakpoints at the call site of an inlined function may not
11303 work, because the call site does not contain any code. @value{GDBN}
11304 may incorrectly move the breakpoint to the next line of the enclosing
11305 function, after the call. This limitation will be removed in a future
11306 version of @value{GDBN}; until then, set a breakpoint on an earlier line
11307 or inside the inlined function instead.
11310 @value{GDBN} cannot locate the return value of inlined calls after
11311 using the @code{finish} command. This is a limitation of compiler-generated
11312 debugging information; after @code{finish}, you can step to the next line
11313 and print a variable where your program stored the return value.
11317 @node Tail Call Frames
11318 @section Tail Call Frames
11319 @cindex tail call frames, debugging
11321 Function @code{B} can call function @code{C} in its very last statement. In
11322 unoptimized compilation the call of @code{C} is immediately followed by return
11323 instruction at the end of @code{B} code. Optimizing compiler may replace the
11324 call and return in function @code{B} into one jump to function @code{C}
11325 instead. Such use of a jump instruction is called @dfn{tail call}.
11327 During execution of function @code{C}, there will be no indication in the
11328 function call stack frames that it was tail-called from @code{B}. If function
11329 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
11330 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
11331 some cases @value{GDBN} can determine that @code{C} was tail-called from
11332 @code{B}, and it will then create fictitious call frame for that, with the
11333 return address set up as if @code{B} called @code{C} normally.
11335 This functionality is currently supported only by DWARF 2 debugging format and
11336 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
11337 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
11340 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
11341 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
11345 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
11347 Stack level 1, frame at 0x7fffffffda30:
11348 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
11349 tail call frame, caller of frame at 0x7fffffffda30
11350 source language c++.
11351 Arglist at unknown address.
11352 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
11355 The detection of all the possible code path executions can find them ambiguous.
11356 There is no execution history stored (possible @ref{Reverse Execution} is never
11357 used for this purpose) and the last known caller could have reached the known
11358 callee by multiple different jump sequences. In such case @value{GDBN} still
11359 tries to show at least all the unambiguous top tail callers and all the
11360 unambiguous bottom tail calees, if any.
11363 @anchor{set debug entry-values}
11364 @item set debug entry-values
11365 @kindex set debug entry-values
11366 When set to on, enables printing of analysis messages for both frame argument
11367 values at function entry and tail calls. It will show all the possible valid
11368 tail calls code paths it has considered. It will also print the intersection
11369 of them with the final unambiguous (possibly partial or even empty) code path
11372 @item show debug entry-values
11373 @kindex show debug entry-values
11374 Show the current state of analysis messages printing for both frame argument
11375 values at function entry and tail calls.
11378 The analysis messages for tail calls can for example show why the virtual tail
11379 call frame for function @code{c} has not been recognized (due to the indirect
11380 reference by variable @code{x}):
11383 static void __attribute__((noinline, noclone)) c (void);
11384 void (*x) (void) = c;
11385 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
11386 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
11387 int main (void) @{ x (); return 0; @}
11389 Breakpoint 1, DW_OP_GNU_entry_value resolving cannot find
11390 DW_TAG_GNU_call_site 0x40039a in main
11392 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
11395 #1 0x000000000040039a in main () at t.c:5
11398 Another possibility is an ambiguous virtual tail call frames resolution:
11402 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
11403 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
11404 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
11405 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
11406 static void __attribute__((noinline, noclone)) b (void)
11407 @{ if (i) c (); else e (); @}
11408 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
11409 int main (void) @{ a (); return 0; @}
11411 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
11412 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
11413 tailcall: reduced: 0x4004d2(a) |
11416 #1 0x00000000004004d2 in a () at t.c:8
11417 #2 0x0000000000400395 in main () at t.c:9
11420 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
11421 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
11423 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
11424 @ifset HAVE_MAKEINFO_CLICK
11425 @set ARROW @click{}
11426 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
11427 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
11429 @ifclear HAVE_MAKEINFO_CLICK
11431 @set CALLSEQ1B @value{CALLSEQ1A}
11432 @set CALLSEQ2B @value{CALLSEQ2A}
11435 Frames #0 and #2 are real, #1 is a virtual tail call frame.
11436 The code can have possible execution paths @value{CALLSEQ1B} or
11437 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
11439 @code{initial:} state shows some random possible calling sequence @value{GDBN}
11440 has found. It then finds another possible calling sequcen - that one is
11441 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
11442 printed as the @code{reduced:} calling sequence. That one could have many
11443 futher @code{compare:} and @code{reduced:} statements as long as there remain
11444 any non-ambiguous sequence entries.
11446 For the frame of function @code{b} in both cases there are different possible
11447 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
11448 also ambigous. The only non-ambiguous frame is the one for function @code{a},
11449 therefore this one is displayed to the user while the ambiguous frames are
11452 There can be also reasons why printing of frame argument values at function
11457 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
11458 static void __attribute__((noinline, noclone)) a (int i);
11459 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
11460 static void __attribute__((noinline, noclone)) a (int i)
11461 @{ if (i) b (i - 1); else c (0); @}
11462 int main (void) @{ a (5); return 0; @}
11465 #0 c (i=i@@entry=0) at t.c:2
11466 #1 0x0000000000400428 in a (DW_OP_GNU_entry_value resolving has found
11467 function "a" at 0x400420 can call itself via tail calls
11468 i=<optimized out>) at t.c:6
11469 #2 0x000000000040036e in main () at t.c:7
11472 @value{GDBN} cannot find out from the inferior state if and how many times did
11473 function @code{a} call itself (via function @code{b}) as these calls would be
11474 tail calls. Such tail calls would modify thue @code{i} variable, therefore
11475 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
11476 prints @code{<optimized out>} instead.
11479 @chapter C Preprocessor Macros
11481 Some languages, such as C and C@t{++}, provide a way to define and invoke
11482 ``preprocessor macros'' which expand into strings of tokens.
11483 @value{GDBN} can evaluate expressions containing macro invocations, show
11484 the result of macro expansion, and show a macro's definition, including
11485 where it was defined.
11487 You may need to compile your program specially to provide @value{GDBN}
11488 with information about preprocessor macros. Most compilers do not
11489 include macros in their debugging information, even when you compile
11490 with the @option{-g} flag. @xref{Compilation}.
11492 A program may define a macro at one point, remove that definition later,
11493 and then provide a different definition after that. Thus, at different
11494 points in the program, a macro may have different definitions, or have
11495 no definition at all. If there is a current stack frame, @value{GDBN}
11496 uses the macros in scope at that frame's source code line. Otherwise,
11497 @value{GDBN} uses the macros in scope at the current listing location;
11500 Whenever @value{GDBN} evaluates an expression, it always expands any
11501 macro invocations present in the expression. @value{GDBN} also provides
11502 the following commands for working with macros explicitly.
11506 @kindex macro expand
11507 @cindex macro expansion, showing the results of preprocessor
11508 @cindex preprocessor macro expansion, showing the results of
11509 @cindex expanding preprocessor macros
11510 @item macro expand @var{expression}
11511 @itemx macro exp @var{expression}
11512 Show the results of expanding all preprocessor macro invocations in
11513 @var{expression}. Since @value{GDBN} simply expands macros, but does
11514 not parse the result, @var{expression} need not be a valid expression;
11515 it can be any string of tokens.
11518 @item macro expand-once @var{expression}
11519 @itemx macro exp1 @var{expression}
11520 @cindex expand macro once
11521 @i{(This command is not yet implemented.)} Show the results of
11522 expanding those preprocessor macro invocations that appear explicitly in
11523 @var{expression}. Macro invocations appearing in that expansion are
11524 left unchanged. This command allows you to see the effect of a
11525 particular macro more clearly, without being confused by further
11526 expansions. Since @value{GDBN} simply expands macros, but does not
11527 parse the result, @var{expression} need not be a valid expression; it
11528 can be any string of tokens.
11531 @cindex macro definition, showing
11532 @cindex definition of a macro, showing
11533 @cindex macros, from debug info
11534 @item info macro [-a|-all] [--] @var{macro}
11535 Show the current definition or all definitions of the named @var{macro},
11536 and describe the source location or compiler command-line where that
11537 definition was established. The optional double dash is to signify the end of
11538 argument processing and the beginning of @var{macro} for non C-like macros where
11539 the macro may begin with a hyphen.
11541 @kindex info macros
11542 @item info macros @var{linespec}
11543 Show all macro definitions that are in effect at the location specified
11544 by @var{linespec}, and describe the source location or compiler
11545 command-line where those definitions were established.
11547 @kindex macro define
11548 @cindex user-defined macros
11549 @cindex defining macros interactively
11550 @cindex macros, user-defined
11551 @item macro define @var{macro} @var{replacement-list}
11552 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
11553 Introduce a definition for a preprocessor macro named @var{macro},
11554 invocations of which are replaced by the tokens given in
11555 @var{replacement-list}. The first form of this command defines an
11556 ``object-like'' macro, which takes no arguments; the second form
11557 defines a ``function-like'' macro, which takes the arguments given in
11560 A definition introduced by this command is in scope in every
11561 expression evaluated in @value{GDBN}, until it is removed with the
11562 @code{macro undef} command, described below. The definition overrides
11563 all definitions for @var{macro} present in the program being debugged,
11564 as well as any previous user-supplied definition.
11566 @kindex macro undef
11567 @item macro undef @var{macro}
11568 Remove any user-supplied definition for the macro named @var{macro}.
11569 This command only affects definitions provided with the @code{macro
11570 define} command, described above; it cannot remove definitions present
11571 in the program being debugged.
11575 List all the macros defined using the @code{macro define} command.
11578 @cindex macros, example of debugging with
11579 Here is a transcript showing the above commands in action. First, we
11580 show our source files:
11585 #include "sample.h"
11588 #define ADD(x) (M + x)
11593 printf ("Hello, world!\n");
11595 printf ("We're so creative.\n");
11597 printf ("Goodbye, world!\n");
11604 Now, we compile the program using the @sc{gnu} C compiler,
11605 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
11606 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
11607 and @option{-gdwarf-4}; we recommend always choosing the most recent
11608 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
11609 includes information about preprocessor macros in the debugging
11613 $ gcc -gdwarf-2 -g3 sample.c -o sample
11617 Now, we start @value{GDBN} on our sample program:
11621 GNU gdb 2002-05-06-cvs
11622 Copyright 2002 Free Software Foundation, Inc.
11623 GDB is free software, @dots{}
11627 We can expand macros and examine their definitions, even when the
11628 program is not running. @value{GDBN} uses the current listing position
11629 to decide which macro definitions are in scope:
11632 (@value{GDBP}) list main
11635 5 #define ADD(x) (M + x)
11640 10 printf ("Hello, world!\n");
11642 12 printf ("We're so creative.\n");
11643 (@value{GDBP}) info macro ADD
11644 Defined at /home/jimb/gdb/macros/play/sample.c:5
11645 #define ADD(x) (M + x)
11646 (@value{GDBP}) info macro Q
11647 Defined at /home/jimb/gdb/macros/play/sample.h:1
11648 included at /home/jimb/gdb/macros/play/sample.c:2
11650 (@value{GDBP}) macro expand ADD(1)
11651 expands to: (42 + 1)
11652 (@value{GDBP}) macro expand-once ADD(1)
11653 expands to: once (M + 1)
11657 In the example above, note that @code{macro expand-once} expands only
11658 the macro invocation explicit in the original text --- the invocation of
11659 @code{ADD} --- but does not expand the invocation of the macro @code{M},
11660 which was introduced by @code{ADD}.
11662 Once the program is running, @value{GDBN} uses the macro definitions in
11663 force at the source line of the current stack frame:
11666 (@value{GDBP}) break main
11667 Breakpoint 1 at 0x8048370: file sample.c, line 10.
11669 Starting program: /home/jimb/gdb/macros/play/sample
11671 Breakpoint 1, main () at sample.c:10
11672 10 printf ("Hello, world!\n");
11676 At line 10, the definition of the macro @code{N} at line 9 is in force:
11679 (@value{GDBP}) info macro N
11680 Defined at /home/jimb/gdb/macros/play/sample.c:9
11682 (@value{GDBP}) macro expand N Q M
11683 expands to: 28 < 42
11684 (@value{GDBP}) print N Q M
11689 As we step over directives that remove @code{N}'s definition, and then
11690 give it a new definition, @value{GDBN} finds the definition (or lack
11691 thereof) in force at each point:
11694 (@value{GDBP}) next
11696 12 printf ("We're so creative.\n");
11697 (@value{GDBP}) info macro N
11698 The symbol `N' has no definition as a C/C++ preprocessor macro
11699 at /home/jimb/gdb/macros/play/sample.c:12
11700 (@value{GDBP}) next
11702 14 printf ("Goodbye, world!\n");
11703 (@value{GDBP}) info macro N
11704 Defined at /home/jimb/gdb/macros/play/sample.c:13
11706 (@value{GDBP}) macro expand N Q M
11707 expands to: 1729 < 42
11708 (@value{GDBP}) print N Q M
11713 In addition to source files, macros can be defined on the compilation command
11714 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
11715 such a way, @value{GDBN} displays the location of their definition as line zero
11716 of the source file submitted to the compiler.
11719 (@value{GDBP}) info macro __STDC__
11720 Defined at /home/jimb/gdb/macros/play/sample.c:0
11727 @chapter Tracepoints
11728 @c This chapter is based on the documentation written by Michael
11729 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
11731 @cindex tracepoints
11732 In some applications, it is not feasible for the debugger to interrupt
11733 the program's execution long enough for the developer to learn
11734 anything helpful about its behavior. If the program's correctness
11735 depends on its real-time behavior, delays introduced by a debugger
11736 might cause the program to change its behavior drastically, or perhaps
11737 fail, even when the code itself is correct. It is useful to be able
11738 to observe the program's behavior without interrupting it.
11740 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
11741 specify locations in the program, called @dfn{tracepoints}, and
11742 arbitrary expressions to evaluate when those tracepoints are reached.
11743 Later, using the @code{tfind} command, you can examine the values
11744 those expressions had when the program hit the tracepoints. The
11745 expressions may also denote objects in memory---structures or arrays,
11746 for example---whose values @value{GDBN} should record; while visiting
11747 a particular tracepoint, you may inspect those objects as if they were
11748 in memory at that moment. However, because @value{GDBN} records these
11749 values without interacting with you, it can do so quickly and
11750 unobtrusively, hopefully not disturbing the program's behavior.
11752 The tracepoint facility is currently available only for remote
11753 targets. @xref{Targets}. In addition, your remote target must know
11754 how to collect trace data. This functionality is implemented in the
11755 remote stub; however, none of the stubs distributed with @value{GDBN}
11756 support tracepoints as of this writing. The format of the remote
11757 packets used to implement tracepoints are described in @ref{Tracepoint
11760 It is also possible to get trace data from a file, in a manner reminiscent
11761 of corefiles; you specify the filename, and use @code{tfind} to search
11762 through the file. @xref{Trace Files}, for more details.
11764 This chapter describes the tracepoint commands and features.
11767 * Set Tracepoints::
11768 * Analyze Collected Data::
11769 * Tracepoint Variables::
11773 @node Set Tracepoints
11774 @section Commands to Set Tracepoints
11776 Before running such a @dfn{trace experiment}, an arbitrary number of
11777 tracepoints can be set. A tracepoint is actually a special type of
11778 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
11779 standard breakpoint commands. For instance, as with breakpoints,
11780 tracepoint numbers are successive integers starting from one, and many
11781 of the commands associated with tracepoints take the tracepoint number
11782 as their argument, to identify which tracepoint to work on.
11784 For each tracepoint, you can specify, in advance, some arbitrary set
11785 of data that you want the target to collect in the trace buffer when
11786 it hits that tracepoint. The collected data can include registers,
11787 local variables, or global data. Later, you can use @value{GDBN}
11788 commands to examine the values these data had at the time the
11789 tracepoint was hit.
11791 Tracepoints do not support every breakpoint feature. Ignore counts on
11792 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
11793 commands when they are hit. Tracepoints may not be thread-specific
11796 @cindex fast tracepoints
11797 Some targets may support @dfn{fast tracepoints}, which are inserted in
11798 a different way (such as with a jump instead of a trap), that is
11799 faster but possibly restricted in where they may be installed.
11801 @cindex static tracepoints
11802 @cindex markers, static tracepoints
11803 @cindex probing markers, static tracepoints
11804 Regular and fast tracepoints are dynamic tracing facilities, meaning
11805 that they can be used to insert tracepoints at (almost) any location
11806 in the target. Some targets may also support controlling @dfn{static
11807 tracepoints} from @value{GDBN}. With static tracing, a set of
11808 instrumentation points, also known as @dfn{markers}, are embedded in
11809 the target program, and can be activated or deactivated by name or
11810 address. These are usually placed at locations which facilitate
11811 investigating what the target is actually doing. @value{GDBN}'s
11812 support for static tracing includes being able to list instrumentation
11813 points, and attach them with @value{GDBN} defined high level
11814 tracepoints that expose the whole range of convenience of
11815 @value{GDBN}'s tracepoints support. Namely, support for collecting
11816 registers values and values of global or local (to the instrumentation
11817 point) variables; tracepoint conditions and trace state variables.
11818 The act of installing a @value{GDBN} static tracepoint on an
11819 instrumentation point, or marker, is referred to as @dfn{probing} a
11820 static tracepoint marker.
11822 @code{gdbserver} supports tracepoints on some target systems.
11823 @xref{Server,,Tracepoints support in @code{gdbserver}}.
11825 This section describes commands to set tracepoints and associated
11826 conditions and actions.
11829 * Create and Delete Tracepoints::
11830 * Enable and Disable Tracepoints::
11831 * Tracepoint Passcounts::
11832 * Tracepoint Conditions::
11833 * Trace State Variables::
11834 * Tracepoint Actions::
11835 * Listing Tracepoints::
11836 * Listing Static Tracepoint Markers::
11837 * Starting and Stopping Trace Experiments::
11838 * Tracepoint Restrictions::
11841 @node Create and Delete Tracepoints
11842 @subsection Create and Delete Tracepoints
11845 @cindex set tracepoint
11847 @item trace @var{location}
11848 The @code{trace} command is very similar to the @code{break} command.
11849 Its argument @var{location} can be a source line, a function name, or
11850 an address in the target program. @xref{Specify Location}. The
11851 @code{trace} command defines a tracepoint, which is a point in the
11852 target program where the debugger will briefly stop, collect some
11853 data, and then allow the program to continue. Setting a tracepoint or
11854 changing its actions takes effect immediately if the remote stub
11855 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
11857 If remote stub doesn't support the @samp{InstallInTrace} feature, all
11858 these changes don't take effect until the next @code{tstart}
11859 command, and once a trace experiment is running, further changes will
11860 not have any effect until the next trace experiment starts. In addition,
11861 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
11862 address is not yet resolved. (This is similar to pending breakpoints.)
11863 Pending tracepoints are not downloaded to the target and not installed
11864 until they are resolved. The resolution of pending tracepoints requires
11865 @value{GDBN} support---when debugging with the remote target, and
11866 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
11867 tracing}), pending tracepoints can not be resolved (and downloaded to
11868 the remote stub) while @value{GDBN} is disconnected.
11870 Here are some examples of using the @code{trace} command:
11873 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
11875 (@value{GDBP}) @b{trace +2} // 2 lines forward
11877 (@value{GDBP}) @b{trace my_function} // first source line of function
11879 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
11881 (@value{GDBP}) @b{trace *0x2117c4} // an address
11885 You can abbreviate @code{trace} as @code{tr}.
11887 @item trace @var{location} if @var{cond}
11888 Set a tracepoint with condition @var{cond}; evaluate the expression
11889 @var{cond} each time the tracepoint is reached, and collect data only
11890 if the value is nonzero---that is, if @var{cond} evaluates as true.
11891 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
11892 information on tracepoint conditions.
11894 @item ftrace @var{location} [ if @var{cond} ]
11895 @cindex set fast tracepoint
11896 @cindex fast tracepoints, setting
11898 The @code{ftrace} command sets a fast tracepoint. For targets that
11899 support them, fast tracepoints will use a more efficient but possibly
11900 less general technique to trigger data collection, such as a jump
11901 instruction instead of a trap, or some sort of hardware support. It
11902 may not be possible to create a fast tracepoint at the desired
11903 location, in which case the command will exit with an explanatory
11906 @value{GDBN} handles arguments to @code{ftrace} exactly as for
11909 On 32-bit x86-architecture systems, fast tracepoints normally need to
11910 be placed at an instruction that is 5 bytes or longer, but can be
11911 placed at 4-byte instructions if the low 64K of memory of the target
11912 program is available to install trampolines. Some Unix-type systems,
11913 such as @sc{gnu}/Linux, exclude low addresses from the program's
11914 address space; but for instance with the Linux kernel it is possible
11915 to let @value{GDBN} use this area by doing a @command{sysctl} command
11916 to set the @code{mmap_min_addr} kernel parameter, as in
11919 sudo sysctl -w vm.mmap_min_addr=32768
11923 which sets the low address to 32K, which leaves plenty of room for
11924 trampolines. The minimum address should be set to a page boundary.
11926 @item strace @var{location} [ if @var{cond} ]
11927 @cindex set static tracepoint
11928 @cindex static tracepoints, setting
11929 @cindex probe static tracepoint marker
11931 The @code{strace} command sets a static tracepoint. For targets that
11932 support it, setting a static tracepoint probes a static
11933 instrumentation point, or marker, found at @var{location}. It may not
11934 be possible to set a static tracepoint at the desired location, in
11935 which case the command will exit with an explanatory message.
11937 @value{GDBN} handles arguments to @code{strace} exactly as for
11938 @code{trace}, with the addition that the user can also specify
11939 @code{-m @var{marker}} as @var{location}. This probes the marker
11940 identified by the @var{marker} string identifier. This identifier
11941 depends on the static tracepoint backend library your program is
11942 using. You can find all the marker identifiers in the @samp{ID} field
11943 of the @code{info static-tracepoint-markers} command output.
11944 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
11945 Markers}. For example, in the following small program using the UST
11951 trace_mark(ust, bar33, "str %s", "FOOBAZ");
11956 the marker id is composed of joining the first two arguments to the
11957 @code{trace_mark} call with a slash, which translates to:
11960 (@value{GDBP}) info static-tracepoint-markers
11961 Cnt Enb ID Address What
11962 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
11968 so you may probe the marker above with:
11971 (@value{GDBP}) strace -m ust/bar33
11974 Static tracepoints accept an extra collect action --- @code{collect
11975 $_sdata}. This collects arbitrary user data passed in the probe point
11976 call to the tracing library. In the UST example above, you'll see
11977 that the third argument to @code{trace_mark} is a printf-like format
11978 string. The user data is then the result of running that formating
11979 string against the following arguments. Note that @code{info
11980 static-tracepoint-markers} command output lists that format string in
11981 the @samp{Data:} field.
11983 You can inspect this data when analyzing the trace buffer, by printing
11984 the $_sdata variable like any other variable available to
11985 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
11988 @cindex last tracepoint number
11989 @cindex recent tracepoint number
11990 @cindex tracepoint number
11991 The convenience variable @code{$tpnum} records the tracepoint number
11992 of the most recently set tracepoint.
11994 @kindex delete tracepoint
11995 @cindex tracepoint deletion
11996 @item delete tracepoint @r{[}@var{num}@r{]}
11997 Permanently delete one or more tracepoints. With no argument, the
11998 default is to delete all tracepoints. Note that the regular
11999 @code{delete} command can remove tracepoints also.
12004 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
12006 (@value{GDBP}) @b{delete trace} // remove all tracepoints
12010 You can abbreviate this command as @code{del tr}.
12013 @node Enable and Disable Tracepoints
12014 @subsection Enable and Disable Tracepoints
12016 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
12019 @kindex disable tracepoint
12020 @item disable tracepoint @r{[}@var{num}@r{]}
12021 Disable tracepoint @var{num}, or all tracepoints if no argument
12022 @var{num} is given. A disabled tracepoint will have no effect during
12023 a trace experiment, but it is not forgotten. You can re-enable
12024 a disabled tracepoint using the @code{enable tracepoint} command.
12025 If the command is issued during a trace experiment and the debug target
12026 has support for disabling tracepoints during a trace experiment, then the
12027 change will be effective immediately. Otherwise, it will be applied to the
12028 next trace experiment.
12030 @kindex enable tracepoint
12031 @item enable tracepoint @r{[}@var{num}@r{]}
12032 Enable tracepoint @var{num}, or all tracepoints. If this command is
12033 issued during a trace experiment and the debug target supports enabling
12034 tracepoints during a trace experiment, then the enabled tracepoints will
12035 become effective immediately. Otherwise, they will become effective the
12036 next time a trace experiment is run.
12039 @node Tracepoint Passcounts
12040 @subsection Tracepoint Passcounts
12044 @cindex tracepoint pass count
12045 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
12046 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
12047 automatically stop a trace experiment. If a tracepoint's passcount is
12048 @var{n}, then the trace experiment will be automatically stopped on
12049 the @var{n}'th time that tracepoint is hit. If the tracepoint number
12050 @var{num} is not specified, the @code{passcount} command sets the
12051 passcount of the most recently defined tracepoint. If no passcount is
12052 given, the trace experiment will run until stopped explicitly by the
12058 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
12059 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
12061 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
12062 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
12063 (@value{GDBP}) @b{trace foo}
12064 (@value{GDBP}) @b{pass 3}
12065 (@value{GDBP}) @b{trace bar}
12066 (@value{GDBP}) @b{pass 2}
12067 (@value{GDBP}) @b{trace baz}
12068 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
12069 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
12070 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
12071 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
12075 @node Tracepoint Conditions
12076 @subsection Tracepoint Conditions
12077 @cindex conditional tracepoints
12078 @cindex tracepoint conditions
12080 The simplest sort of tracepoint collects data every time your program
12081 reaches a specified place. You can also specify a @dfn{condition} for
12082 a tracepoint. A condition is just a Boolean expression in your
12083 programming language (@pxref{Expressions, ,Expressions}). A
12084 tracepoint with a condition evaluates the expression each time your
12085 program reaches it, and data collection happens only if the condition
12088 Tracepoint conditions can be specified when a tracepoint is set, by
12089 using @samp{if} in the arguments to the @code{trace} command.
12090 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
12091 also be set or changed at any time with the @code{condition} command,
12092 just as with breakpoints.
12094 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
12095 the conditional expression itself. Instead, @value{GDBN} encodes the
12096 expression into an agent expression (@pxref{Agent Expressions})
12097 suitable for execution on the target, independently of @value{GDBN}.
12098 Global variables become raw memory locations, locals become stack
12099 accesses, and so forth.
12101 For instance, suppose you have a function that is usually called
12102 frequently, but should not be called after an error has occurred. You
12103 could use the following tracepoint command to collect data about calls
12104 of that function that happen while the error code is propagating
12105 through the program; an unconditional tracepoint could end up
12106 collecting thousands of useless trace frames that you would have to
12110 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
12113 @node Trace State Variables
12114 @subsection Trace State Variables
12115 @cindex trace state variables
12117 A @dfn{trace state variable} is a special type of variable that is
12118 created and managed by target-side code. The syntax is the same as
12119 that for GDB's convenience variables (a string prefixed with ``$''),
12120 but they are stored on the target. They must be created explicitly,
12121 using a @code{tvariable} command. They are always 64-bit signed
12124 Trace state variables are remembered by @value{GDBN}, and downloaded
12125 to the target along with tracepoint information when the trace
12126 experiment starts. There are no intrinsic limits on the number of
12127 trace state variables, beyond memory limitations of the target.
12129 @cindex convenience variables, and trace state variables
12130 Although trace state variables are managed by the target, you can use
12131 them in print commands and expressions as if they were convenience
12132 variables; @value{GDBN} will get the current value from the target
12133 while the trace experiment is running. Trace state variables share
12134 the same namespace as other ``$'' variables, which means that you
12135 cannot have trace state variables with names like @code{$23} or
12136 @code{$pc}, nor can you have a trace state variable and a convenience
12137 variable with the same name.
12141 @item tvariable $@var{name} [ = @var{expression} ]
12143 The @code{tvariable} command creates a new trace state variable named
12144 @code{$@var{name}}, and optionally gives it an initial value of
12145 @var{expression}. The @var{expression} is evaluated when this command is
12146 entered; the result will be converted to an integer if possible,
12147 otherwise @value{GDBN} will report an error. A subsequent
12148 @code{tvariable} command specifying the same name does not create a
12149 variable, but instead assigns the supplied initial value to the
12150 existing variable of that name, overwriting any previous initial
12151 value. The default initial value is 0.
12153 @item info tvariables
12154 @kindex info tvariables
12155 List all the trace state variables along with their initial values.
12156 Their current values may also be displayed, if the trace experiment is
12159 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
12160 @kindex delete tvariable
12161 Delete the given trace state variables, or all of them if no arguments
12166 @node Tracepoint Actions
12167 @subsection Tracepoint Action Lists
12171 @cindex tracepoint actions
12172 @item actions @r{[}@var{num}@r{]}
12173 This command will prompt for a list of actions to be taken when the
12174 tracepoint is hit. If the tracepoint number @var{num} is not
12175 specified, this command sets the actions for the one that was most
12176 recently defined (so that you can define a tracepoint and then say
12177 @code{actions} without bothering about its number). You specify the
12178 actions themselves on the following lines, one action at a time, and
12179 terminate the actions list with a line containing just @code{end}. So
12180 far, the only defined actions are @code{collect}, @code{teval}, and
12181 @code{while-stepping}.
12183 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
12184 Commands, ,Breakpoint Command Lists}), except that only the defined
12185 actions are allowed; any other @value{GDBN} command is rejected.
12187 @cindex remove actions from a tracepoint
12188 To remove all actions from a tracepoint, type @samp{actions @var{num}}
12189 and follow it immediately with @samp{end}.
12192 (@value{GDBP}) @b{collect @var{data}} // collect some data
12194 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
12196 (@value{GDBP}) @b{end} // signals the end of actions.
12199 In the following example, the action list begins with @code{collect}
12200 commands indicating the things to be collected when the tracepoint is
12201 hit. Then, in order to single-step and collect additional data
12202 following the tracepoint, a @code{while-stepping} command is used,
12203 followed by the list of things to be collected after each step in a
12204 sequence of single steps. The @code{while-stepping} command is
12205 terminated by its own separate @code{end} command. Lastly, the action
12206 list is terminated by an @code{end} command.
12209 (@value{GDBP}) @b{trace foo}
12210 (@value{GDBP}) @b{actions}
12211 Enter actions for tracepoint 1, one per line:
12214 > while-stepping 12
12215 > collect $pc, arr[i]
12220 @kindex collect @r{(tracepoints)}
12221 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
12222 Collect values of the given expressions when the tracepoint is hit.
12223 This command accepts a comma-separated list of any valid expressions.
12224 In addition to global, static, or local variables, the following
12225 special arguments are supported:
12229 Collect all registers.
12232 Collect all function arguments.
12235 Collect all local variables.
12238 Collect the return address. This is helpful if you want to see more
12242 Collects the number of arguments from the static probe at which the
12243 tracepoint is located.
12244 @xref{Static Probe Points}.
12246 @item $_probe_arg@var{n}
12247 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
12248 from the static probe at which the tracepoint is located.
12249 @xref{Static Probe Points}.
12252 @vindex $_sdata@r{, collect}
12253 Collect static tracepoint marker specific data. Only available for
12254 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
12255 Lists}. On the UST static tracepoints library backend, an
12256 instrumentation point resembles a @code{printf} function call. The
12257 tracing library is able to collect user specified data formatted to a
12258 character string using the format provided by the programmer that
12259 instrumented the program. Other backends have similar mechanisms.
12260 Here's an example of a UST marker call:
12263 const char master_name[] = "$your_name";
12264 trace_mark(channel1, marker1, "hello %s", master_name)
12267 In this case, collecting @code{$_sdata} collects the string
12268 @samp{hello $yourname}. When analyzing the trace buffer, you can
12269 inspect @samp{$_sdata} like any other variable available to
12273 You can give several consecutive @code{collect} commands, each one
12274 with a single argument, or one @code{collect} command with several
12275 arguments separated by commas; the effect is the same.
12277 The optional @var{mods} changes the usual handling of the arguments.
12278 @code{s} requests that pointers to chars be handled as strings, in
12279 particular collecting the contents of the memory being pointed at, up
12280 to the first zero. The upper bound is by default the value of the
12281 @code{print elements} variable; if @code{s} is followed by a decimal
12282 number, that is the upper bound instead. So for instance
12283 @samp{collect/s25 mystr} collects as many as 25 characters at
12286 The command @code{info scope} (@pxref{Symbols, info scope}) is
12287 particularly useful for figuring out what data to collect.
12289 @kindex teval @r{(tracepoints)}
12290 @item teval @var{expr1}, @var{expr2}, @dots{}
12291 Evaluate the given expressions when the tracepoint is hit. This
12292 command accepts a comma-separated list of expressions. The results
12293 are discarded, so this is mainly useful for assigning values to trace
12294 state variables (@pxref{Trace State Variables}) without adding those
12295 values to the trace buffer, as would be the case if the @code{collect}
12298 @kindex while-stepping @r{(tracepoints)}
12299 @item while-stepping @var{n}
12300 Perform @var{n} single-step instruction traces after the tracepoint,
12301 collecting new data after each step. The @code{while-stepping}
12302 command is followed by the list of what to collect while stepping
12303 (followed by its own @code{end} command):
12306 > while-stepping 12
12307 > collect $regs, myglobal
12313 Note that @code{$pc} is not automatically collected by
12314 @code{while-stepping}; you need to explicitly collect that register if
12315 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
12318 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
12319 @kindex set default-collect
12320 @cindex default collection action
12321 This variable is a list of expressions to collect at each tracepoint
12322 hit. It is effectively an additional @code{collect} action prepended
12323 to every tracepoint action list. The expressions are parsed
12324 individually for each tracepoint, so for instance a variable named
12325 @code{xyz} may be interpreted as a global for one tracepoint, and a
12326 local for another, as appropriate to the tracepoint's location.
12328 @item show default-collect
12329 @kindex show default-collect
12330 Show the list of expressions that are collected by default at each
12335 @node Listing Tracepoints
12336 @subsection Listing Tracepoints
12339 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
12340 @kindex info tp @r{[}@var{n}@dots{}@r{]}
12341 @cindex information about tracepoints
12342 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
12343 Display information about the tracepoint @var{num}. If you don't
12344 specify a tracepoint number, displays information about all the
12345 tracepoints defined so far. The format is similar to that used for
12346 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
12347 command, simply restricting itself to tracepoints.
12349 A tracepoint's listing may include additional information specific to
12354 its passcount as given by the @code{passcount @var{n}} command
12357 the state about installed on target of each location
12361 (@value{GDBP}) @b{info trace}
12362 Num Type Disp Enb Address What
12363 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
12365 collect globfoo, $regs
12370 2 tracepoint keep y <MULTIPLE>
12372 2.1 y 0x0804859c in func4 at change-loc.h:35
12373 installed on target
12374 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
12375 installed on target
12376 2.3 y <PENDING> set_tracepoint
12377 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
12378 not installed on target
12383 This command can be abbreviated @code{info tp}.
12386 @node Listing Static Tracepoint Markers
12387 @subsection Listing Static Tracepoint Markers
12390 @kindex info static-tracepoint-markers
12391 @cindex information about static tracepoint markers
12392 @item info static-tracepoint-markers
12393 Display information about all static tracepoint markers defined in the
12396 For each marker, the following columns are printed:
12400 An incrementing counter, output to help readability. This is not a
12403 The marker ID, as reported by the target.
12404 @item Enabled or Disabled
12405 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
12406 that are not enabled.
12408 Where the marker is in your program, as a memory address.
12410 Where the marker is in the source for your program, as a file and line
12411 number. If the debug information included in the program does not
12412 allow @value{GDBN} to locate the source of the marker, this column
12413 will be left blank.
12417 In addition, the following information may be printed for each marker:
12421 User data passed to the tracing library by the marker call. In the
12422 UST backend, this is the format string passed as argument to the
12424 @item Static tracepoints probing the marker
12425 The list of static tracepoints attached to the marker.
12429 (@value{GDBP}) info static-tracepoint-markers
12430 Cnt ID Enb Address What
12431 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
12432 Data: number1 %d number2 %d
12433 Probed by static tracepoints: #2
12434 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
12440 @node Starting and Stopping Trace Experiments
12441 @subsection Starting and Stopping Trace Experiments
12444 @kindex tstart [ @var{notes} ]
12445 @cindex start a new trace experiment
12446 @cindex collected data discarded
12448 This command starts the trace experiment, and begins collecting data.
12449 It has the side effect of discarding all the data collected in the
12450 trace buffer during the previous trace experiment. If any arguments
12451 are supplied, they are taken as a note and stored with the trace
12452 experiment's state. The notes may be arbitrary text, and are
12453 especially useful with disconnected tracing in a multi-user context;
12454 the notes can explain what the trace is doing, supply user contact
12455 information, and so forth.
12457 @kindex tstop [ @var{notes} ]
12458 @cindex stop a running trace experiment
12460 This command stops the trace experiment. If any arguments are
12461 supplied, they are recorded with the experiment as a note. This is
12462 useful if you are stopping a trace started by someone else, for
12463 instance if the trace is interfering with the system's behavior and
12464 needs to be stopped quickly.
12466 @strong{Note}: a trace experiment and data collection may stop
12467 automatically if any tracepoint's passcount is reached
12468 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
12471 @cindex status of trace data collection
12472 @cindex trace experiment, status of
12474 This command displays the status of the current trace data
12478 Here is an example of the commands we described so far:
12481 (@value{GDBP}) @b{trace gdb_c_test}
12482 (@value{GDBP}) @b{actions}
12483 Enter actions for tracepoint #1, one per line.
12484 > collect $regs,$locals,$args
12485 > while-stepping 11
12489 (@value{GDBP}) @b{tstart}
12490 [time passes @dots{}]
12491 (@value{GDBP}) @b{tstop}
12494 @anchor{disconnected tracing}
12495 @cindex disconnected tracing
12496 You can choose to continue running the trace experiment even if
12497 @value{GDBN} disconnects from the target, voluntarily or
12498 involuntarily. For commands such as @code{detach}, the debugger will
12499 ask what you want to do with the trace. But for unexpected
12500 terminations (@value{GDBN} crash, network outage), it would be
12501 unfortunate to lose hard-won trace data, so the variable
12502 @code{disconnected-tracing} lets you decide whether the trace should
12503 continue running without @value{GDBN}.
12506 @item set disconnected-tracing on
12507 @itemx set disconnected-tracing off
12508 @kindex set disconnected-tracing
12509 Choose whether a tracing run should continue to run if @value{GDBN}
12510 has disconnected from the target. Note that @code{detach} or
12511 @code{quit} will ask you directly what to do about a running trace no
12512 matter what this variable's setting, so the variable is mainly useful
12513 for handling unexpected situations, such as loss of the network.
12515 @item show disconnected-tracing
12516 @kindex show disconnected-tracing
12517 Show the current choice for disconnected tracing.
12521 When you reconnect to the target, the trace experiment may or may not
12522 still be running; it might have filled the trace buffer in the
12523 meantime, or stopped for one of the other reasons. If it is running,
12524 it will continue after reconnection.
12526 Upon reconnection, the target will upload information about the
12527 tracepoints in effect. @value{GDBN} will then compare that
12528 information to the set of tracepoints currently defined, and attempt
12529 to match them up, allowing for the possibility that the numbers may
12530 have changed due to creation and deletion in the meantime. If one of
12531 the target's tracepoints does not match any in @value{GDBN}, the
12532 debugger will create a new tracepoint, so that you have a number with
12533 which to specify that tracepoint. This matching-up process is
12534 necessarily heuristic, and it may result in useless tracepoints being
12535 created; you may simply delete them if they are of no use.
12537 @cindex circular trace buffer
12538 If your target agent supports a @dfn{circular trace buffer}, then you
12539 can run a trace experiment indefinitely without filling the trace
12540 buffer; when space runs out, the agent deletes already-collected trace
12541 frames, oldest first, until there is enough room to continue
12542 collecting. This is especially useful if your tracepoints are being
12543 hit too often, and your trace gets terminated prematurely because the
12544 buffer is full. To ask for a circular trace buffer, simply set
12545 @samp{circular-trace-buffer} to on. You can set this at any time,
12546 including during tracing; if the agent can do it, it will change
12547 buffer handling on the fly, otherwise it will not take effect until
12551 @item set circular-trace-buffer on
12552 @itemx set circular-trace-buffer off
12553 @kindex set circular-trace-buffer
12554 Choose whether a tracing run should use a linear or circular buffer
12555 for trace data. A linear buffer will not lose any trace data, but may
12556 fill up prematurely, while a circular buffer will discard old trace
12557 data, but it will have always room for the latest tracepoint hits.
12559 @item show circular-trace-buffer
12560 @kindex show circular-trace-buffer
12561 Show the current choice for the trace buffer. Note that this may not
12562 match the agent's current buffer handling, nor is it guaranteed to
12563 match the setting that might have been in effect during a past run,
12564 for instance if you are looking at frames from a trace file.
12569 @item set trace-buffer-size @var{n}
12570 @itemx set trace-buffer-size unlimited
12571 @kindex set trace-buffer-size
12572 Request that the target use a trace buffer of @var{n} bytes. Not all
12573 targets will honor the request; they may have a compiled-in size for
12574 the trace buffer, or some other limitation. Set to a value of
12575 @code{unlimited} or @code{-1} to let the target use whatever size it
12576 likes. This is also the default.
12578 @item show trace-buffer-size
12579 @kindex show trace-buffer-size
12580 Show the current requested size for the trace buffer. Note that this
12581 will only match the actual size if the target supports size-setting,
12582 and was able to handle the requested size. For instance, if the
12583 target can only change buffer size between runs, this variable will
12584 not reflect the change until the next run starts. Use @code{tstatus}
12585 to get a report of the actual buffer size.
12589 @item set trace-user @var{text}
12590 @kindex set trace-user
12592 @item show trace-user
12593 @kindex show trace-user
12595 @item set trace-notes @var{text}
12596 @kindex set trace-notes
12597 Set the trace run's notes.
12599 @item show trace-notes
12600 @kindex show trace-notes
12601 Show the trace run's notes.
12603 @item set trace-stop-notes @var{text}
12604 @kindex set trace-stop-notes
12605 Set the trace run's stop notes. The handling of the note is as for
12606 @code{tstop} arguments; the set command is convenient way to fix a
12607 stop note that is mistaken or incomplete.
12609 @item show trace-stop-notes
12610 @kindex show trace-stop-notes
12611 Show the trace run's stop notes.
12615 @node Tracepoint Restrictions
12616 @subsection Tracepoint Restrictions
12618 @cindex tracepoint restrictions
12619 There are a number of restrictions on the use of tracepoints. As
12620 described above, tracepoint data gathering occurs on the target
12621 without interaction from @value{GDBN}. Thus the full capabilities of
12622 the debugger are not available during data gathering, and then at data
12623 examination time, you will be limited by only having what was
12624 collected. The following items describe some common problems, but it
12625 is not exhaustive, and you may run into additional difficulties not
12631 Tracepoint expressions are intended to gather objects (lvalues). Thus
12632 the full flexibility of GDB's expression evaluator is not available.
12633 You cannot call functions, cast objects to aggregate types, access
12634 convenience variables or modify values (except by assignment to trace
12635 state variables). Some language features may implicitly call
12636 functions (for instance Objective-C fields with accessors), and therefore
12637 cannot be collected either.
12640 Collection of local variables, either individually or in bulk with
12641 @code{$locals} or @code{$args}, during @code{while-stepping} may
12642 behave erratically. The stepping action may enter a new scope (for
12643 instance by stepping into a function), or the location of the variable
12644 may change (for instance it is loaded into a register). The
12645 tracepoint data recorded uses the location information for the
12646 variables that is correct for the tracepoint location. When the
12647 tracepoint is created, it is not possible, in general, to determine
12648 where the steps of a @code{while-stepping} sequence will advance the
12649 program---particularly if a conditional branch is stepped.
12652 Collection of an incompletely-initialized or partially-destroyed object
12653 may result in something that @value{GDBN} cannot display, or displays
12654 in a misleading way.
12657 When @value{GDBN} displays a pointer to character it automatically
12658 dereferences the pointer to also display characters of the string
12659 being pointed to. However, collecting the pointer during tracing does
12660 not automatically collect the string. You need to explicitly
12661 dereference the pointer and provide size information if you want to
12662 collect not only the pointer, but the memory pointed to. For example,
12663 @code{*ptr@@50} can be used to collect the 50 element array pointed to
12667 It is not possible to collect a complete stack backtrace at a
12668 tracepoint. Instead, you may collect the registers and a few hundred
12669 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
12670 (adjust to use the name of the actual stack pointer register on your
12671 target architecture, and the amount of stack you wish to capture).
12672 Then the @code{backtrace} command will show a partial backtrace when
12673 using a trace frame. The number of stack frames that can be examined
12674 depends on the sizes of the frames in the collected stack. Note that
12675 if you ask for a block so large that it goes past the bottom of the
12676 stack, the target agent may report an error trying to read from an
12680 If you do not collect registers at a tracepoint, @value{GDBN} can
12681 infer that the value of @code{$pc} must be the same as the address of
12682 the tracepoint and use that when you are looking at a trace frame
12683 for that tracepoint. However, this cannot work if the tracepoint has
12684 multiple locations (for instance if it was set in a function that was
12685 inlined), or if it has a @code{while-stepping} loop. In those cases
12686 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
12691 @node Analyze Collected Data
12692 @section Using the Collected Data
12694 After the tracepoint experiment ends, you use @value{GDBN} commands
12695 for examining the trace data. The basic idea is that each tracepoint
12696 collects a trace @dfn{snapshot} every time it is hit and another
12697 snapshot every time it single-steps. All these snapshots are
12698 consecutively numbered from zero and go into a buffer, and you can
12699 examine them later. The way you examine them is to @dfn{focus} on a
12700 specific trace snapshot. When the remote stub is focused on a trace
12701 snapshot, it will respond to all @value{GDBN} requests for memory and
12702 registers by reading from the buffer which belongs to that snapshot,
12703 rather than from @emph{real} memory or registers of the program being
12704 debugged. This means that @strong{all} @value{GDBN} commands
12705 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
12706 behave as if we were currently debugging the program state as it was
12707 when the tracepoint occurred. Any requests for data that are not in
12708 the buffer will fail.
12711 * tfind:: How to select a trace snapshot
12712 * tdump:: How to display all data for a snapshot
12713 * save tracepoints:: How to save tracepoints for a future run
12717 @subsection @code{tfind @var{n}}
12720 @cindex select trace snapshot
12721 @cindex find trace snapshot
12722 The basic command for selecting a trace snapshot from the buffer is
12723 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
12724 counting from zero. If no argument @var{n} is given, the next
12725 snapshot is selected.
12727 Here are the various forms of using the @code{tfind} command.
12731 Find the first snapshot in the buffer. This is a synonym for
12732 @code{tfind 0} (since 0 is the number of the first snapshot).
12735 Stop debugging trace snapshots, resume @emph{live} debugging.
12738 Same as @samp{tfind none}.
12741 No argument means find the next trace snapshot.
12744 Find the previous trace snapshot before the current one. This permits
12745 retracing earlier steps.
12747 @item tfind tracepoint @var{num}
12748 Find the next snapshot associated with tracepoint @var{num}. Search
12749 proceeds forward from the last examined trace snapshot. If no
12750 argument @var{num} is given, it means find the next snapshot collected
12751 for the same tracepoint as the current snapshot.
12753 @item tfind pc @var{addr}
12754 Find the next snapshot associated with the value @var{addr} of the
12755 program counter. Search proceeds forward from the last examined trace
12756 snapshot. If no argument @var{addr} is given, it means find the next
12757 snapshot with the same value of PC as the current snapshot.
12759 @item tfind outside @var{addr1}, @var{addr2}
12760 Find the next snapshot whose PC is outside the given range of
12761 addresses (exclusive).
12763 @item tfind range @var{addr1}, @var{addr2}
12764 Find the next snapshot whose PC is between @var{addr1} and
12765 @var{addr2} (inclusive).
12767 @item tfind line @r{[}@var{file}:@r{]}@var{n}
12768 Find the next snapshot associated with the source line @var{n}. If
12769 the optional argument @var{file} is given, refer to line @var{n} in
12770 that source file. Search proceeds forward from the last examined
12771 trace snapshot. If no argument @var{n} is given, it means find the
12772 next line other than the one currently being examined; thus saying
12773 @code{tfind line} repeatedly can appear to have the same effect as
12774 stepping from line to line in a @emph{live} debugging session.
12777 The default arguments for the @code{tfind} commands are specifically
12778 designed to make it easy to scan through the trace buffer. For
12779 instance, @code{tfind} with no argument selects the next trace
12780 snapshot, and @code{tfind -} with no argument selects the previous
12781 trace snapshot. So, by giving one @code{tfind} command, and then
12782 simply hitting @key{RET} repeatedly you can examine all the trace
12783 snapshots in order. Or, by saying @code{tfind -} and then hitting
12784 @key{RET} repeatedly you can examine the snapshots in reverse order.
12785 The @code{tfind line} command with no argument selects the snapshot
12786 for the next source line executed. The @code{tfind pc} command with
12787 no argument selects the next snapshot with the same program counter
12788 (PC) as the current frame. The @code{tfind tracepoint} command with
12789 no argument selects the next trace snapshot collected by the same
12790 tracepoint as the current one.
12792 In addition to letting you scan through the trace buffer manually,
12793 these commands make it easy to construct @value{GDBN} scripts that
12794 scan through the trace buffer and print out whatever collected data
12795 you are interested in. Thus, if we want to examine the PC, FP, and SP
12796 registers from each trace frame in the buffer, we can say this:
12799 (@value{GDBP}) @b{tfind start}
12800 (@value{GDBP}) @b{while ($trace_frame != -1)}
12801 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
12802 $trace_frame, $pc, $sp, $fp
12806 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
12807 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
12808 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
12809 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
12810 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
12811 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
12812 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
12813 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
12814 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
12815 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
12816 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
12819 Or, if we want to examine the variable @code{X} at each source line in
12823 (@value{GDBP}) @b{tfind start}
12824 (@value{GDBP}) @b{while ($trace_frame != -1)}
12825 > printf "Frame %d, X == %d\n", $trace_frame, X
12835 @subsection @code{tdump}
12837 @cindex dump all data collected at tracepoint
12838 @cindex tracepoint data, display
12840 This command takes no arguments. It prints all the data collected at
12841 the current trace snapshot.
12844 (@value{GDBP}) @b{trace 444}
12845 (@value{GDBP}) @b{actions}
12846 Enter actions for tracepoint #2, one per line:
12847 > collect $regs, $locals, $args, gdb_long_test
12850 (@value{GDBP}) @b{tstart}
12852 (@value{GDBP}) @b{tfind line 444}
12853 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
12855 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
12857 (@value{GDBP}) @b{tdump}
12858 Data collected at tracepoint 2, trace frame 1:
12859 d0 0xc4aa0085 -995491707
12863 d4 0x71aea3d 119204413
12866 d7 0x380035 3670069
12867 a0 0x19e24a 1696330
12868 a1 0x3000668 50333288
12870 a3 0x322000 3284992
12871 a4 0x3000698 50333336
12872 a5 0x1ad3cc 1758156
12873 fp 0x30bf3c 0x30bf3c
12874 sp 0x30bf34 0x30bf34
12876 pc 0x20b2c8 0x20b2c8
12880 p = 0x20e5b4 "gdb-test"
12887 gdb_long_test = 17 '\021'
12892 @code{tdump} works by scanning the tracepoint's current collection
12893 actions and printing the value of each expression listed. So
12894 @code{tdump} can fail, if after a run, you change the tracepoint's
12895 actions to mention variables that were not collected during the run.
12897 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
12898 uses the collected value of @code{$pc} to distinguish between trace
12899 frames that were collected at the tracepoint hit, and frames that were
12900 collected while stepping. This allows it to correctly choose whether
12901 to display the basic list of collections, or the collections from the
12902 body of the while-stepping loop. However, if @code{$pc} was not collected,
12903 then @code{tdump} will always attempt to dump using the basic collection
12904 list, and may fail if a while-stepping frame does not include all the
12905 same data that is collected at the tracepoint hit.
12906 @c This is getting pretty arcane, example would be good.
12908 @node save tracepoints
12909 @subsection @code{save tracepoints @var{filename}}
12910 @kindex save tracepoints
12911 @kindex save-tracepoints
12912 @cindex save tracepoints for future sessions
12914 This command saves all current tracepoint definitions together with
12915 their actions and passcounts, into a file @file{@var{filename}}
12916 suitable for use in a later debugging session. To read the saved
12917 tracepoint definitions, use the @code{source} command (@pxref{Command
12918 Files}). The @w{@code{save-tracepoints}} command is a deprecated
12919 alias for @w{@code{save tracepoints}}
12921 @node Tracepoint Variables
12922 @section Convenience Variables for Tracepoints
12923 @cindex tracepoint variables
12924 @cindex convenience variables for tracepoints
12927 @vindex $trace_frame
12928 @item (int) $trace_frame
12929 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
12930 snapshot is selected.
12932 @vindex $tracepoint
12933 @item (int) $tracepoint
12934 The tracepoint for the current trace snapshot.
12936 @vindex $trace_line
12937 @item (int) $trace_line
12938 The line number for the current trace snapshot.
12940 @vindex $trace_file
12941 @item (char []) $trace_file
12942 The source file for the current trace snapshot.
12944 @vindex $trace_func
12945 @item (char []) $trace_func
12946 The name of the function containing @code{$tracepoint}.
12949 Note: @code{$trace_file} is not suitable for use in @code{printf},
12950 use @code{output} instead.
12952 Here's a simple example of using these convenience variables for
12953 stepping through all the trace snapshots and printing some of their
12954 data. Note that these are not the same as trace state variables,
12955 which are managed by the target.
12958 (@value{GDBP}) @b{tfind start}
12960 (@value{GDBP}) @b{while $trace_frame != -1}
12961 > output $trace_file
12962 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
12968 @section Using Trace Files
12969 @cindex trace files
12971 In some situations, the target running a trace experiment may no
12972 longer be available; perhaps it crashed, or the hardware was needed
12973 for a different activity. To handle these cases, you can arrange to
12974 dump the trace data into a file, and later use that file as a source
12975 of trace data, via the @code{target tfile} command.
12980 @item tsave [ -r ] @var{filename}
12981 @itemx tsave [-ctf] @var{dirname}
12982 Save the trace data to @var{filename}. By default, this command
12983 assumes that @var{filename} refers to the host filesystem, so if
12984 necessary @value{GDBN} will copy raw trace data up from the target and
12985 then save it. If the target supports it, you can also supply the
12986 optional argument @code{-r} (``remote'') to direct the target to save
12987 the data directly into @var{filename} in its own filesystem, which may be
12988 more efficient if the trace buffer is very large. (Note, however, that
12989 @code{target tfile} can only read from files accessible to the host.)
12990 By default, this command will save trace frame in tfile format.
12991 You can supply the optional argument @code{-ctf} to save date in CTF
12992 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
12993 that can be shared by multiple debugging and tracing tools. Please go to
12994 @indicateurl{http://www.efficios.com/ctf} to get more information.
12996 @kindex target tfile
13000 @item target tfile @var{filename}
13001 @itemx target ctf @var{dirname}
13002 Use the file named @var{filename} or directory named @var{dirname} as
13003 a source of trace data. Commands that examine data work as they do with
13004 a live target, but it is not possible to run any new trace experiments.
13005 @code{tstatus} will report the state of the trace run at the moment
13006 the data was saved, as well as the current trace frame you are examining.
13007 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
13011 (@value{GDBP}) target ctf ctf.ctf
13012 (@value{GDBP}) tfind
13013 Found trace frame 0, tracepoint 2
13014 39 ++a; /* set tracepoint 1 here */
13015 (@value{GDBP}) tdump
13016 Data collected at tracepoint 2, trace frame 0:
13020 c = @{"123", "456", "789", "123", "456", "789"@}
13021 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
13029 @chapter Debugging Programs That Use Overlays
13032 If your program is too large to fit completely in your target system's
13033 memory, you can sometimes use @dfn{overlays} to work around this
13034 problem. @value{GDBN} provides some support for debugging programs that
13038 * How Overlays Work:: A general explanation of overlays.
13039 * Overlay Commands:: Managing overlays in @value{GDBN}.
13040 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
13041 mapped by asking the inferior.
13042 * Overlay Sample Program:: A sample program using overlays.
13045 @node How Overlays Work
13046 @section How Overlays Work
13047 @cindex mapped overlays
13048 @cindex unmapped overlays
13049 @cindex load address, overlay's
13050 @cindex mapped address
13051 @cindex overlay area
13053 Suppose you have a computer whose instruction address space is only 64
13054 kilobytes long, but which has much more memory which can be accessed by
13055 other means: special instructions, segment registers, or memory
13056 management hardware, for example. Suppose further that you want to
13057 adapt a program which is larger than 64 kilobytes to run on this system.
13059 One solution is to identify modules of your program which are relatively
13060 independent, and need not call each other directly; call these modules
13061 @dfn{overlays}. Separate the overlays from the main program, and place
13062 their machine code in the larger memory. Place your main program in
13063 instruction memory, but leave at least enough space there to hold the
13064 largest overlay as well.
13066 Now, to call a function located in an overlay, you must first copy that
13067 overlay's machine code from the large memory into the space set aside
13068 for it in the instruction memory, and then jump to its entry point
13071 @c NB: In the below the mapped area's size is greater or equal to the
13072 @c size of all overlays. This is intentional to remind the developer
13073 @c that overlays don't necessarily need to be the same size.
13077 Data Instruction Larger
13078 Address Space Address Space Address Space
13079 +-----------+ +-----------+ +-----------+
13081 +-----------+ +-----------+ +-----------+<-- overlay 1
13082 | program | | main | .----| overlay 1 | load address
13083 | variables | | program | | +-----------+
13084 | and heap | | | | | |
13085 +-----------+ | | | +-----------+<-- overlay 2
13086 | | +-----------+ | | | load address
13087 +-----------+ | | | .-| overlay 2 |
13089 mapped --->+-----------+ | | +-----------+
13090 address | | | | | |
13091 | overlay | <-' | | |
13092 | area | <---' +-----------+<-- overlay 3
13093 | | <---. | | load address
13094 +-----------+ `--| overlay 3 |
13101 @anchor{A code overlay}A code overlay
13105 The diagram (@pxref{A code overlay}) shows a system with separate data
13106 and instruction address spaces. To map an overlay, the program copies
13107 its code from the larger address space to the instruction address space.
13108 Since the overlays shown here all use the same mapped address, only one
13109 may be mapped at a time. For a system with a single address space for
13110 data and instructions, the diagram would be similar, except that the
13111 program variables and heap would share an address space with the main
13112 program and the overlay area.
13114 An overlay loaded into instruction memory and ready for use is called a
13115 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
13116 instruction memory. An overlay not present (or only partially present)
13117 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
13118 is its address in the larger memory. The mapped address is also called
13119 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
13120 called the @dfn{load memory address}, or @dfn{LMA}.
13122 Unfortunately, overlays are not a completely transparent way to adapt a
13123 program to limited instruction memory. They introduce a new set of
13124 global constraints you must keep in mind as you design your program:
13129 Before calling or returning to a function in an overlay, your program
13130 must make sure that overlay is actually mapped. Otherwise, the call or
13131 return will transfer control to the right address, but in the wrong
13132 overlay, and your program will probably crash.
13135 If the process of mapping an overlay is expensive on your system, you
13136 will need to choose your overlays carefully to minimize their effect on
13137 your program's performance.
13140 The executable file you load onto your system must contain each
13141 overlay's instructions, appearing at the overlay's load address, not its
13142 mapped address. However, each overlay's instructions must be relocated
13143 and its symbols defined as if the overlay were at its mapped address.
13144 You can use GNU linker scripts to specify different load and relocation
13145 addresses for pieces of your program; see @ref{Overlay Description,,,
13146 ld.info, Using ld: the GNU linker}.
13149 The procedure for loading executable files onto your system must be able
13150 to load their contents into the larger address space as well as the
13151 instruction and data spaces.
13155 The overlay system described above is rather simple, and could be
13156 improved in many ways:
13161 If your system has suitable bank switch registers or memory management
13162 hardware, you could use those facilities to make an overlay's load area
13163 contents simply appear at their mapped address in instruction space.
13164 This would probably be faster than copying the overlay to its mapped
13165 area in the usual way.
13168 If your overlays are small enough, you could set aside more than one
13169 overlay area, and have more than one overlay mapped at a time.
13172 You can use overlays to manage data, as well as instructions. In
13173 general, data overlays are even less transparent to your design than
13174 code overlays: whereas code overlays only require care when you call or
13175 return to functions, data overlays require care every time you access
13176 the data. Also, if you change the contents of a data overlay, you
13177 must copy its contents back out to its load address before you can copy a
13178 different data overlay into the same mapped area.
13183 @node Overlay Commands
13184 @section Overlay Commands
13186 To use @value{GDBN}'s overlay support, each overlay in your program must
13187 correspond to a separate section of the executable file. The section's
13188 virtual memory address and load memory address must be the overlay's
13189 mapped and load addresses. Identifying overlays with sections allows
13190 @value{GDBN} to determine the appropriate address of a function or
13191 variable, depending on whether the overlay is mapped or not.
13193 @value{GDBN}'s overlay commands all start with the word @code{overlay};
13194 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
13199 Disable @value{GDBN}'s overlay support. When overlay support is
13200 disabled, @value{GDBN} assumes that all functions and variables are
13201 always present at their mapped addresses. By default, @value{GDBN}'s
13202 overlay support is disabled.
13204 @item overlay manual
13205 @cindex manual overlay debugging
13206 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
13207 relies on you to tell it which overlays are mapped, and which are not,
13208 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
13209 commands described below.
13211 @item overlay map-overlay @var{overlay}
13212 @itemx overlay map @var{overlay}
13213 @cindex map an overlay
13214 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
13215 be the name of the object file section containing the overlay. When an
13216 overlay is mapped, @value{GDBN} assumes it can find the overlay's
13217 functions and variables at their mapped addresses. @value{GDBN} assumes
13218 that any other overlays whose mapped ranges overlap that of
13219 @var{overlay} are now unmapped.
13221 @item overlay unmap-overlay @var{overlay}
13222 @itemx overlay unmap @var{overlay}
13223 @cindex unmap an overlay
13224 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
13225 must be the name of the object file section containing the overlay.
13226 When an overlay is unmapped, @value{GDBN} assumes it can find the
13227 overlay's functions and variables at their load addresses.
13230 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
13231 consults a data structure the overlay manager maintains in the inferior
13232 to see which overlays are mapped. For details, see @ref{Automatic
13233 Overlay Debugging}.
13235 @item overlay load-target
13236 @itemx overlay load
13237 @cindex reloading the overlay table
13238 Re-read the overlay table from the inferior. Normally, @value{GDBN}
13239 re-reads the table @value{GDBN} automatically each time the inferior
13240 stops, so this command should only be necessary if you have changed the
13241 overlay mapping yourself using @value{GDBN}. This command is only
13242 useful when using automatic overlay debugging.
13244 @item overlay list-overlays
13245 @itemx overlay list
13246 @cindex listing mapped overlays
13247 Display a list of the overlays currently mapped, along with their mapped
13248 addresses, load addresses, and sizes.
13252 Normally, when @value{GDBN} prints a code address, it includes the name
13253 of the function the address falls in:
13256 (@value{GDBP}) print main
13257 $3 = @{int ()@} 0x11a0 <main>
13260 When overlay debugging is enabled, @value{GDBN} recognizes code in
13261 unmapped overlays, and prints the names of unmapped functions with
13262 asterisks around them. For example, if @code{foo} is a function in an
13263 unmapped overlay, @value{GDBN} prints it this way:
13266 (@value{GDBP}) overlay list
13267 No sections are mapped.
13268 (@value{GDBP}) print foo
13269 $5 = @{int (int)@} 0x100000 <*foo*>
13272 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
13276 (@value{GDBP}) overlay list
13277 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
13278 mapped at 0x1016 - 0x104a
13279 (@value{GDBP}) print foo
13280 $6 = @{int (int)@} 0x1016 <foo>
13283 When overlay debugging is enabled, @value{GDBN} can find the correct
13284 address for functions and variables in an overlay, whether or not the
13285 overlay is mapped. This allows most @value{GDBN} commands, like
13286 @code{break} and @code{disassemble}, to work normally, even on unmapped
13287 code. However, @value{GDBN}'s breakpoint support has some limitations:
13291 @cindex breakpoints in overlays
13292 @cindex overlays, setting breakpoints in
13293 You can set breakpoints in functions in unmapped overlays, as long as
13294 @value{GDBN} can write to the overlay at its load address.
13296 @value{GDBN} can not set hardware or simulator-based breakpoints in
13297 unmapped overlays. However, if you set a breakpoint at the end of your
13298 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
13299 you are using manual overlay management), @value{GDBN} will re-set its
13300 breakpoints properly.
13304 @node Automatic Overlay Debugging
13305 @section Automatic Overlay Debugging
13306 @cindex automatic overlay debugging
13308 @value{GDBN} can automatically track which overlays are mapped and which
13309 are not, given some simple co-operation from the overlay manager in the
13310 inferior. If you enable automatic overlay debugging with the
13311 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
13312 looks in the inferior's memory for certain variables describing the
13313 current state of the overlays.
13315 Here are the variables your overlay manager must define to support
13316 @value{GDBN}'s automatic overlay debugging:
13320 @item @code{_ovly_table}:
13321 This variable must be an array of the following structures:
13326 /* The overlay's mapped address. */
13329 /* The size of the overlay, in bytes. */
13330 unsigned long size;
13332 /* The overlay's load address. */
13335 /* Non-zero if the overlay is currently mapped;
13337 unsigned long mapped;
13341 @item @code{_novlys}:
13342 This variable must be a four-byte signed integer, holding the total
13343 number of elements in @code{_ovly_table}.
13347 To decide whether a particular overlay is mapped or not, @value{GDBN}
13348 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
13349 @code{lma} members equal the VMA and LMA of the overlay's section in the
13350 executable file. When @value{GDBN} finds a matching entry, it consults
13351 the entry's @code{mapped} member to determine whether the overlay is
13354 In addition, your overlay manager may define a function called
13355 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
13356 will silently set a breakpoint there. If the overlay manager then
13357 calls this function whenever it has changed the overlay table, this
13358 will enable @value{GDBN} to accurately keep track of which overlays
13359 are in program memory, and update any breakpoints that may be set
13360 in overlays. This will allow breakpoints to work even if the
13361 overlays are kept in ROM or other non-writable memory while they
13362 are not being executed.
13364 @node Overlay Sample Program
13365 @section Overlay Sample Program
13366 @cindex overlay example program
13368 When linking a program which uses overlays, you must place the overlays
13369 at their load addresses, while relocating them to run at their mapped
13370 addresses. To do this, you must write a linker script (@pxref{Overlay
13371 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
13372 since linker scripts are specific to a particular host system, target
13373 architecture, and target memory layout, this manual cannot provide
13374 portable sample code demonstrating @value{GDBN}'s overlay support.
13376 However, the @value{GDBN} source distribution does contain an overlaid
13377 program, with linker scripts for a few systems, as part of its test
13378 suite. The program consists of the following files from
13379 @file{gdb/testsuite/gdb.base}:
13383 The main program file.
13385 A simple overlay manager, used by @file{overlays.c}.
13390 Overlay modules, loaded and used by @file{overlays.c}.
13393 Linker scripts for linking the test program on the @code{d10v-elf}
13394 and @code{m32r-elf} targets.
13397 You can build the test program using the @code{d10v-elf} GCC
13398 cross-compiler like this:
13401 $ d10v-elf-gcc -g -c overlays.c
13402 $ d10v-elf-gcc -g -c ovlymgr.c
13403 $ d10v-elf-gcc -g -c foo.c
13404 $ d10v-elf-gcc -g -c bar.c
13405 $ d10v-elf-gcc -g -c baz.c
13406 $ d10v-elf-gcc -g -c grbx.c
13407 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
13408 baz.o grbx.o -Wl,-Td10v.ld -o overlays
13411 The build process is identical for any other architecture, except that
13412 you must substitute the appropriate compiler and linker script for the
13413 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
13417 @chapter Using @value{GDBN} with Different Languages
13420 Although programming languages generally have common aspects, they are
13421 rarely expressed in the same manner. For instance, in ANSI C,
13422 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
13423 Modula-2, it is accomplished by @code{p^}. Values can also be
13424 represented (and displayed) differently. Hex numbers in C appear as
13425 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
13427 @cindex working language
13428 Language-specific information is built into @value{GDBN} for some languages,
13429 allowing you to express operations like the above in your program's
13430 native language, and allowing @value{GDBN} to output values in a manner
13431 consistent with the syntax of your program's native language. The
13432 language you use to build expressions is called the @dfn{working
13436 * Setting:: Switching between source languages
13437 * Show:: Displaying the language
13438 * Checks:: Type and range checks
13439 * Supported Languages:: Supported languages
13440 * Unsupported Languages:: Unsupported languages
13444 @section Switching Between Source Languages
13446 There are two ways to control the working language---either have @value{GDBN}
13447 set it automatically, or select it manually yourself. You can use the
13448 @code{set language} command for either purpose. On startup, @value{GDBN}
13449 defaults to setting the language automatically. The working language is
13450 used to determine how expressions you type are interpreted, how values
13453 In addition to the working language, every source file that
13454 @value{GDBN} knows about has its own working language. For some object
13455 file formats, the compiler might indicate which language a particular
13456 source file is in. However, most of the time @value{GDBN} infers the
13457 language from the name of the file. The language of a source file
13458 controls whether C@t{++} names are demangled---this way @code{backtrace} can
13459 show each frame appropriately for its own language. There is no way to
13460 set the language of a source file from within @value{GDBN}, but you can
13461 set the language associated with a filename extension. @xref{Show, ,
13462 Displaying the Language}.
13464 This is most commonly a problem when you use a program, such
13465 as @code{cfront} or @code{f2c}, that generates C but is written in
13466 another language. In that case, make the
13467 program use @code{#line} directives in its C output; that way
13468 @value{GDBN} will know the correct language of the source code of the original
13469 program, and will display that source code, not the generated C code.
13472 * Filenames:: Filename extensions and languages.
13473 * Manually:: Setting the working language manually
13474 * Automatically:: Having @value{GDBN} infer the source language
13478 @subsection List of Filename Extensions and Languages
13480 If a source file name ends in one of the following extensions, then
13481 @value{GDBN} infers that its language is the one indicated.
13499 C@t{++} source file
13505 Objective-C source file
13509 Fortran source file
13512 Modula-2 source file
13516 Assembler source file. This actually behaves almost like C, but
13517 @value{GDBN} does not skip over function prologues when stepping.
13520 In addition, you may set the language associated with a filename
13521 extension. @xref{Show, , Displaying the Language}.
13524 @subsection Setting the Working Language
13526 If you allow @value{GDBN} to set the language automatically,
13527 expressions are interpreted the same way in your debugging session and
13530 @kindex set language
13531 If you wish, you may set the language manually. To do this, issue the
13532 command @samp{set language @var{lang}}, where @var{lang} is the name of
13533 a language, such as
13534 @code{c} or @code{modula-2}.
13535 For a list of the supported languages, type @samp{set language}.
13537 Setting the language manually prevents @value{GDBN} from updating the working
13538 language automatically. This can lead to confusion if you try
13539 to debug a program when the working language is not the same as the
13540 source language, when an expression is acceptable to both
13541 languages---but means different things. For instance, if the current
13542 source file were written in C, and @value{GDBN} was parsing Modula-2, a
13550 might not have the effect you intended. In C, this means to add
13551 @code{b} and @code{c} and place the result in @code{a}. The result
13552 printed would be the value of @code{a}. In Modula-2, this means to compare
13553 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
13555 @node Automatically
13556 @subsection Having @value{GDBN} Infer the Source Language
13558 To have @value{GDBN} set the working language automatically, use
13559 @samp{set language local} or @samp{set language auto}. @value{GDBN}
13560 then infers the working language. That is, when your program stops in a
13561 frame (usually by encountering a breakpoint), @value{GDBN} sets the
13562 working language to the language recorded for the function in that
13563 frame. If the language for a frame is unknown (that is, if the function
13564 or block corresponding to the frame was defined in a source file that
13565 does not have a recognized extension), the current working language is
13566 not changed, and @value{GDBN} issues a warning.
13568 This may not seem necessary for most programs, which are written
13569 entirely in one source language. However, program modules and libraries
13570 written in one source language can be used by a main program written in
13571 a different source language. Using @samp{set language auto} in this
13572 case frees you from having to set the working language manually.
13575 @section Displaying the Language
13577 The following commands help you find out which language is the
13578 working language, and also what language source files were written in.
13581 @item show language
13582 @anchor{show language}
13583 @kindex show language
13584 Display the current working language. This is the
13585 language you can use with commands such as @code{print} to
13586 build and compute expressions that may involve variables in your program.
13589 @kindex info frame@r{, show the source language}
13590 Display the source language for this frame. This language becomes the
13591 working language if you use an identifier from this frame.
13592 @xref{Frame Info, ,Information about a Frame}, to identify the other
13593 information listed here.
13596 @kindex info source@r{, show the source language}
13597 Display the source language of this source file.
13598 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
13599 information listed here.
13602 In unusual circumstances, you may have source files with extensions
13603 not in the standard list. You can then set the extension associated
13604 with a language explicitly:
13607 @item set extension-language @var{ext} @var{language}
13608 @kindex set extension-language
13609 Tell @value{GDBN} that source files with extension @var{ext} are to be
13610 assumed as written in the source language @var{language}.
13612 @item info extensions
13613 @kindex info extensions
13614 List all the filename extensions and the associated languages.
13618 @section Type and Range Checking
13620 Some languages are designed to guard you against making seemingly common
13621 errors through a series of compile- and run-time checks. These include
13622 checking the type of arguments to functions and operators and making
13623 sure mathematical overflows are caught at run time. Checks such as
13624 these help to ensure a program's correctness once it has been compiled
13625 by eliminating type mismatches and providing active checks for range
13626 errors when your program is running.
13628 By default @value{GDBN} checks for these errors according to the
13629 rules of the current source language. Although @value{GDBN} does not check
13630 the statements in your program, it can check expressions entered directly
13631 into @value{GDBN} for evaluation via the @code{print} command, for example.
13634 * Type Checking:: An overview of type checking
13635 * Range Checking:: An overview of range checking
13638 @cindex type checking
13639 @cindex checks, type
13640 @node Type Checking
13641 @subsection An Overview of Type Checking
13643 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
13644 arguments to operators and functions have to be of the correct type,
13645 otherwise an error occurs. These checks prevent type mismatch
13646 errors from ever causing any run-time problems. For example,
13649 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
13651 (@value{GDBP}) print obj.my_method (0)
13654 (@value{GDBP}) print obj.my_method (0x1234)
13655 Cannot resolve method klass::my_method to any overloaded instance
13658 The second example fails because in C@t{++} the integer constant
13659 @samp{0x1234} is not type-compatible with the pointer parameter type.
13661 For the expressions you use in @value{GDBN} commands, you can tell
13662 @value{GDBN} to not enforce strict type checking or
13663 to treat any mismatches as errors and abandon the expression;
13664 When type checking is disabled, @value{GDBN} successfully evaluates
13665 expressions like the second example above.
13667 Even if type checking is off, there may be other reasons
13668 related to type that prevent @value{GDBN} from evaluating an expression.
13669 For instance, @value{GDBN} does not know how to add an @code{int} and
13670 a @code{struct foo}. These particular type errors have nothing to do
13671 with the language in use and usually arise from expressions which make
13672 little sense to evaluate anyway.
13674 @value{GDBN} provides some additional commands for controlling type checking:
13676 @kindex set check type
13677 @kindex show check type
13679 @item set check type on
13680 @itemx set check type off
13681 Set strict type checking on or off. If any type mismatches occur in
13682 evaluating an expression while type checking is on, @value{GDBN} prints a
13683 message and aborts evaluation of the expression.
13685 @item show check type
13686 Show the current setting of type checking and whether @value{GDBN}
13687 is enforcing strict type checking rules.
13690 @cindex range checking
13691 @cindex checks, range
13692 @node Range Checking
13693 @subsection An Overview of Range Checking
13695 In some languages (such as Modula-2), it is an error to exceed the
13696 bounds of a type; this is enforced with run-time checks. Such range
13697 checking is meant to ensure program correctness by making sure
13698 computations do not overflow, or indices on an array element access do
13699 not exceed the bounds of the array.
13701 For expressions you use in @value{GDBN} commands, you can tell
13702 @value{GDBN} to treat range errors in one of three ways: ignore them,
13703 always treat them as errors and abandon the expression, or issue
13704 warnings but evaluate the expression anyway.
13706 A range error can result from numerical overflow, from exceeding an
13707 array index bound, or when you type a constant that is not a member
13708 of any type. Some languages, however, do not treat overflows as an
13709 error. In many implementations of C, mathematical overflow causes the
13710 result to ``wrap around'' to lower values---for example, if @var{m} is
13711 the largest integer value, and @var{s} is the smallest, then
13714 @var{m} + 1 @result{} @var{s}
13717 This, too, is specific to individual languages, and in some cases
13718 specific to individual compilers or machines. @xref{Supported Languages, ,
13719 Supported Languages}, for further details on specific languages.
13721 @value{GDBN} provides some additional commands for controlling the range checker:
13723 @kindex set check range
13724 @kindex show check range
13726 @item set check range auto
13727 Set range checking on or off based on the current working language.
13728 @xref{Supported Languages, ,Supported Languages}, for the default settings for
13731 @item set check range on
13732 @itemx set check range off
13733 Set range checking on or off, overriding the default setting for the
13734 current working language. A warning is issued if the setting does not
13735 match the language default. If a range error occurs and range checking is on,
13736 then a message is printed and evaluation of the expression is aborted.
13738 @item set check range warn
13739 Output messages when the @value{GDBN} range checker detects a range error,
13740 but attempt to evaluate the expression anyway. Evaluating the
13741 expression may still be impossible for other reasons, such as accessing
13742 memory that the process does not own (a typical example from many Unix
13746 Show the current setting of the range checker, and whether or not it is
13747 being set automatically by @value{GDBN}.
13750 @node Supported Languages
13751 @section Supported Languages
13753 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran, Java,
13754 OpenCL C, Pascal, assembly, Modula-2, and Ada.
13755 @c This is false ...
13756 Some @value{GDBN} features may be used in expressions regardless of the
13757 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
13758 and the @samp{@{type@}addr} construct (@pxref{Expressions,
13759 ,Expressions}) can be used with the constructs of any supported
13762 The following sections detail to what degree each source language is
13763 supported by @value{GDBN}. These sections are not meant to be language
13764 tutorials or references, but serve only as a reference guide to what the
13765 @value{GDBN} expression parser accepts, and what input and output
13766 formats should look like for different languages. There are many good
13767 books written on each of these languages; please look to these for a
13768 language reference or tutorial.
13771 * C:: C and C@t{++}
13774 * Objective-C:: Objective-C
13775 * OpenCL C:: OpenCL C
13776 * Fortran:: Fortran
13778 * Modula-2:: Modula-2
13783 @subsection C and C@t{++}
13785 @cindex C and C@t{++}
13786 @cindex expressions in C or C@t{++}
13788 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
13789 to both languages. Whenever this is the case, we discuss those languages
13793 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
13794 @cindex @sc{gnu} C@t{++}
13795 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
13796 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
13797 effectively, you must compile your C@t{++} programs with a supported
13798 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
13799 compiler (@code{aCC}).
13802 * C Operators:: C and C@t{++} operators
13803 * C Constants:: C and C@t{++} constants
13804 * C Plus Plus Expressions:: C@t{++} expressions
13805 * C Defaults:: Default settings for C and C@t{++}
13806 * C Checks:: C and C@t{++} type and range checks
13807 * Debugging C:: @value{GDBN} and C
13808 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
13809 * Decimal Floating Point:: Numbers in Decimal Floating Point format
13813 @subsubsection C and C@t{++} Operators
13815 @cindex C and C@t{++} operators
13817 Operators must be defined on values of specific types. For instance,
13818 @code{+} is defined on numbers, but not on structures. Operators are
13819 often defined on groups of types.
13821 For the purposes of C and C@t{++}, the following definitions hold:
13826 @emph{Integral types} include @code{int} with any of its storage-class
13827 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
13830 @emph{Floating-point types} include @code{float}, @code{double}, and
13831 @code{long double} (if supported by the target platform).
13834 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
13837 @emph{Scalar types} include all of the above.
13842 The following operators are supported. They are listed here
13843 in order of increasing precedence:
13847 The comma or sequencing operator. Expressions in a comma-separated list
13848 are evaluated from left to right, with the result of the entire
13849 expression being the last expression evaluated.
13852 Assignment. The value of an assignment expression is the value
13853 assigned. Defined on scalar types.
13856 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
13857 and translated to @w{@code{@var{a} = @var{a op b}}}.
13858 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
13859 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
13860 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
13863 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
13864 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
13865 should be of an integral type.
13868 Logical @sc{or}. Defined on integral types.
13871 Logical @sc{and}. Defined on integral types.
13874 Bitwise @sc{or}. Defined on integral types.
13877 Bitwise exclusive-@sc{or}. Defined on integral types.
13880 Bitwise @sc{and}. Defined on integral types.
13883 Equality and inequality. Defined on scalar types. The value of these
13884 expressions is 0 for false and non-zero for true.
13886 @item <@r{, }>@r{, }<=@r{, }>=
13887 Less than, greater than, less than or equal, greater than or equal.
13888 Defined on scalar types. The value of these expressions is 0 for false
13889 and non-zero for true.
13892 left shift, and right shift. Defined on integral types.
13895 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
13898 Addition and subtraction. Defined on integral types, floating-point types and
13901 @item *@r{, }/@r{, }%
13902 Multiplication, division, and modulus. Multiplication and division are
13903 defined on integral and floating-point types. Modulus is defined on
13907 Increment and decrement. When appearing before a variable, the
13908 operation is performed before the variable is used in an expression;
13909 when appearing after it, the variable's value is used before the
13910 operation takes place.
13913 Pointer dereferencing. Defined on pointer types. Same precedence as
13917 Address operator. Defined on variables. Same precedence as @code{++}.
13919 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
13920 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
13921 to examine the address
13922 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
13926 Negative. Defined on integral and floating-point types. Same
13927 precedence as @code{++}.
13930 Logical negation. Defined on integral types. Same precedence as
13934 Bitwise complement operator. Defined on integral types. Same precedence as
13939 Structure member, and pointer-to-structure member. For convenience,
13940 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
13941 pointer based on the stored type information.
13942 Defined on @code{struct} and @code{union} data.
13945 Dereferences of pointers to members.
13948 Array indexing. @code{@var{a}[@var{i}]} is defined as
13949 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
13952 Function parameter list. Same precedence as @code{->}.
13955 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
13956 and @code{class} types.
13959 Doubled colons also represent the @value{GDBN} scope operator
13960 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
13964 If an operator is redefined in the user code, @value{GDBN} usually
13965 attempts to invoke the redefined version instead of using the operator's
13966 predefined meaning.
13969 @subsubsection C and C@t{++} Constants
13971 @cindex C and C@t{++} constants
13973 @value{GDBN} allows you to express the constants of C and C@t{++} in the
13978 Integer constants are a sequence of digits. Octal constants are
13979 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
13980 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
13981 @samp{l}, specifying that the constant should be treated as a
13985 Floating point constants are a sequence of digits, followed by a decimal
13986 point, followed by a sequence of digits, and optionally followed by an
13987 exponent. An exponent is of the form:
13988 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
13989 sequence of digits. The @samp{+} is optional for positive exponents.
13990 A floating-point constant may also end with a letter @samp{f} or
13991 @samp{F}, specifying that the constant should be treated as being of
13992 the @code{float} (as opposed to the default @code{double}) type; or with
13993 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
13997 Enumerated constants consist of enumerated identifiers, or their
13998 integral equivalents.
14001 Character constants are a single character surrounded by single quotes
14002 (@code{'}), or a number---the ordinal value of the corresponding character
14003 (usually its @sc{ascii} value). Within quotes, the single character may
14004 be represented by a letter or by @dfn{escape sequences}, which are of
14005 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
14006 of the character's ordinal value; or of the form @samp{\@var{x}}, where
14007 @samp{@var{x}} is a predefined special character---for example,
14008 @samp{\n} for newline.
14010 Wide character constants can be written by prefixing a character
14011 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
14012 form of @samp{x}. The target wide character set is used when
14013 computing the value of this constant (@pxref{Character Sets}).
14016 String constants are a sequence of character constants surrounded by
14017 double quotes (@code{"}). Any valid character constant (as described
14018 above) may appear. Double quotes within the string must be preceded by
14019 a backslash, so for instance @samp{"a\"b'c"} is a string of five
14022 Wide string constants can be written by prefixing a string constant
14023 with @samp{L}, as in C. The target wide character set is used when
14024 computing the value of this constant (@pxref{Character Sets}).
14027 Pointer constants are an integral value. You can also write pointers
14028 to constants using the C operator @samp{&}.
14031 Array constants are comma-separated lists surrounded by braces @samp{@{}
14032 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
14033 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
14034 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
14037 @node C Plus Plus Expressions
14038 @subsubsection C@t{++} Expressions
14040 @cindex expressions in C@t{++}
14041 @value{GDBN} expression handling can interpret most C@t{++} expressions.
14043 @cindex debugging C@t{++} programs
14044 @cindex C@t{++} compilers
14045 @cindex debug formats and C@t{++}
14046 @cindex @value{NGCC} and C@t{++}
14048 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
14049 the proper compiler and the proper debug format. Currently,
14050 @value{GDBN} works best when debugging C@t{++} code that is compiled
14051 with the most recent version of @value{NGCC} possible. The DWARF
14052 debugging format is preferred; @value{NGCC} defaults to this on most
14053 popular platforms. Other compilers and/or debug formats are likely to
14054 work badly or not at all when using @value{GDBN} to debug C@t{++}
14055 code. @xref{Compilation}.
14060 @cindex member functions
14062 Member function calls are allowed; you can use expressions like
14065 count = aml->GetOriginal(x, y)
14068 @vindex this@r{, inside C@t{++} member functions}
14069 @cindex namespace in C@t{++}
14071 While a member function is active (in the selected stack frame), your
14072 expressions have the same namespace available as the member function;
14073 that is, @value{GDBN} allows implicit references to the class instance
14074 pointer @code{this} following the same rules as C@t{++}. @code{using}
14075 declarations in the current scope are also respected by @value{GDBN}.
14077 @cindex call overloaded functions
14078 @cindex overloaded functions, calling
14079 @cindex type conversions in C@t{++}
14081 You can call overloaded functions; @value{GDBN} resolves the function
14082 call to the right definition, with some restrictions. @value{GDBN} does not
14083 perform overload resolution involving user-defined type conversions,
14084 calls to constructors, or instantiations of templates that do not exist
14085 in the program. It also cannot handle ellipsis argument lists or
14088 It does perform integral conversions and promotions, floating-point
14089 promotions, arithmetic conversions, pointer conversions, conversions of
14090 class objects to base classes, and standard conversions such as those of
14091 functions or arrays to pointers; it requires an exact match on the
14092 number of function arguments.
14094 Overload resolution is always performed, unless you have specified
14095 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
14096 ,@value{GDBN} Features for C@t{++}}.
14098 You must specify @code{set overload-resolution off} in order to use an
14099 explicit function signature to call an overloaded function, as in
14101 p 'foo(char,int)'('x', 13)
14104 The @value{GDBN} command-completion facility can simplify this;
14105 see @ref{Completion, ,Command Completion}.
14107 @cindex reference declarations
14109 @value{GDBN} understands variables declared as C@t{++} references; you can use
14110 them in expressions just as you do in C@t{++} source---they are automatically
14113 In the parameter list shown when @value{GDBN} displays a frame, the values of
14114 reference variables are not displayed (unlike other variables); this
14115 avoids clutter, since references are often used for large structures.
14116 The @emph{address} of a reference variable is always shown, unless
14117 you have specified @samp{set print address off}.
14120 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
14121 expressions can use it just as expressions in your program do. Since
14122 one scope may be defined in another, you can use @code{::} repeatedly if
14123 necessary, for example in an expression like
14124 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
14125 resolving name scope by reference to source files, in both C and C@t{++}
14126 debugging (@pxref{Variables, ,Program Variables}).
14129 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
14134 @subsubsection C and C@t{++} Defaults
14136 @cindex C and C@t{++} defaults
14138 If you allow @value{GDBN} to set range checking automatically, it
14139 defaults to @code{off} whenever the working language changes to
14140 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
14141 selects the working language.
14143 If you allow @value{GDBN} to set the language automatically, it
14144 recognizes source files whose names end with @file{.c}, @file{.C}, or
14145 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
14146 these files, it sets the working language to C or C@t{++}.
14147 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
14148 for further details.
14151 @subsubsection C and C@t{++} Type and Range Checks
14153 @cindex C and C@t{++} checks
14155 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
14156 checking is used. However, if you turn type checking off, @value{GDBN}
14157 will allow certain non-standard conversions, such as promoting integer
14158 constants to pointers.
14160 Range checking, if turned on, is done on mathematical operations. Array
14161 indices are not checked, since they are often used to index a pointer
14162 that is not itself an array.
14165 @subsubsection @value{GDBN} and C
14167 The @code{set print union} and @code{show print union} commands apply to
14168 the @code{union} type. When set to @samp{on}, any @code{union} that is
14169 inside a @code{struct} or @code{class} is also printed. Otherwise, it
14170 appears as @samp{@{...@}}.
14172 The @code{@@} operator aids in the debugging of dynamic arrays, formed
14173 with pointers and a memory allocation function. @xref{Expressions,
14176 @node Debugging C Plus Plus
14177 @subsubsection @value{GDBN} Features for C@t{++}
14179 @cindex commands for C@t{++}
14181 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
14182 designed specifically for use with C@t{++}. Here is a summary:
14185 @cindex break in overloaded functions
14186 @item @r{breakpoint menus}
14187 When you want a breakpoint in a function whose name is overloaded,
14188 @value{GDBN} has the capability to display a menu of possible breakpoint
14189 locations to help you specify which function definition you want.
14190 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
14192 @cindex overloading in C@t{++}
14193 @item rbreak @var{regex}
14194 Setting breakpoints using regular expressions is helpful for setting
14195 breakpoints on overloaded functions that are not members of any special
14197 @xref{Set Breaks, ,Setting Breakpoints}.
14199 @cindex C@t{++} exception handling
14201 @itemx catch rethrow
14203 Debug C@t{++} exception handling using these commands. @xref{Set
14204 Catchpoints, , Setting Catchpoints}.
14206 @cindex inheritance
14207 @item ptype @var{typename}
14208 Print inheritance relationships as well as other information for type
14210 @xref{Symbols, ,Examining the Symbol Table}.
14212 @item info vtbl @var{expression}.
14213 The @code{info vtbl} command can be used to display the virtual
14214 method tables of the object computed by @var{expression}. This shows
14215 one entry per virtual table; there may be multiple virtual tables when
14216 multiple inheritance is in use.
14218 @cindex C@t{++} symbol display
14219 @item set print demangle
14220 @itemx show print demangle
14221 @itemx set print asm-demangle
14222 @itemx show print asm-demangle
14223 Control whether C@t{++} symbols display in their source form, both when
14224 displaying code as C@t{++} source and when displaying disassemblies.
14225 @xref{Print Settings, ,Print Settings}.
14227 @item set print object
14228 @itemx show print object
14229 Choose whether to print derived (actual) or declared types of objects.
14230 @xref{Print Settings, ,Print Settings}.
14232 @item set print vtbl
14233 @itemx show print vtbl
14234 Control the format for printing virtual function tables.
14235 @xref{Print Settings, ,Print Settings}.
14236 (The @code{vtbl} commands do not work on programs compiled with the HP
14237 ANSI C@t{++} compiler (@code{aCC}).)
14239 @kindex set overload-resolution
14240 @cindex overloaded functions, overload resolution
14241 @item set overload-resolution on
14242 Enable overload resolution for C@t{++} expression evaluation. The default
14243 is on. For overloaded functions, @value{GDBN} evaluates the arguments
14244 and searches for a function whose signature matches the argument types,
14245 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
14246 Expressions, ,C@t{++} Expressions}, for details).
14247 If it cannot find a match, it emits a message.
14249 @item set overload-resolution off
14250 Disable overload resolution for C@t{++} expression evaluation. For
14251 overloaded functions that are not class member functions, @value{GDBN}
14252 chooses the first function of the specified name that it finds in the
14253 symbol table, whether or not its arguments are of the correct type. For
14254 overloaded functions that are class member functions, @value{GDBN}
14255 searches for a function whose signature @emph{exactly} matches the
14258 @kindex show overload-resolution
14259 @item show overload-resolution
14260 Show the current setting of overload resolution.
14262 @item @r{Overloaded symbol names}
14263 You can specify a particular definition of an overloaded symbol, using
14264 the same notation that is used to declare such symbols in C@t{++}: type
14265 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
14266 also use the @value{GDBN} command-line word completion facilities to list the
14267 available choices, or to finish the type list for you.
14268 @xref{Completion,, Command Completion}, for details on how to do this.
14271 @node Decimal Floating Point
14272 @subsubsection Decimal Floating Point format
14273 @cindex decimal floating point format
14275 @value{GDBN} can examine, set and perform computations with numbers in
14276 decimal floating point format, which in the C language correspond to the
14277 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
14278 specified by the extension to support decimal floating-point arithmetic.
14280 There are two encodings in use, depending on the architecture: BID (Binary
14281 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
14282 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
14285 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
14286 to manipulate decimal floating point numbers, it is not possible to convert
14287 (using a cast, for example) integers wider than 32-bit to decimal float.
14289 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
14290 point computations, error checking in decimal float operations ignores
14291 underflow, overflow and divide by zero exceptions.
14293 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
14294 to inspect @code{_Decimal128} values stored in floating point registers.
14295 See @ref{PowerPC,,PowerPC} for more details.
14301 @value{GDBN} can be used to debug programs written in D and compiled with
14302 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
14303 specific feature --- dynamic arrays.
14308 @cindex Go (programming language)
14309 @value{GDBN} can be used to debug programs written in Go and compiled with
14310 @file{gccgo} or @file{6g} compilers.
14312 Here is a summary of the Go-specific features and restrictions:
14315 @cindex current Go package
14316 @item The current Go package
14317 The name of the current package does not need to be specified when
14318 specifying global variables and functions.
14320 For example, given the program:
14324 var myglob = "Shall we?"
14330 When stopped inside @code{main} either of these work:
14334 (gdb) p main.myglob
14337 @cindex builtin Go types
14338 @item Builtin Go types
14339 The @code{string} type is recognized by @value{GDBN} and is printed
14342 @cindex builtin Go functions
14343 @item Builtin Go functions
14344 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
14345 function and handles it internally.
14347 @cindex restrictions on Go expressions
14348 @item Restrictions on Go expressions
14349 All Go operators are supported except @code{&^}.
14350 The Go @code{_} ``blank identifier'' is not supported.
14351 Automatic dereferencing of pointers is not supported.
14355 @subsection Objective-C
14357 @cindex Objective-C
14358 This section provides information about some commands and command
14359 options that are useful for debugging Objective-C code. See also
14360 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
14361 few more commands specific to Objective-C support.
14364 * Method Names in Commands::
14365 * The Print Command with Objective-C::
14368 @node Method Names in Commands
14369 @subsubsection Method Names in Commands
14371 The following commands have been extended to accept Objective-C method
14372 names as line specifications:
14374 @kindex clear@r{, and Objective-C}
14375 @kindex break@r{, and Objective-C}
14376 @kindex info line@r{, and Objective-C}
14377 @kindex jump@r{, and Objective-C}
14378 @kindex list@r{, and Objective-C}
14382 @item @code{info line}
14387 A fully qualified Objective-C method name is specified as
14390 -[@var{Class} @var{methodName}]
14393 where the minus sign is used to indicate an instance method and a
14394 plus sign (not shown) is used to indicate a class method. The class
14395 name @var{Class} and method name @var{methodName} are enclosed in
14396 brackets, similar to the way messages are specified in Objective-C
14397 source code. For example, to set a breakpoint at the @code{create}
14398 instance method of class @code{Fruit} in the program currently being
14402 break -[Fruit create]
14405 To list ten program lines around the @code{initialize} class method,
14409 list +[NSText initialize]
14412 In the current version of @value{GDBN}, the plus or minus sign is
14413 required. In future versions of @value{GDBN}, the plus or minus
14414 sign will be optional, but you can use it to narrow the search. It
14415 is also possible to specify just a method name:
14421 You must specify the complete method name, including any colons. If
14422 your program's source files contain more than one @code{create} method,
14423 you'll be presented with a numbered list of classes that implement that
14424 method. Indicate your choice by number, or type @samp{0} to exit if
14427 As another example, to clear a breakpoint established at the
14428 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
14431 clear -[NSWindow makeKeyAndOrderFront:]
14434 @node The Print Command with Objective-C
14435 @subsubsection The Print Command With Objective-C
14436 @cindex Objective-C, print objects
14437 @kindex print-object
14438 @kindex po @r{(@code{print-object})}
14440 The print command has also been extended to accept methods. For example:
14443 print -[@var{object} hash]
14446 @cindex print an Objective-C object description
14447 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
14449 will tell @value{GDBN} to send the @code{hash} message to @var{object}
14450 and print the result. Also, an additional command has been added,
14451 @code{print-object} or @code{po} for short, which is meant to print
14452 the description of an object. However, this command may only work
14453 with certain Objective-C libraries that have a particular hook
14454 function, @code{_NSPrintForDebugger}, defined.
14457 @subsection OpenCL C
14460 This section provides information about @value{GDBN}s OpenCL C support.
14463 * OpenCL C Datatypes::
14464 * OpenCL C Expressions::
14465 * OpenCL C Operators::
14468 @node OpenCL C Datatypes
14469 @subsubsection OpenCL C Datatypes
14471 @cindex OpenCL C Datatypes
14472 @value{GDBN} supports the builtin scalar and vector datatypes specified
14473 by OpenCL 1.1. In addition the half- and double-precision floating point
14474 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
14475 extensions are also known to @value{GDBN}.
14477 @node OpenCL C Expressions
14478 @subsubsection OpenCL C Expressions
14480 @cindex OpenCL C Expressions
14481 @value{GDBN} supports accesses to vector components including the access as
14482 lvalue where possible. Since OpenCL C is based on C99 most C expressions
14483 supported by @value{GDBN} can be used as well.
14485 @node OpenCL C Operators
14486 @subsubsection OpenCL C Operators
14488 @cindex OpenCL C Operators
14489 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
14493 @subsection Fortran
14494 @cindex Fortran-specific support in @value{GDBN}
14496 @value{GDBN} can be used to debug programs written in Fortran, but it
14497 currently supports only the features of Fortran 77 language.
14499 @cindex trailing underscore, in Fortran symbols
14500 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
14501 among them) append an underscore to the names of variables and
14502 functions. When you debug programs compiled by those compilers, you
14503 will need to refer to variables and functions with a trailing
14507 * Fortran Operators:: Fortran operators and expressions
14508 * Fortran Defaults:: Default settings for Fortran
14509 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
14512 @node Fortran Operators
14513 @subsubsection Fortran Operators and Expressions
14515 @cindex Fortran operators and expressions
14517 Operators must be defined on values of specific types. For instance,
14518 @code{+} is defined on numbers, but not on characters or other non-
14519 arithmetic types. Operators are often defined on groups of types.
14523 The exponentiation operator. It raises the first operand to the power
14527 The range operator. Normally used in the form of array(low:high) to
14528 represent a section of array.
14531 The access component operator. Normally used to access elements in derived
14532 types. Also suitable for unions. As unions aren't part of regular Fortran,
14533 this can only happen when accessing a register that uses a gdbarch-defined
14537 @node Fortran Defaults
14538 @subsubsection Fortran Defaults
14540 @cindex Fortran Defaults
14542 Fortran symbols are usually case-insensitive, so @value{GDBN} by
14543 default uses case-insensitive matches for Fortran symbols. You can
14544 change that with the @samp{set case-insensitive} command, see
14545 @ref{Symbols}, for the details.
14547 @node Special Fortran Commands
14548 @subsubsection Special Fortran Commands
14550 @cindex Special Fortran commands
14552 @value{GDBN} has some commands to support Fortran-specific features,
14553 such as displaying common blocks.
14556 @cindex @code{COMMON} blocks, Fortran
14557 @kindex info common
14558 @item info common @r{[}@var{common-name}@r{]}
14559 This command prints the values contained in the Fortran @code{COMMON}
14560 block whose name is @var{common-name}. With no argument, the names of
14561 all @code{COMMON} blocks visible at the current program location are
14568 @cindex Pascal support in @value{GDBN}, limitations
14569 Debugging Pascal programs which use sets, subranges, file variables, or
14570 nested functions does not currently work. @value{GDBN} does not support
14571 entering expressions, printing values, or similar features using Pascal
14574 The Pascal-specific command @code{set print pascal_static-members}
14575 controls whether static members of Pascal objects are displayed.
14576 @xref{Print Settings, pascal_static-members}.
14579 @subsection Modula-2
14581 @cindex Modula-2, @value{GDBN} support
14583 The extensions made to @value{GDBN} to support Modula-2 only support
14584 output from the @sc{gnu} Modula-2 compiler (which is currently being
14585 developed). Other Modula-2 compilers are not currently supported, and
14586 attempting to debug executables produced by them is most likely
14587 to give an error as @value{GDBN} reads in the executable's symbol
14590 @cindex expressions in Modula-2
14592 * M2 Operators:: Built-in operators
14593 * Built-In Func/Proc:: Built-in functions and procedures
14594 * M2 Constants:: Modula-2 constants
14595 * M2 Types:: Modula-2 types
14596 * M2 Defaults:: Default settings for Modula-2
14597 * Deviations:: Deviations from standard Modula-2
14598 * M2 Checks:: Modula-2 type and range checks
14599 * M2 Scope:: The scope operators @code{::} and @code{.}
14600 * GDB/M2:: @value{GDBN} and Modula-2
14604 @subsubsection Operators
14605 @cindex Modula-2 operators
14607 Operators must be defined on values of specific types. For instance,
14608 @code{+} is defined on numbers, but not on structures. Operators are
14609 often defined on groups of types. For the purposes of Modula-2, the
14610 following definitions hold:
14615 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
14619 @emph{Character types} consist of @code{CHAR} and its subranges.
14622 @emph{Floating-point types} consist of @code{REAL}.
14625 @emph{Pointer types} consist of anything declared as @code{POINTER TO
14629 @emph{Scalar types} consist of all of the above.
14632 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
14635 @emph{Boolean types} consist of @code{BOOLEAN}.
14639 The following operators are supported, and appear in order of
14640 increasing precedence:
14644 Function argument or array index separator.
14647 Assignment. The value of @var{var} @code{:=} @var{value} is
14651 Less than, greater than on integral, floating-point, or enumerated
14655 Less than or equal to, greater than or equal to
14656 on integral, floating-point and enumerated types, or set inclusion on
14657 set types. Same precedence as @code{<}.
14659 @item =@r{, }<>@r{, }#
14660 Equality and two ways of expressing inequality, valid on scalar types.
14661 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
14662 available for inequality, since @code{#} conflicts with the script
14666 Set membership. Defined on set types and the types of their members.
14667 Same precedence as @code{<}.
14670 Boolean disjunction. Defined on boolean types.
14673 Boolean conjunction. Defined on boolean types.
14676 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
14679 Addition and subtraction on integral and floating-point types, or union
14680 and difference on set types.
14683 Multiplication on integral and floating-point types, or set intersection
14687 Division on floating-point types, or symmetric set difference on set
14688 types. Same precedence as @code{*}.
14691 Integer division and remainder. Defined on integral types. Same
14692 precedence as @code{*}.
14695 Negative. Defined on @code{INTEGER} and @code{REAL} data.
14698 Pointer dereferencing. Defined on pointer types.
14701 Boolean negation. Defined on boolean types. Same precedence as
14705 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
14706 precedence as @code{^}.
14709 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
14712 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
14716 @value{GDBN} and Modula-2 scope operators.
14720 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
14721 treats the use of the operator @code{IN}, or the use of operators
14722 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
14723 @code{<=}, and @code{>=} on sets as an error.
14727 @node Built-In Func/Proc
14728 @subsubsection Built-in Functions and Procedures
14729 @cindex Modula-2 built-ins
14731 Modula-2 also makes available several built-in procedures and functions.
14732 In describing these, the following metavariables are used:
14737 represents an @code{ARRAY} variable.
14740 represents a @code{CHAR} constant or variable.
14743 represents a variable or constant of integral type.
14746 represents an identifier that belongs to a set. Generally used in the
14747 same function with the metavariable @var{s}. The type of @var{s} should
14748 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
14751 represents a variable or constant of integral or floating-point type.
14754 represents a variable or constant of floating-point type.
14760 represents a variable.
14763 represents a variable or constant of one of many types. See the
14764 explanation of the function for details.
14767 All Modula-2 built-in procedures also return a result, described below.
14771 Returns the absolute value of @var{n}.
14774 If @var{c} is a lower case letter, it returns its upper case
14775 equivalent, otherwise it returns its argument.
14778 Returns the character whose ordinal value is @var{i}.
14781 Decrements the value in the variable @var{v} by one. Returns the new value.
14783 @item DEC(@var{v},@var{i})
14784 Decrements the value in the variable @var{v} by @var{i}. Returns the
14787 @item EXCL(@var{m},@var{s})
14788 Removes the element @var{m} from the set @var{s}. Returns the new
14791 @item FLOAT(@var{i})
14792 Returns the floating point equivalent of the integer @var{i}.
14794 @item HIGH(@var{a})
14795 Returns the index of the last member of @var{a}.
14798 Increments the value in the variable @var{v} by one. Returns the new value.
14800 @item INC(@var{v},@var{i})
14801 Increments the value in the variable @var{v} by @var{i}. Returns the
14804 @item INCL(@var{m},@var{s})
14805 Adds the element @var{m} to the set @var{s} if it is not already
14806 there. Returns the new set.
14809 Returns the maximum value of the type @var{t}.
14812 Returns the minimum value of the type @var{t}.
14815 Returns boolean TRUE if @var{i} is an odd number.
14818 Returns the ordinal value of its argument. For example, the ordinal
14819 value of a character is its @sc{ascii} value (on machines supporting
14820 the @sc{ascii} character set). The argument @var{x} must be of an
14821 ordered type, which include integral, character and enumerated types.
14823 @item SIZE(@var{x})
14824 Returns the size of its argument. The argument @var{x} can be a
14825 variable or a type.
14827 @item TRUNC(@var{r})
14828 Returns the integral part of @var{r}.
14830 @item TSIZE(@var{x})
14831 Returns the size of its argument. The argument @var{x} can be a
14832 variable or a type.
14834 @item VAL(@var{t},@var{i})
14835 Returns the member of the type @var{t} whose ordinal value is @var{i}.
14839 @emph{Warning:} Sets and their operations are not yet supported, so
14840 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
14844 @cindex Modula-2 constants
14846 @subsubsection Constants
14848 @value{GDBN} allows you to express the constants of Modula-2 in the following
14854 Integer constants are simply a sequence of digits. When used in an
14855 expression, a constant is interpreted to be type-compatible with the
14856 rest of the expression. Hexadecimal integers are specified by a
14857 trailing @samp{H}, and octal integers by a trailing @samp{B}.
14860 Floating point constants appear as a sequence of digits, followed by a
14861 decimal point and another sequence of digits. An optional exponent can
14862 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
14863 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
14864 digits of the floating point constant must be valid decimal (base 10)
14868 Character constants consist of a single character enclosed by a pair of
14869 like quotes, either single (@code{'}) or double (@code{"}). They may
14870 also be expressed by their ordinal value (their @sc{ascii} value, usually)
14871 followed by a @samp{C}.
14874 String constants consist of a sequence of characters enclosed by a
14875 pair of like quotes, either single (@code{'}) or double (@code{"}).
14876 Escape sequences in the style of C are also allowed. @xref{C
14877 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
14881 Enumerated constants consist of an enumerated identifier.
14884 Boolean constants consist of the identifiers @code{TRUE} and
14888 Pointer constants consist of integral values only.
14891 Set constants are not yet supported.
14895 @subsubsection Modula-2 Types
14896 @cindex Modula-2 types
14898 Currently @value{GDBN} can print the following data types in Modula-2
14899 syntax: array types, record types, set types, pointer types, procedure
14900 types, enumerated types, subrange types and base types. You can also
14901 print the contents of variables declared using these type.
14902 This section gives a number of simple source code examples together with
14903 sample @value{GDBN} sessions.
14905 The first example contains the following section of code:
14914 and you can request @value{GDBN} to interrogate the type and value of
14915 @code{r} and @code{s}.
14918 (@value{GDBP}) print s
14920 (@value{GDBP}) ptype s
14922 (@value{GDBP}) print r
14924 (@value{GDBP}) ptype r
14929 Likewise if your source code declares @code{s} as:
14933 s: SET ['A'..'Z'] ;
14937 then you may query the type of @code{s} by:
14940 (@value{GDBP}) ptype s
14941 type = SET ['A'..'Z']
14945 Note that at present you cannot interactively manipulate set
14946 expressions using the debugger.
14948 The following example shows how you might declare an array in Modula-2
14949 and how you can interact with @value{GDBN} to print its type and contents:
14953 s: ARRAY [-10..10] OF CHAR ;
14957 (@value{GDBP}) ptype s
14958 ARRAY [-10..10] OF CHAR
14961 Note that the array handling is not yet complete and although the type
14962 is printed correctly, expression handling still assumes that all
14963 arrays have a lower bound of zero and not @code{-10} as in the example
14966 Here are some more type related Modula-2 examples:
14970 colour = (blue, red, yellow, green) ;
14971 t = [blue..yellow] ;
14979 The @value{GDBN} interaction shows how you can query the data type
14980 and value of a variable.
14983 (@value{GDBP}) print s
14985 (@value{GDBP}) ptype t
14986 type = [blue..yellow]
14990 In this example a Modula-2 array is declared and its contents
14991 displayed. Observe that the contents are written in the same way as
14992 their @code{C} counterparts.
14996 s: ARRAY [1..5] OF CARDINAL ;
15002 (@value{GDBP}) print s
15003 $1 = @{1, 0, 0, 0, 0@}
15004 (@value{GDBP}) ptype s
15005 type = ARRAY [1..5] OF CARDINAL
15008 The Modula-2 language interface to @value{GDBN} also understands
15009 pointer types as shown in this example:
15013 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
15020 and you can request that @value{GDBN} describes the type of @code{s}.
15023 (@value{GDBP}) ptype s
15024 type = POINTER TO ARRAY [1..5] OF CARDINAL
15027 @value{GDBN} handles compound types as we can see in this example.
15028 Here we combine array types, record types, pointer types and subrange
15039 myarray = ARRAY myrange OF CARDINAL ;
15040 myrange = [-2..2] ;
15042 s: POINTER TO ARRAY myrange OF foo ;
15046 and you can ask @value{GDBN} to describe the type of @code{s} as shown
15050 (@value{GDBP}) ptype s
15051 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
15054 f3 : ARRAY [-2..2] OF CARDINAL;
15059 @subsubsection Modula-2 Defaults
15060 @cindex Modula-2 defaults
15062 If type and range checking are set automatically by @value{GDBN}, they
15063 both default to @code{on} whenever the working language changes to
15064 Modula-2. This happens regardless of whether you or @value{GDBN}
15065 selected the working language.
15067 If you allow @value{GDBN} to set the language automatically, then entering
15068 code compiled from a file whose name ends with @file{.mod} sets the
15069 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
15070 Infer the Source Language}, for further details.
15073 @subsubsection Deviations from Standard Modula-2
15074 @cindex Modula-2, deviations from
15076 A few changes have been made to make Modula-2 programs easier to debug.
15077 This is done primarily via loosening its type strictness:
15081 Unlike in standard Modula-2, pointer constants can be formed by
15082 integers. This allows you to modify pointer variables during
15083 debugging. (In standard Modula-2, the actual address contained in a
15084 pointer variable is hidden from you; it can only be modified
15085 through direct assignment to another pointer variable or expression that
15086 returned a pointer.)
15089 C escape sequences can be used in strings and characters to represent
15090 non-printable characters. @value{GDBN} prints out strings with these
15091 escape sequences embedded. Single non-printable characters are
15092 printed using the @samp{CHR(@var{nnn})} format.
15095 The assignment operator (@code{:=}) returns the value of its right-hand
15099 All built-in procedures both modify @emph{and} return their argument.
15103 @subsubsection Modula-2 Type and Range Checks
15104 @cindex Modula-2 checks
15107 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
15110 @c FIXME remove warning when type/range checks added
15112 @value{GDBN} considers two Modula-2 variables type equivalent if:
15116 They are of types that have been declared equivalent via a @code{TYPE
15117 @var{t1} = @var{t2}} statement
15120 They have been declared on the same line. (Note: This is true of the
15121 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
15124 As long as type checking is enabled, any attempt to combine variables
15125 whose types are not equivalent is an error.
15127 Range checking is done on all mathematical operations, assignment, array
15128 index bounds, and all built-in functions and procedures.
15131 @subsubsection The Scope Operators @code{::} and @code{.}
15133 @cindex @code{.}, Modula-2 scope operator
15134 @cindex colon, doubled as scope operator
15136 @vindex colon-colon@r{, in Modula-2}
15137 @c Info cannot handle :: but TeX can.
15140 @vindex ::@r{, in Modula-2}
15143 There are a few subtle differences between the Modula-2 scope operator
15144 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
15149 @var{module} . @var{id}
15150 @var{scope} :: @var{id}
15154 where @var{scope} is the name of a module or a procedure,
15155 @var{module} the name of a module, and @var{id} is any declared
15156 identifier within your program, except another module.
15158 Using the @code{::} operator makes @value{GDBN} search the scope
15159 specified by @var{scope} for the identifier @var{id}. If it is not
15160 found in the specified scope, then @value{GDBN} searches all scopes
15161 enclosing the one specified by @var{scope}.
15163 Using the @code{.} operator makes @value{GDBN} search the current scope for
15164 the identifier specified by @var{id} that was imported from the
15165 definition module specified by @var{module}. With this operator, it is
15166 an error if the identifier @var{id} was not imported from definition
15167 module @var{module}, or if @var{id} is not an identifier in
15171 @subsubsection @value{GDBN} and Modula-2
15173 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
15174 Five subcommands of @code{set print} and @code{show print} apply
15175 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
15176 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
15177 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
15178 analogue in Modula-2.
15180 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
15181 with any language, is not useful with Modula-2. Its
15182 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
15183 created in Modula-2 as they can in C or C@t{++}. However, because an
15184 address can be specified by an integral constant, the construct
15185 @samp{@{@var{type}@}@var{adrexp}} is still useful.
15187 @cindex @code{#} in Modula-2
15188 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
15189 interpreted as the beginning of a comment. Use @code{<>} instead.
15195 The extensions made to @value{GDBN} for Ada only support
15196 output from the @sc{gnu} Ada (GNAT) compiler.
15197 Other Ada compilers are not currently supported, and
15198 attempting to debug executables produced by them is most likely
15202 @cindex expressions in Ada
15204 * Ada Mode Intro:: General remarks on the Ada syntax
15205 and semantics supported by Ada mode
15207 * Omissions from Ada:: Restrictions on the Ada expression syntax.
15208 * Additions to Ada:: Extensions of the Ada expression syntax.
15209 * Stopping Before Main Program:: Debugging the program during elaboration.
15210 * Ada Exceptions:: Ada Exceptions
15211 * Ada Tasks:: Listing and setting breakpoints in tasks.
15212 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
15213 * Ravenscar Profile:: Tasking Support when using the Ravenscar
15215 * Ada Glitches:: Known peculiarities of Ada mode.
15218 @node Ada Mode Intro
15219 @subsubsection Introduction
15220 @cindex Ada mode, general
15222 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
15223 syntax, with some extensions.
15224 The philosophy behind the design of this subset is
15228 That @value{GDBN} should provide basic literals and access to operations for
15229 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
15230 leaving more sophisticated computations to subprograms written into the
15231 program (which therefore may be called from @value{GDBN}).
15234 That type safety and strict adherence to Ada language restrictions
15235 are not particularly important to the @value{GDBN} user.
15238 That brevity is important to the @value{GDBN} user.
15241 Thus, for brevity, the debugger acts as if all names declared in
15242 user-written packages are directly visible, even if they are not visible
15243 according to Ada rules, thus making it unnecessary to fully qualify most
15244 names with their packages, regardless of context. Where this causes
15245 ambiguity, @value{GDBN} asks the user's intent.
15247 The debugger will start in Ada mode if it detects an Ada main program.
15248 As for other languages, it will enter Ada mode when stopped in a program that
15249 was translated from an Ada source file.
15251 While in Ada mode, you may use `@t{--}' for comments. This is useful
15252 mostly for documenting command files. The standard @value{GDBN} comment
15253 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
15254 middle (to allow based literals).
15256 The debugger supports limited overloading. Given a subprogram call in which
15257 the function symbol has multiple definitions, it will use the number of
15258 actual parameters and some information about their types to attempt to narrow
15259 the set of definitions. It also makes very limited use of context, preferring
15260 procedures to functions in the context of the @code{call} command, and
15261 functions to procedures elsewhere.
15263 @node Omissions from Ada
15264 @subsubsection Omissions from Ada
15265 @cindex Ada, omissions from
15267 Here are the notable omissions from the subset:
15271 Only a subset of the attributes are supported:
15275 @t{'First}, @t{'Last}, and @t{'Length}
15276 on array objects (not on types and subtypes).
15279 @t{'Min} and @t{'Max}.
15282 @t{'Pos} and @t{'Val}.
15288 @t{'Range} on array objects (not subtypes), but only as the right
15289 operand of the membership (@code{in}) operator.
15292 @t{'Access}, @t{'Unchecked_Access}, and
15293 @t{'Unrestricted_Access} (a GNAT extension).
15301 @code{Characters.Latin_1} are not available and
15302 concatenation is not implemented. Thus, escape characters in strings are
15303 not currently available.
15306 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
15307 equality of representations. They will generally work correctly
15308 for strings and arrays whose elements have integer or enumeration types.
15309 They may not work correctly for arrays whose element
15310 types have user-defined equality, for arrays of real values
15311 (in particular, IEEE-conformant floating point, because of negative
15312 zeroes and NaNs), and for arrays whose elements contain unused bits with
15313 indeterminate values.
15316 The other component-by-component array operations (@code{and}, @code{or},
15317 @code{xor}, @code{not}, and relational tests other than equality)
15318 are not implemented.
15321 @cindex array aggregates (Ada)
15322 @cindex record aggregates (Ada)
15323 @cindex aggregates (Ada)
15324 There is limited support for array and record aggregates. They are
15325 permitted only on the right sides of assignments, as in these examples:
15328 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
15329 (@value{GDBP}) set An_Array := (1, others => 0)
15330 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
15331 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
15332 (@value{GDBP}) set A_Record := (1, "Peter", True);
15333 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
15337 discriminant's value by assigning an aggregate has an
15338 undefined effect if that discriminant is used within the record.
15339 However, you can first modify discriminants by directly assigning to
15340 them (which normally would not be allowed in Ada), and then performing an
15341 aggregate assignment. For example, given a variable @code{A_Rec}
15342 declared to have a type such as:
15345 type Rec (Len : Small_Integer := 0) is record
15347 Vals : IntArray (1 .. Len);
15351 you can assign a value with a different size of @code{Vals} with two
15355 (@value{GDBP}) set A_Rec.Len := 4
15356 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
15359 As this example also illustrates, @value{GDBN} is very loose about the usual
15360 rules concerning aggregates. You may leave out some of the
15361 components of an array or record aggregate (such as the @code{Len}
15362 component in the assignment to @code{A_Rec} above); they will retain their
15363 original values upon assignment. You may freely use dynamic values as
15364 indices in component associations. You may even use overlapping or
15365 redundant component associations, although which component values are
15366 assigned in such cases is not defined.
15369 Calls to dispatching subprograms are not implemented.
15372 The overloading algorithm is much more limited (i.e., less selective)
15373 than that of real Ada. It makes only limited use of the context in
15374 which a subexpression appears to resolve its meaning, and it is much
15375 looser in its rules for allowing type matches. As a result, some
15376 function calls will be ambiguous, and the user will be asked to choose
15377 the proper resolution.
15380 The @code{new} operator is not implemented.
15383 Entry calls are not implemented.
15386 Aside from printing, arithmetic operations on the native VAX floating-point
15387 formats are not supported.
15390 It is not possible to slice a packed array.
15393 The names @code{True} and @code{False}, when not part of a qualified name,
15394 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
15396 Should your program
15397 redefine these names in a package or procedure (at best a dubious practice),
15398 you will have to use fully qualified names to access their new definitions.
15401 @node Additions to Ada
15402 @subsubsection Additions to Ada
15403 @cindex Ada, deviations from
15405 As it does for other languages, @value{GDBN} makes certain generic
15406 extensions to Ada (@pxref{Expressions}):
15410 If the expression @var{E} is a variable residing in memory (typically
15411 a local variable or array element) and @var{N} is a positive integer,
15412 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
15413 @var{N}-1 adjacent variables following it in memory as an array. In
15414 Ada, this operator is generally not necessary, since its prime use is
15415 in displaying parts of an array, and slicing will usually do this in
15416 Ada. However, there are occasional uses when debugging programs in
15417 which certain debugging information has been optimized away.
15420 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
15421 appears in function or file @var{B}.'' When @var{B} is a file name,
15422 you must typically surround it in single quotes.
15425 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
15426 @var{type} that appears at address @var{addr}.''
15429 A name starting with @samp{$} is a convenience variable
15430 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
15433 In addition, @value{GDBN} provides a few other shortcuts and outright
15434 additions specific to Ada:
15438 The assignment statement is allowed as an expression, returning
15439 its right-hand operand as its value. Thus, you may enter
15442 (@value{GDBP}) set x := y + 3
15443 (@value{GDBP}) print A(tmp := y + 1)
15447 The semicolon is allowed as an ``operator,'' returning as its value
15448 the value of its right-hand operand.
15449 This allows, for example,
15450 complex conditional breaks:
15453 (@value{GDBP}) break f
15454 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
15458 Rather than use catenation and symbolic character names to introduce special
15459 characters into strings, one may instead use a special bracket notation,
15460 which is also used to print strings. A sequence of characters of the form
15461 @samp{["@var{XX}"]} within a string or character literal denotes the
15462 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
15463 sequence of characters @samp{["""]} also denotes a single quotation mark
15464 in strings. For example,
15466 "One line.["0a"]Next line.["0a"]"
15469 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
15473 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
15474 @t{'Max} is optional (and is ignored in any case). For example, it is valid
15478 (@value{GDBP}) print 'max(x, y)
15482 When printing arrays, @value{GDBN} uses positional notation when the
15483 array has a lower bound of 1, and uses a modified named notation otherwise.
15484 For example, a one-dimensional array of three integers with a lower bound
15485 of 3 might print as
15492 That is, in contrast to valid Ada, only the first component has a @code{=>}
15496 You may abbreviate attributes in expressions with any unique,
15497 multi-character subsequence of
15498 their names (an exact match gets preference).
15499 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
15500 in place of @t{a'length}.
15503 @cindex quoting Ada internal identifiers
15504 Since Ada is case-insensitive, the debugger normally maps identifiers you type
15505 to lower case. The GNAT compiler uses upper-case characters for
15506 some of its internal identifiers, which are normally of no interest to users.
15507 For the rare occasions when you actually have to look at them,
15508 enclose them in angle brackets to avoid the lower-case mapping.
15511 (@value{GDBP}) print <JMPBUF_SAVE>[0]
15515 Printing an object of class-wide type or dereferencing an
15516 access-to-class-wide value will display all the components of the object's
15517 specific type (as indicated by its run-time tag). Likewise, component
15518 selection on such a value will operate on the specific type of the
15523 @node Stopping Before Main Program
15524 @subsubsection Stopping at the Very Beginning
15526 @cindex breakpointing Ada elaboration code
15527 It is sometimes necessary to debug the program during elaboration, and
15528 before reaching the main procedure.
15529 As defined in the Ada Reference
15530 Manual, the elaboration code is invoked from a procedure called
15531 @code{adainit}. To run your program up to the beginning of
15532 elaboration, simply use the following two commands:
15533 @code{tbreak adainit} and @code{run}.
15535 @node Ada Exceptions
15536 @subsubsection Ada Exceptions
15538 A command is provided to list all Ada exceptions:
15541 @kindex info exceptions
15542 @item info exceptions
15543 @itemx info exceptions @var{regexp}
15544 The @code{info exceptions} command allows you to list all Ada exceptions
15545 defined within the program being debugged, as well as their addresses.
15546 With a regular expression, @var{regexp}, as argument, only those exceptions
15547 whose names match @var{regexp} are listed.
15550 Below is a small example, showing how the command can be used, first
15551 without argument, and next with a regular expression passed as an
15555 (@value{GDBP}) info exceptions
15556 All defined Ada exceptions:
15557 constraint_error: 0x613da0
15558 program_error: 0x613d20
15559 storage_error: 0x613ce0
15560 tasking_error: 0x613ca0
15561 const.aint_global_e: 0x613b00
15562 (@value{GDBP}) info exceptions const.aint
15563 All Ada exceptions matching regular expression "const.aint":
15564 constraint_error: 0x613da0
15565 const.aint_global_e: 0x613b00
15568 It is also possible to ask @value{GDBN} to stop your program's execution
15569 when an exception is raised. For more details, see @ref{Set Catchpoints}.
15572 @subsubsection Extensions for Ada Tasks
15573 @cindex Ada, tasking
15575 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
15576 @value{GDBN} provides the following task-related commands:
15581 This command shows a list of current Ada tasks, as in the following example:
15588 (@value{GDBP}) info tasks
15589 ID TID P-ID Pri State Name
15590 1 8088000 0 15 Child Activation Wait main_task
15591 2 80a4000 1 15 Accept Statement b
15592 3 809a800 1 15 Child Activation Wait a
15593 * 4 80ae800 3 15 Runnable c
15598 In this listing, the asterisk before the last task indicates it to be the
15599 task currently being inspected.
15603 Represents @value{GDBN}'s internal task number.
15609 The parent's task ID (@value{GDBN}'s internal task number).
15612 The base priority of the task.
15615 Current state of the task.
15619 The task has been created but has not been activated. It cannot be
15623 The task is not blocked for any reason known to Ada. (It may be waiting
15624 for a mutex, though.) It is conceptually "executing" in normal mode.
15627 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
15628 that were waiting on terminate alternatives have been awakened and have
15629 terminated themselves.
15631 @item Child Activation Wait
15632 The task is waiting for created tasks to complete activation.
15634 @item Accept Statement
15635 The task is waiting on an accept or selective wait statement.
15637 @item Waiting on entry call
15638 The task is waiting on an entry call.
15640 @item Async Select Wait
15641 The task is waiting to start the abortable part of an asynchronous
15645 The task is waiting on a select statement with only a delay
15648 @item Child Termination Wait
15649 The task is sleeping having completed a master within itself, and is
15650 waiting for the tasks dependent on that master to become terminated or
15651 waiting on a terminate Phase.
15653 @item Wait Child in Term Alt
15654 The task is sleeping waiting for tasks on terminate alternatives to
15655 finish terminating.
15657 @item Accepting RV with @var{taskno}
15658 The task is accepting a rendez-vous with the task @var{taskno}.
15662 Name of the task in the program.
15666 @kindex info task @var{taskno}
15667 @item info task @var{taskno}
15668 This command shows detailled informations on the specified task, as in
15669 the following example:
15674 (@value{GDBP}) info tasks
15675 ID TID P-ID Pri State Name
15676 1 8077880 0 15 Child Activation Wait main_task
15677 * 2 807c468 1 15 Runnable task_1
15678 (@value{GDBP}) info task 2
15679 Ada Task: 0x807c468
15682 Parent: 1 (main_task)
15688 @kindex task@r{ (Ada)}
15689 @cindex current Ada task ID
15690 This command prints the ID of the current task.
15696 (@value{GDBP}) info tasks
15697 ID TID P-ID Pri State Name
15698 1 8077870 0 15 Child Activation Wait main_task
15699 * 2 807c458 1 15 Runnable t
15700 (@value{GDBP}) task
15701 [Current task is 2]
15704 @item task @var{taskno}
15705 @cindex Ada task switching
15706 This command is like the @code{thread @var{threadno}}
15707 command (@pxref{Threads}). It switches the context of debugging
15708 from the current task to the given task.
15714 (@value{GDBP}) info tasks
15715 ID TID P-ID Pri State Name
15716 1 8077870 0 15 Child Activation Wait main_task
15717 * 2 807c458 1 15 Runnable t
15718 (@value{GDBP}) task 1
15719 [Switching to task 1]
15720 #0 0x8067726 in pthread_cond_wait ()
15722 #0 0x8067726 in pthread_cond_wait ()
15723 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
15724 #2 0x805cb63 in system.task_primitives.operations.sleep ()
15725 #3 0x806153e in system.tasking.stages.activate_tasks ()
15726 #4 0x804aacc in un () at un.adb:5
15729 @item break @var{linespec} task @var{taskno}
15730 @itemx break @var{linespec} task @var{taskno} if @dots{}
15731 @cindex breakpoints and tasks, in Ada
15732 @cindex task breakpoints, in Ada
15733 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
15734 These commands are like the @code{break @dots{} thread @dots{}}
15735 command (@pxref{Thread Stops}). The
15736 @var{linespec} argument specifies source lines, as described
15737 in @ref{Specify Location}.
15739 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
15740 to specify that you only want @value{GDBN} to stop the program when a
15741 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
15742 numeric task identifiers assigned by @value{GDBN}, shown in the first
15743 column of the @samp{info tasks} display.
15745 If you do not specify @samp{task @var{taskno}} when you set a
15746 breakpoint, the breakpoint applies to @emph{all} tasks of your
15749 You can use the @code{task} qualifier on conditional breakpoints as
15750 well; in this case, place @samp{task @var{taskno}} before the
15751 breakpoint condition (before the @code{if}).
15759 (@value{GDBP}) info tasks
15760 ID TID P-ID Pri State Name
15761 1 140022020 0 15 Child Activation Wait main_task
15762 2 140045060 1 15 Accept/Select Wait t2
15763 3 140044840 1 15 Runnable t1
15764 * 4 140056040 1 15 Runnable t3
15765 (@value{GDBP}) b 15 task 2
15766 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
15767 (@value{GDBP}) cont
15772 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
15774 (@value{GDBP}) info tasks
15775 ID TID P-ID Pri State Name
15776 1 140022020 0 15 Child Activation Wait main_task
15777 * 2 140045060 1 15 Runnable t2
15778 3 140044840 1 15 Runnable t1
15779 4 140056040 1 15 Delay Sleep t3
15783 @node Ada Tasks and Core Files
15784 @subsubsection Tasking Support when Debugging Core Files
15785 @cindex Ada tasking and core file debugging
15787 When inspecting a core file, as opposed to debugging a live program,
15788 tasking support may be limited or even unavailable, depending on
15789 the platform being used.
15790 For instance, on x86-linux, the list of tasks is available, but task
15791 switching is not supported.
15793 On certain platforms, the debugger needs to perform some
15794 memory writes in order to provide Ada tasking support. When inspecting
15795 a core file, this means that the core file must be opened with read-write
15796 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
15797 Under these circumstances, you should make a backup copy of the core
15798 file before inspecting it with @value{GDBN}.
15800 @node Ravenscar Profile
15801 @subsubsection Tasking Support when using the Ravenscar Profile
15802 @cindex Ravenscar Profile
15804 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
15805 specifically designed for systems with safety-critical real-time
15809 @kindex set ravenscar task-switching on
15810 @cindex task switching with program using Ravenscar Profile
15811 @item set ravenscar task-switching on
15812 Allows task switching when debugging a program that uses the Ravenscar
15813 Profile. This is the default.
15815 @kindex set ravenscar task-switching off
15816 @item set ravenscar task-switching off
15817 Turn off task switching when debugging a program that uses the Ravenscar
15818 Profile. This is mostly intended to disable the code that adds support
15819 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
15820 the Ravenscar runtime is preventing @value{GDBN} from working properly.
15821 To be effective, this command should be run before the program is started.
15823 @kindex show ravenscar task-switching
15824 @item show ravenscar task-switching
15825 Show whether it is possible to switch from task to task in a program
15826 using the Ravenscar Profile.
15831 @subsubsection Known Peculiarities of Ada Mode
15832 @cindex Ada, problems
15834 Besides the omissions listed previously (@pxref{Omissions from Ada}),
15835 we know of several problems with and limitations of Ada mode in
15837 some of which will be fixed with planned future releases of the debugger
15838 and the GNU Ada compiler.
15842 Static constants that the compiler chooses not to materialize as objects in
15843 storage are invisible to the debugger.
15846 Named parameter associations in function argument lists are ignored (the
15847 argument lists are treated as positional).
15850 Many useful library packages are currently invisible to the debugger.
15853 Fixed-point arithmetic, conversions, input, and output is carried out using
15854 floating-point arithmetic, and may give results that only approximate those on
15858 The GNAT compiler never generates the prefix @code{Standard} for any of
15859 the standard symbols defined by the Ada language. @value{GDBN} knows about
15860 this: it will strip the prefix from names when you use it, and will never
15861 look for a name you have so qualified among local symbols, nor match against
15862 symbols in other packages or subprograms. If you have
15863 defined entities anywhere in your program other than parameters and
15864 local variables whose simple names match names in @code{Standard},
15865 GNAT's lack of qualification here can cause confusion. When this happens,
15866 you can usually resolve the confusion
15867 by qualifying the problematic names with package
15868 @code{Standard} explicitly.
15871 Older versions of the compiler sometimes generate erroneous debugging
15872 information, resulting in the debugger incorrectly printing the value
15873 of affected entities. In some cases, the debugger is able to work
15874 around an issue automatically. In other cases, the debugger is able
15875 to work around the issue, but the work-around has to be specifically
15878 @kindex set ada trust-PAD-over-XVS
15879 @kindex show ada trust-PAD-over-XVS
15882 @item set ada trust-PAD-over-XVS on
15883 Configure GDB to strictly follow the GNAT encoding when computing the
15884 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
15885 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
15886 a complete description of the encoding used by the GNAT compiler).
15887 This is the default.
15889 @item set ada trust-PAD-over-XVS off
15890 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
15891 sometimes prints the wrong value for certain entities, changing @code{ada
15892 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
15893 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
15894 @code{off}, but this incurs a slight performance penalty, so it is
15895 recommended to leave this setting to @code{on} unless necessary.
15899 @cindex GNAT descriptive types
15900 @cindex GNAT encoding
15901 Internally, the debugger also relies on the compiler following a number
15902 of conventions known as the @samp{GNAT Encoding}, all documented in
15903 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
15904 how the debugging information should be generated for certain types.
15905 In particular, this convention makes use of @dfn{descriptive types},
15906 which are artificial types generated purely to help the debugger.
15908 These encodings were defined at a time when the debugging information
15909 format used was not powerful enough to describe some of the more complex
15910 types available in Ada. Since DWARF allows us to express nearly all
15911 Ada features, the long-term goal is to slowly replace these descriptive
15912 types by their pure DWARF equivalent. To facilitate that transition,
15913 a new maintenance option is available to force the debugger to ignore
15914 those descriptive types. It allows the user to quickly evaluate how
15915 well @value{GDBN} works without them.
15919 @kindex maint ada set ignore-descriptive-types
15920 @item maintenance ada set ignore-descriptive-types [on|off]
15921 Control whether the debugger should ignore descriptive types.
15922 The default is not to ignore descriptives types (@code{off}).
15924 @kindex maint ada show ignore-descriptive-types
15925 @item maintenance ada show ignore-descriptive-types
15926 Show if descriptive types are ignored by @value{GDBN}.
15930 @node Unsupported Languages
15931 @section Unsupported Languages
15933 @cindex unsupported languages
15934 @cindex minimal language
15935 In addition to the other fully-supported programming languages,
15936 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
15937 It does not represent a real programming language, but provides a set
15938 of capabilities close to what the C or assembly languages provide.
15939 This should allow most simple operations to be performed while debugging
15940 an application that uses a language currently not supported by @value{GDBN}.
15942 If the language is set to @code{auto}, @value{GDBN} will automatically
15943 select this language if the current frame corresponds to an unsupported
15947 @chapter Examining the Symbol Table
15949 The commands described in this chapter allow you to inquire about the
15950 symbols (names of variables, functions and types) defined in your
15951 program. This information is inherent in the text of your program and
15952 does not change as your program executes. @value{GDBN} finds it in your
15953 program's symbol table, in the file indicated when you started @value{GDBN}
15954 (@pxref{File Options, ,Choosing Files}), or by one of the
15955 file-management commands (@pxref{Files, ,Commands to Specify Files}).
15957 @cindex symbol names
15958 @cindex names of symbols
15959 @cindex quoting names
15960 Occasionally, you may need to refer to symbols that contain unusual
15961 characters, which @value{GDBN} ordinarily treats as word delimiters. The
15962 most frequent case is in referring to static variables in other
15963 source files (@pxref{Variables,,Program Variables}). File names
15964 are recorded in object files as debugging symbols, but @value{GDBN} would
15965 ordinarily parse a typical file name, like @file{foo.c}, as the three words
15966 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
15967 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
15974 looks up the value of @code{x} in the scope of the file @file{foo.c}.
15977 @cindex case-insensitive symbol names
15978 @cindex case sensitivity in symbol names
15979 @kindex set case-sensitive
15980 @item set case-sensitive on
15981 @itemx set case-sensitive off
15982 @itemx set case-sensitive auto
15983 Normally, when @value{GDBN} looks up symbols, it matches their names
15984 with case sensitivity determined by the current source language.
15985 Occasionally, you may wish to control that. The command @code{set
15986 case-sensitive} lets you do that by specifying @code{on} for
15987 case-sensitive matches or @code{off} for case-insensitive ones. If
15988 you specify @code{auto}, case sensitivity is reset to the default
15989 suitable for the source language. The default is case-sensitive
15990 matches for all languages except for Fortran, for which the default is
15991 case-insensitive matches.
15993 @kindex show case-sensitive
15994 @item show case-sensitive
15995 This command shows the current setting of case sensitivity for symbols
15998 @kindex set print type methods
15999 @item set print type methods
16000 @itemx set print type methods on
16001 @itemx set print type methods off
16002 Normally, when @value{GDBN} prints a class, it displays any methods
16003 declared in that class. You can control this behavior either by
16004 passing the appropriate flag to @code{ptype}, or using @command{set
16005 print type methods}. Specifying @code{on} will cause @value{GDBN} to
16006 display the methods; this is the default. Specifying @code{off} will
16007 cause @value{GDBN} to omit the methods.
16009 @kindex show print type methods
16010 @item show print type methods
16011 This command shows the current setting of method display when printing
16014 @kindex set print type typedefs
16015 @item set print type typedefs
16016 @itemx set print type typedefs on
16017 @itemx set print type typedefs off
16019 Normally, when @value{GDBN} prints a class, it displays any typedefs
16020 defined in that class. You can control this behavior either by
16021 passing the appropriate flag to @code{ptype}, or using @command{set
16022 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
16023 display the typedef definitions; this is the default. Specifying
16024 @code{off} will cause @value{GDBN} to omit the typedef definitions.
16025 Note that this controls whether the typedef definition itself is
16026 printed, not whether typedef names are substituted when printing other
16029 @kindex show print type typedefs
16030 @item show print type typedefs
16031 This command shows the current setting of typedef display when
16034 @kindex info address
16035 @cindex address of a symbol
16036 @item info address @var{symbol}
16037 Describe where the data for @var{symbol} is stored. For a register
16038 variable, this says which register it is kept in. For a non-register
16039 local variable, this prints the stack-frame offset at which the variable
16042 Note the contrast with @samp{print &@var{symbol}}, which does not work
16043 at all for a register variable, and for a stack local variable prints
16044 the exact address of the current instantiation of the variable.
16046 @kindex info symbol
16047 @cindex symbol from address
16048 @cindex closest symbol and offset for an address
16049 @item info symbol @var{addr}
16050 Print the name of a symbol which is stored at the address @var{addr}.
16051 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
16052 nearest symbol and an offset from it:
16055 (@value{GDBP}) info symbol 0x54320
16056 _initialize_vx + 396 in section .text
16060 This is the opposite of the @code{info address} command. You can use
16061 it to find out the name of a variable or a function given its address.
16063 For dynamically linked executables, the name of executable or shared
16064 library containing the symbol is also printed:
16067 (@value{GDBP}) info symbol 0x400225
16068 _start + 5 in section .text of /tmp/a.out
16069 (@value{GDBP}) info symbol 0x2aaaac2811cf
16070 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
16074 @item whatis[/@var{flags}] [@var{arg}]
16075 Print the data type of @var{arg}, which can be either an expression
16076 or a name of a data type. With no argument, print the data type of
16077 @code{$}, the last value in the value history.
16079 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
16080 is not actually evaluated, and any side-effecting operations (such as
16081 assignments or function calls) inside it do not take place.
16083 If @var{arg} is a variable or an expression, @code{whatis} prints its
16084 literal type as it is used in the source code. If the type was
16085 defined using a @code{typedef}, @code{whatis} will @emph{not} print
16086 the data type underlying the @code{typedef}. If the type of the
16087 variable or the expression is a compound data type, such as
16088 @code{struct} or @code{class}, @code{whatis} never prints their
16089 fields or methods. It just prints the @code{struct}/@code{class}
16090 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
16091 such a compound data type, use @code{ptype}.
16093 If @var{arg} is a type name that was defined using @code{typedef},
16094 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
16095 Unrolling means that @code{whatis} will show the underlying type used
16096 in the @code{typedef} declaration of @var{arg}. However, if that
16097 underlying type is also a @code{typedef}, @code{whatis} will not
16100 For C code, the type names may also have the form @samp{class
16101 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
16102 @var{union-tag}} or @samp{enum @var{enum-tag}}.
16104 @var{flags} can be used to modify how the type is displayed.
16105 Available flags are:
16109 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
16110 parameters and typedefs defined in a class when printing the class'
16111 members. The @code{/r} flag disables this.
16114 Do not print methods defined in the class.
16117 Print methods defined in the class. This is the default, but the flag
16118 exists in case you change the default with @command{set print type methods}.
16121 Do not print typedefs defined in the class. Note that this controls
16122 whether the typedef definition itself is printed, not whether typedef
16123 names are substituted when printing other types.
16126 Print typedefs defined in the class. This is the default, but the flag
16127 exists in case you change the default with @command{set print type typedefs}.
16131 @item ptype[/@var{flags}] [@var{arg}]
16132 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
16133 detailed description of the type, instead of just the name of the type.
16134 @xref{Expressions, ,Expressions}.
16136 Contrary to @code{whatis}, @code{ptype} always unrolls any
16137 @code{typedef}s in its argument declaration, whether the argument is
16138 a variable, expression, or a data type. This means that @code{ptype}
16139 of a variable or an expression will not print literally its type as
16140 present in the source code---use @code{whatis} for that. @code{typedef}s at
16141 the pointer or reference targets are also unrolled. Only @code{typedef}s of
16142 fields, methods and inner @code{class typedef}s of @code{struct}s,
16143 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
16145 For example, for this variable declaration:
16148 typedef double real_t;
16149 struct complex @{ real_t real; double imag; @};
16150 typedef struct complex complex_t;
16152 real_t *real_pointer_var;
16156 the two commands give this output:
16160 (@value{GDBP}) whatis var
16162 (@value{GDBP}) ptype var
16163 type = struct complex @{
16167 (@value{GDBP}) whatis complex_t
16168 type = struct complex
16169 (@value{GDBP}) whatis struct complex
16170 type = struct complex
16171 (@value{GDBP}) ptype struct complex
16172 type = struct complex @{
16176 (@value{GDBP}) whatis real_pointer_var
16178 (@value{GDBP}) ptype real_pointer_var
16184 As with @code{whatis}, using @code{ptype} without an argument refers to
16185 the type of @code{$}, the last value in the value history.
16187 @cindex incomplete type
16188 Sometimes, programs use opaque data types or incomplete specifications
16189 of complex data structure. If the debug information included in the
16190 program does not allow @value{GDBN} to display a full declaration of
16191 the data type, it will say @samp{<incomplete type>}. For example,
16192 given these declarations:
16196 struct foo *fooptr;
16200 but no definition for @code{struct foo} itself, @value{GDBN} will say:
16203 (@value{GDBP}) ptype foo
16204 $1 = <incomplete type>
16208 ``Incomplete type'' is C terminology for data types that are not
16209 completely specified.
16212 @item info types @var{regexp}
16214 Print a brief description of all types whose names match the regular
16215 expression @var{regexp} (or all types in your program, if you supply
16216 no argument). Each complete typename is matched as though it were a
16217 complete line; thus, @samp{i type value} gives information on all
16218 types in your program whose names include the string @code{value}, but
16219 @samp{i type ^value$} gives information only on types whose complete
16220 name is @code{value}.
16222 This command differs from @code{ptype} in two ways: first, like
16223 @code{whatis}, it does not print a detailed description; second, it
16224 lists all source files where a type is defined.
16226 @kindex info type-printers
16227 @item info type-printers
16228 Versions of @value{GDBN} that ship with Python scripting enabled may
16229 have ``type printers'' available. When using @command{ptype} or
16230 @command{whatis}, these printers are consulted when the name of a type
16231 is needed. @xref{Type Printing API}, for more information on writing
16234 @code{info type-printers} displays all the available type printers.
16236 @kindex enable type-printer
16237 @kindex disable type-printer
16238 @item enable type-printer @var{name}@dots{}
16239 @item disable type-printer @var{name}@dots{}
16240 These commands can be used to enable or disable type printers.
16243 @cindex local variables
16244 @item info scope @var{location}
16245 List all the variables local to a particular scope. This command
16246 accepts a @var{location} argument---a function name, a source line, or
16247 an address preceded by a @samp{*}, and prints all the variables local
16248 to the scope defined by that location. (@xref{Specify Location}, for
16249 details about supported forms of @var{location}.) For example:
16252 (@value{GDBP}) @b{info scope command_line_handler}
16253 Scope for command_line_handler:
16254 Symbol rl is an argument at stack/frame offset 8, length 4.
16255 Symbol linebuffer is in static storage at address 0x150a18, length 4.
16256 Symbol linelength is in static storage at address 0x150a1c, length 4.
16257 Symbol p is a local variable in register $esi, length 4.
16258 Symbol p1 is a local variable in register $ebx, length 4.
16259 Symbol nline is a local variable in register $edx, length 4.
16260 Symbol repeat is a local variable at frame offset -8, length 4.
16264 This command is especially useful for determining what data to collect
16265 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
16268 @kindex info source
16270 Show information about the current source file---that is, the source file for
16271 the function containing the current point of execution:
16274 the name of the source file, and the directory containing it,
16276 the directory it was compiled in,
16278 its length, in lines,
16280 which programming language it is written in,
16282 whether the executable includes debugging information for that file, and
16283 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
16285 whether the debugging information includes information about
16286 preprocessor macros.
16290 @kindex info sources
16292 Print the names of all source files in your program for which there is
16293 debugging information, organized into two lists: files whose symbols
16294 have already been read, and files whose symbols will be read when needed.
16296 @kindex info functions
16297 @item info functions
16298 Print the names and data types of all defined functions.
16300 @item info functions @var{regexp}
16301 Print the names and data types of all defined functions
16302 whose names contain a match for regular expression @var{regexp}.
16303 Thus, @samp{info fun step} finds all functions whose names
16304 include @code{step}; @samp{info fun ^step} finds those whose names
16305 start with @code{step}. If a function name contains characters
16306 that conflict with the regular expression language (e.g.@:
16307 @samp{operator*()}), they may be quoted with a backslash.
16309 @kindex info variables
16310 @item info variables
16311 Print the names and data types of all variables that are defined
16312 outside of functions (i.e.@: excluding local variables).
16314 @item info variables @var{regexp}
16315 Print the names and data types of all variables (except for local
16316 variables) whose names contain a match for regular expression
16319 @kindex info classes
16320 @cindex Objective-C, classes and selectors
16322 @itemx info classes @var{regexp}
16323 Display all Objective-C classes in your program, or
16324 (with the @var{regexp} argument) all those matching a particular regular
16327 @kindex info selectors
16328 @item info selectors
16329 @itemx info selectors @var{regexp}
16330 Display all Objective-C selectors in your program, or
16331 (with the @var{regexp} argument) all those matching a particular regular
16335 This was never implemented.
16336 @kindex info methods
16338 @itemx info methods @var{regexp}
16339 The @code{info methods} command permits the user to examine all defined
16340 methods within C@t{++} program, or (with the @var{regexp} argument) a
16341 specific set of methods found in the various C@t{++} classes. Many
16342 C@t{++} classes provide a large number of methods. Thus, the output
16343 from the @code{ptype} command can be overwhelming and hard to use. The
16344 @code{info-methods} command filters the methods, printing only those
16345 which match the regular-expression @var{regexp}.
16348 @cindex opaque data types
16349 @kindex set opaque-type-resolution
16350 @item set opaque-type-resolution on
16351 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
16352 declared as a pointer to a @code{struct}, @code{class}, or
16353 @code{union}---for example, @code{struct MyType *}---that is used in one
16354 source file although the full declaration of @code{struct MyType} is in
16355 another source file. The default is on.
16357 A change in the setting of this subcommand will not take effect until
16358 the next time symbols for a file are loaded.
16360 @item set opaque-type-resolution off
16361 Tell @value{GDBN} not to resolve opaque types. In this case, the type
16362 is printed as follows:
16364 @{<no data fields>@}
16367 @kindex show opaque-type-resolution
16368 @item show opaque-type-resolution
16369 Show whether opaque types are resolved or not.
16371 @kindex set print symbol-loading
16372 @cindex print messages when symbols are loaded
16373 @item set print symbol-loading
16374 @itemx set print symbol-loading full
16375 @itemx set print symbol-loading brief
16376 @itemx set print symbol-loading off
16377 The @code{set print symbol-loading} command allows you to control the
16378 printing of messages when @value{GDBN} loads symbol information.
16379 By default a message is printed for the executable and one for each
16380 shared library, and normally this is what you want. However, when
16381 debugging apps with large numbers of shared libraries these messages
16383 When set to @code{brief} a message is printed for each executable,
16384 and when @value{GDBN} loads a collection of shared libraries at once
16385 it will only print one message regardless of the number of shared
16386 libraries. When set to @code{off} no messages are printed.
16388 @kindex show print symbol-loading
16389 @item show print symbol-loading
16390 Show whether messages will be printed when a @value{GDBN} command
16391 entered from the keyboard causes symbol information to be loaded.
16393 @kindex maint print symbols
16394 @cindex symbol dump
16395 @kindex maint print psymbols
16396 @cindex partial symbol dump
16397 @kindex maint print msymbols
16398 @cindex minimal symbol dump
16399 @item maint print symbols @var{filename}
16400 @itemx maint print psymbols @var{filename}
16401 @itemx maint print msymbols @var{filename}
16402 Write a dump of debugging symbol data into the file @var{filename}.
16403 These commands are used to debug the @value{GDBN} symbol-reading code. Only
16404 symbols with debugging data are included. If you use @samp{maint print
16405 symbols}, @value{GDBN} includes all the symbols for which it has already
16406 collected full details: that is, @var{filename} reflects symbols for
16407 only those files whose symbols @value{GDBN} has read. You can use the
16408 command @code{info sources} to find out which files these are. If you
16409 use @samp{maint print psymbols} instead, the dump shows information about
16410 symbols that @value{GDBN} only knows partially---that is, symbols defined in
16411 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
16412 @samp{maint print msymbols} dumps just the minimal symbol information
16413 required for each object file from which @value{GDBN} has read some symbols.
16414 @xref{Files, ,Commands to Specify Files}, for a discussion of how
16415 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
16417 @kindex maint info symtabs
16418 @kindex maint info psymtabs
16419 @cindex listing @value{GDBN}'s internal symbol tables
16420 @cindex symbol tables, listing @value{GDBN}'s internal
16421 @cindex full symbol tables, listing @value{GDBN}'s internal
16422 @cindex partial symbol tables, listing @value{GDBN}'s internal
16423 @item maint info symtabs @r{[} @var{regexp} @r{]}
16424 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
16426 List the @code{struct symtab} or @code{struct partial_symtab}
16427 structures whose names match @var{regexp}. If @var{regexp} is not
16428 given, list them all. The output includes expressions which you can
16429 copy into a @value{GDBN} debugging this one to examine a particular
16430 structure in more detail. For example:
16433 (@value{GDBP}) maint info psymtabs dwarf2read
16434 @{ objfile /home/gnu/build/gdb/gdb
16435 ((struct objfile *) 0x82e69d0)
16436 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
16437 ((struct partial_symtab *) 0x8474b10)
16440 text addresses 0x814d3c8 -- 0x8158074
16441 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
16442 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
16443 dependencies (none)
16446 (@value{GDBP}) maint info symtabs
16450 We see that there is one partial symbol table whose filename contains
16451 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
16452 and we see that @value{GDBN} has not read in any symtabs yet at all.
16453 If we set a breakpoint on a function, that will cause @value{GDBN} to
16454 read the symtab for the compilation unit containing that function:
16457 (@value{GDBP}) break dwarf2_psymtab_to_symtab
16458 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
16460 (@value{GDBP}) maint info symtabs
16461 @{ objfile /home/gnu/build/gdb/gdb
16462 ((struct objfile *) 0x82e69d0)
16463 @{ symtab /home/gnu/src/gdb/dwarf2read.c
16464 ((struct symtab *) 0x86c1f38)
16467 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
16468 linetable ((struct linetable *) 0x8370fa0)
16469 debugformat DWARF 2
16478 @chapter Altering Execution
16480 Once you think you have found an error in your program, you might want to
16481 find out for certain whether correcting the apparent error would lead to
16482 correct results in the rest of the run. You can find the answer by
16483 experiment, using the @value{GDBN} features for altering execution of the
16486 For example, you can store new values into variables or memory
16487 locations, give your program a signal, restart it at a different
16488 address, or even return prematurely from a function.
16491 * Assignment:: Assignment to variables
16492 * Jumping:: Continuing at a different address
16493 * Signaling:: Giving your program a signal
16494 * Returning:: Returning from a function
16495 * Calling:: Calling your program's functions
16496 * Patching:: Patching your program
16497 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
16501 @section Assignment to Variables
16504 @cindex setting variables
16505 To alter the value of a variable, evaluate an assignment expression.
16506 @xref{Expressions, ,Expressions}. For example,
16513 stores the value 4 into the variable @code{x}, and then prints the
16514 value of the assignment expression (which is 4).
16515 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
16516 information on operators in supported languages.
16518 @kindex set variable
16519 @cindex variables, setting
16520 If you are not interested in seeing the value of the assignment, use the
16521 @code{set} command instead of the @code{print} command. @code{set} is
16522 really the same as @code{print} except that the expression's value is
16523 not printed and is not put in the value history (@pxref{Value History,
16524 ,Value History}). The expression is evaluated only for its effects.
16526 If the beginning of the argument string of the @code{set} command
16527 appears identical to a @code{set} subcommand, use the @code{set
16528 variable} command instead of just @code{set}. This command is identical
16529 to @code{set} except for its lack of subcommands. For example, if your
16530 program has a variable @code{width}, you get an error if you try to set
16531 a new value with just @samp{set width=13}, because @value{GDBN} has the
16532 command @code{set width}:
16535 (@value{GDBP}) whatis width
16537 (@value{GDBP}) p width
16539 (@value{GDBP}) set width=47
16540 Invalid syntax in expression.
16544 The invalid expression, of course, is @samp{=47}. In
16545 order to actually set the program's variable @code{width}, use
16548 (@value{GDBP}) set var width=47
16551 Because the @code{set} command has many subcommands that can conflict
16552 with the names of program variables, it is a good idea to use the
16553 @code{set variable} command instead of just @code{set}. For example, if
16554 your program has a variable @code{g}, you run into problems if you try
16555 to set a new value with just @samp{set g=4}, because @value{GDBN} has
16556 the command @code{set gnutarget}, abbreviated @code{set g}:
16560 (@value{GDBP}) whatis g
16564 (@value{GDBP}) set g=4
16568 The program being debugged has been started already.
16569 Start it from the beginning? (y or n) y
16570 Starting program: /home/smith/cc_progs/a.out
16571 "/home/smith/cc_progs/a.out": can't open to read symbols:
16572 Invalid bfd target.
16573 (@value{GDBP}) show g
16574 The current BFD target is "=4".
16579 The program variable @code{g} did not change, and you silently set the
16580 @code{gnutarget} to an invalid value. In order to set the variable
16584 (@value{GDBP}) set var g=4
16587 @value{GDBN} allows more implicit conversions in assignments than C; you can
16588 freely store an integer value into a pointer variable or vice versa,
16589 and you can convert any structure to any other structure that is the
16590 same length or shorter.
16591 @comment FIXME: how do structs align/pad in these conversions?
16592 @comment /doc@cygnus.com 18dec1990
16594 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
16595 construct to generate a value of specified type at a specified address
16596 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
16597 to memory location @code{0x83040} as an integer (which implies a certain size
16598 and representation in memory), and
16601 set @{int@}0x83040 = 4
16605 stores the value 4 into that memory location.
16608 @section Continuing at a Different Address
16610 Ordinarily, when you continue your program, you do so at the place where
16611 it stopped, with the @code{continue} command. You can instead continue at
16612 an address of your own choosing, with the following commands:
16616 @kindex j @r{(@code{jump})}
16617 @item jump @var{linespec}
16618 @itemx j @var{linespec}
16619 @itemx jump @var{location}
16620 @itemx j @var{location}
16621 Resume execution at line @var{linespec} or at address given by
16622 @var{location}. Execution stops again immediately if there is a
16623 breakpoint there. @xref{Specify Location}, for a description of the
16624 different forms of @var{linespec} and @var{location}. It is common
16625 practice to use the @code{tbreak} command in conjunction with
16626 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
16628 The @code{jump} command does not change the current stack frame, or
16629 the stack pointer, or the contents of any memory location or any
16630 register other than the program counter. If line @var{linespec} is in
16631 a different function from the one currently executing, the results may
16632 be bizarre if the two functions expect different patterns of arguments or
16633 of local variables. For this reason, the @code{jump} command requests
16634 confirmation if the specified line is not in the function currently
16635 executing. However, even bizarre results are predictable if you are
16636 well acquainted with the machine-language code of your program.
16639 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
16640 On many systems, you can get much the same effect as the @code{jump}
16641 command by storing a new value into the register @code{$pc}. The
16642 difference is that this does not start your program running; it only
16643 changes the address of where it @emph{will} run when you continue. For
16651 makes the next @code{continue} command or stepping command execute at
16652 address @code{0x485}, rather than at the address where your program stopped.
16653 @xref{Continuing and Stepping, ,Continuing and Stepping}.
16655 The most common occasion to use the @code{jump} command is to back
16656 up---perhaps with more breakpoints set---over a portion of a program
16657 that has already executed, in order to examine its execution in more
16662 @section Giving your Program a Signal
16663 @cindex deliver a signal to a program
16667 @item signal @var{signal}
16668 Resume execution where your program is stopped, but immediately give it the
16669 signal @var{signal}. The @var{signal} can be the name or the number of a
16670 signal. For example, on many systems @code{signal 2} and @code{signal
16671 SIGINT} are both ways of sending an interrupt signal.
16673 Alternatively, if @var{signal} is zero, continue execution without
16674 giving a signal. This is useful when your program stopped on account of
16675 a signal and would ordinarily see the signal when resumed with the
16676 @code{continue} command; @samp{signal 0} causes it to resume without a
16679 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
16680 delivered to the currently selected thread, not the thread that last
16681 reported a stop. This includes the situation where a thread was
16682 stopped due to a signal. So if you want to continue execution
16683 suppressing the signal that stopped a thread, you should select that
16684 same thread before issuing the @samp{signal 0} command. If you issue
16685 the @samp{signal 0} command with another thread as the selected one,
16686 @value{GDBN} detects that and asks for confirmation.
16688 Invoking the @code{signal} command is not the same as invoking the
16689 @code{kill} utility from the shell. Sending a signal with @code{kill}
16690 causes @value{GDBN} to decide what to do with the signal depending on
16691 the signal handling tables (@pxref{Signals}). The @code{signal} command
16692 passes the signal directly to your program.
16694 @code{signal} does not repeat when you press @key{RET} a second time
16695 after executing the command.
16697 @kindex queue-signal
16698 @item queue-signal @var{signal}
16699 Queue @var{signal} to be delivered immediately to the current thread
16700 when execution of the thread resumes. The @var{signal} can be the name or
16701 the number of a signal. For example, on many systems @code{signal 2} and
16702 @code{signal SIGINT} are both ways of sending an interrupt signal.
16703 The handling of the signal must be set to pass the signal to the program,
16704 otherwise @value{GDBN} will report an error.
16705 You can control the handling of signals from @value{GDBN} with the
16706 @code{handle} command (@pxref{Signals}).
16708 Alternatively, if @var{signal} is zero, any currently queued signal
16709 for the current thread is discarded and when execution resumes no signal
16710 will be delivered. This is useful when your program stopped on account
16711 of a signal and would ordinarily see the signal when resumed with the
16712 @code{continue} command.
16714 This command differs from the @code{signal} command in that the signal
16715 is just queued, execution is not resumed. And @code{queue-signal} cannot
16716 be used to pass a signal whose handling state has been set to @code{nopass}
16721 @xref{stepping into signal handlers}, for information on how stepping
16722 commands behave when the thread has a signal queued.
16725 @section Returning from a Function
16728 @cindex returning from a function
16731 @itemx return @var{expression}
16732 You can cancel execution of a function call with the @code{return}
16733 command. If you give an
16734 @var{expression} argument, its value is used as the function's return
16738 When you use @code{return}, @value{GDBN} discards the selected stack frame
16739 (and all frames within it). You can think of this as making the
16740 discarded frame return prematurely. If you wish to specify a value to
16741 be returned, give that value as the argument to @code{return}.
16743 This pops the selected stack frame (@pxref{Selection, ,Selecting a
16744 Frame}), and any other frames inside of it, leaving its caller as the
16745 innermost remaining frame. That frame becomes selected. The
16746 specified value is stored in the registers used for returning values
16749 The @code{return} command does not resume execution; it leaves the
16750 program stopped in the state that would exist if the function had just
16751 returned. In contrast, the @code{finish} command (@pxref{Continuing
16752 and Stepping, ,Continuing and Stepping}) resumes execution until the
16753 selected stack frame returns naturally.
16755 @value{GDBN} needs to know how the @var{expression} argument should be set for
16756 the inferior. The concrete registers assignment depends on the OS ABI and the
16757 type being returned by the selected stack frame. For example it is common for
16758 OS ABI to return floating point values in FPU registers while integer values in
16759 CPU registers. Still some ABIs return even floating point values in CPU
16760 registers. Larger integer widths (such as @code{long long int}) also have
16761 specific placement rules. @value{GDBN} already knows the OS ABI from its
16762 current target so it needs to find out also the type being returned to make the
16763 assignment into the right register(s).
16765 Normally, the selected stack frame has debug info. @value{GDBN} will always
16766 use the debug info instead of the implicit type of @var{expression} when the
16767 debug info is available. For example, if you type @kbd{return -1}, and the
16768 function in the current stack frame is declared to return a @code{long long
16769 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
16770 into a @code{long long int}:
16773 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
16775 (@value{GDBP}) return -1
16776 Make func return now? (y or n) y
16777 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
16778 43 printf ("result=%lld\n", func ());
16782 However, if the selected stack frame does not have a debug info, e.g., if the
16783 function was compiled without debug info, @value{GDBN} has to find out the type
16784 to return from user. Specifying a different type by mistake may set the value
16785 in different inferior registers than the caller code expects. For example,
16786 typing @kbd{return -1} with its implicit type @code{int} would set only a part
16787 of a @code{long long int} result for a debug info less function (on 32-bit
16788 architectures). Therefore the user is required to specify the return type by
16789 an appropriate cast explicitly:
16792 Breakpoint 2, 0x0040050b in func ()
16793 (@value{GDBP}) return -1
16794 Return value type not available for selected stack frame.
16795 Please use an explicit cast of the value to return.
16796 (@value{GDBP}) return (long long int) -1
16797 Make selected stack frame return now? (y or n) y
16798 #0 0x00400526 in main ()
16803 @section Calling Program Functions
16806 @cindex calling functions
16807 @cindex inferior functions, calling
16808 @item print @var{expr}
16809 Evaluate the expression @var{expr} and display the resulting value.
16810 The expression may include calls to functions in the program being
16814 @item call @var{expr}
16815 Evaluate the expression @var{expr} without displaying @code{void}
16818 You can use this variant of the @code{print} command if you want to
16819 execute a function from your program that does not return anything
16820 (a.k.a.@: @dfn{a void function}), but without cluttering the output
16821 with @code{void} returned values that @value{GDBN} will otherwise
16822 print. If the result is not void, it is printed and saved in the
16826 It is possible for the function you call via the @code{print} or
16827 @code{call} command to generate a signal (e.g., if there's a bug in
16828 the function, or if you passed it incorrect arguments). What happens
16829 in that case is controlled by the @code{set unwindonsignal} command.
16831 Similarly, with a C@t{++} program it is possible for the function you
16832 call via the @code{print} or @code{call} command to generate an
16833 exception that is not handled due to the constraints of the dummy
16834 frame. In this case, any exception that is raised in the frame, but has
16835 an out-of-frame exception handler will not be found. GDB builds a
16836 dummy-frame for the inferior function call, and the unwinder cannot
16837 seek for exception handlers outside of this dummy-frame. What happens
16838 in that case is controlled by the
16839 @code{set unwind-on-terminating-exception} command.
16842 @item set unwindonsignal
16843 @kindex set unwindonsignal
16844 @cindex unwind stack in called functions
16845 @cindex call dummy stack unwinding
16846 Set unwinding of the stack if a signal is received while in a function
16847 that @value{GDBN} called in the program being debugged. If set to on,
16848 @value{GDBN} unwinds the stack it created for the call and restores
16849 the context to what it was before the call. If set to off (the
16850 default), @value{GDBN} stops in the frame where the signal was
16853 @item show unwindonsignal
16854 @kindex show unwindonsignal
16855 Show the current setting of stack unwinding in the functions called by
16858 @item set unwind-on-terminating-exception
16859 @kindex set unwind-on-terminating-exception
16860 @cindex unwind stack in called functions with unhandled exceptions
16861 @cindex call dummy stack unwinding on unhandled exception.
16862 Set unwinding of the stack if a C@t{++} exception is raised, but left
16863 unhandled while in a function that @value{GDBN} called in the program being
16864 debugged. If set to on (the default), @value{GDBN} unwinds the stack
16865 it created for the call and restores the context to what it was before
16866 the call. If set to off, @value{GDBN} the exception is delivered to
16867 the default C@t{++} exception handler and the inferior terminated.
16869 @item show unwind-on-terminating-exception
16870 @kindex show unwind-on-terminating-exception
16871 Show the current setting of stack unwinding in the functions called by
16876 @cindex weak alias functions
16877 Sometimes, a function you wish to call is actually a @dfn{weak alias}
16878 for another function. In such case, @value{GDBN} might not pick up
16879 the type information, including the types of the function arguments,
16880 which causes @value{GDBN} to call the inferior function incorrectly.
16881 As a result, the called function will function erroneously and may
16882 even crash. A solution to that is to use the name of the aliased
16886 @section Patching Programs
16888 @cindex patching binaries
16889 @cindex writing into executables
16890 @cindex writing into corefiles
16892 By default, @value{GDBN} opens the file containing your program's
16893 executable code (or the corefile) read-only. This prevents accidental
16894 alterations to machine code; but it also prevents you from intentionally
16895 patching your program's binary.
16897 If you'd like to be able to patch the binary, you can specify that
16898 explicitly with the @code{set write} command. For example, you might
16899 want to turn on internal debugging flags, or even to make emergency
16905 @itemx set write off
16906 If you specify @samp{set write on}, @value{GDBN} opens executable and
16907 core files for both reading and writing; if you specify @kbd{set write
16908 off} (the default), @value{GDBN} opens them read-only.
16910 If you have already loaded a file, you must load it again (using the
16911 @code{exec-file} or @code{core-file} command) after changing @code{set
16912 write}, for your new setting to take effect.
16916 Display whether executable files and core files are opened for writing
16917 as well as reading.
16920 @node Compiling and Injecting Code
16921 @section Compiling and injecting code in @value{GDBN}
16922 @cindex injecting code
16923 @cindex writing into executables
16924 @cindex compiling code
16926 @value{GDBN} supports on-demand compilation and code injection into
16927 programs running under @value{GDBN}. GCC 5.0 or higher built with
16928 @file{libcc1.so} must be installed for this functionality to be enabled.
16929 This functionality is implemented with the following commands.
16932 @kindex compile code
16933 @item compile code @var{source-code}
16934 @itemx compile code -raw @var{--} @var{source-code}
16935 Compile @var{source-code} with the compiler language found as the current
16936 language in @value{GDBN} (@pxref{Languages}). If compilation and
16937 injection is not supported with the current language specified in
16938 @value{GDBN}, or the compiler does not support this feature, an error
16939 message will be printed. If @var{source-code} compiles and links
16940 successfully, @value{GDBN} will load the object-code emitted,
16941 and execute it within the context of the currently selected inferior.
16942 It is important to note that the compiled code is executed immediately.
16943 After execution, the compiled code is removed from @value{GDBN} and any
16944 new types or variables you have defined will be deleted.
16946 The command allows you to specify @var{source-code} in two ways.
16947 The simplest method is to provide a single line of code to the command.
16951 compile code printf ("hello world\n");
16954 If you specify options on the command line as well as source code, they
16955 may conflict. The @samp{--} delimiter can be used to separate options
16956 from actual source code. E.g.:
16959 compile code -r -- printf ("hello world\n");
16962 Alternatively you can enter source code as multiple lines of text. To
16963 enter this mode, invoke the @samp{compile code} command without any text
16964 following the command. This will start the multiple-line editor and
16965 allow you to type as many lines of source code as required. When you
16966 have completed typing, enter @samp{end} on its own line to exit the
16971 >printf ("hello\n");
16972 >printf ("world\n");
16976 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
16977 provided @var{source-code} in a callable scope. In this case, you must
16978 specify the entry point of the code by defining a function named
16979 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
16980 inferior. Using @samp{-raw} option may be needed for example when
16981 @var{source-code} requires @samp{#include} lines which may conflict with
16982 inferior symbols otherwise.
16984 @kindex compile file
16985 @item compile file @var{filename}
16986 @itemx compile file -raw @var{filename}
16987 Like @code{compile code}, but take the source code from @var{filename}.
16990 compile file /home/user/example.c
16994 @subsection Caveats when using the @code{compile} command
16996 There are a few caveats to keep in mind when using the @code{compile}
16997 command. As the caveats are different per language, the table below
16998 highlights specific issues on a per language basis.
17001 @item C code examples and caveats
17002 When the language in @value{GDBN} is set to @samp{C}, the compiler will
17003 attempt to compile the source code with a @samp{C} compiler. The source
17004 code provided to the @code{compile} command will have much the same
17005 access to variables and types as it normally would if it were part of
17006 the program currently being debugged in @value{GDBN}.
17008 Below is a sample program that forms the basis of the examples that
17009 follow. This program has been compiled and loaded into @value{GDBN},
17010 much like any other normal debugging session.
17013 void function1 (void)
17016 printf ("function 1\n");
17019 void function2 (void)
17034 For the purposes of the examples in this section, the program above has
17035 been compiled, loaded into @value{GDBN}, stopped at the function
17036 @code{main}, and @value{GDBN} is awaiting input from the user.
17038 To access variables and types for any program in @value{GDBN}, the
17039 program must be compiled and packaged with debug information. The
17040 @code{compile} command is not an exception to this rule. Without debug
17041 information, you can still use the @code{compile} command, but you will
17042 be very limited in what variables and types you can access.
17044 So with that in mind, the example above has been compiled with debug
17045 information enabled. The @code{compile} command will have access to
17046 all variables and types (except those that may have been optimized
17047 out). Currently, as @value{GDBN} has stopped the program in the
17048 @code{main} function, the @code{compile} command would have access to
17049 the variable @code{k}. You could invoke the @code{compile} command
17050 and type some source code to set the value of @code{k}. You can also
17051 read it, or do anything with that variable you would normally do in
17052 @code{C}. Be aware that changes to inferior variables in the
17053 @code{compile} command are persistent. In the following example:
17056 compile code k = 3;
17060 the variable @code{k} is now 3. It will retain that value until
17061 something else in the example program changes it, or another
17062 @code{compile} command changes it.
17064 Normal scope and access rules apply to source code compiled and
17065 injected by the @code{compile} command. In the example, the variables
17066 @code{j} and @code{k} are not accessible yet, because the program is
17067 currently stopped in the @code{main} function, where these variables
17068 are not in scope. Therefore, the following command
17071 compile code j = 3;
17075 will result in a compilation error message.
17077 Once the program is continued, execution will bring these variables in
17078 scope, and they will become accessible; then the code you specify via
17079 the @code{compile} command will be able to access them.
17081 You can create variables and types with the @code{compile} command as
17082 part of your source code. Variables and types that are created as part
17083 of the @code{compile} command are not visible to the rest of the program for
17084 the duration of its run. This example is valid:
17087 compile code int ff = 5; printf ("ff is %d\n", ff);
17090 However, if you were to type the following into @value{GDBN} after that
17091 command has completed:
17094 compile code printf ("ff is %d\n'', ff);
17098 a compiler error would be raised as the variable @code{ff} no longer
17099 exists. Object code generated and injected by the @code{compile}
17100 command is removed when its execution ends. Caution is advised
17101 when assigning to program variables values of variables created by the
17102 code submitted to the @code{compile} command. This example is valid:
17105 compile code int ff = 5; k = ff;
17108 The value of the variable @code{ff} is assigned to @code{k}. The variable
17109 @code{k} does not require the existence of @code{ff} to maintain the value
17110 it has been assigned. However, pointers require particular care in
17111 assignment. If the source code compiled with the @code{compile} command
17112 changed the address of a pointer in the example program, perhaps to a
17113 variable created in the @code{compile} command, that pointer would point
17114 to an invalid location when the command exits. The following example
17115 would likely cause issues with your debugged program:
17118 compile code int ff = 5; p = &ff;
17121 In this example, @code{p} would point to @code{ff} when the
17122 @code{compile} command is executing the source code provided to it.
17123 However, as variables in the (example) program persist with their
17124 assigned values, the variable @code{p} would point to an invalid
17125 location when the command exists. A general rule should be followed
17126 in that you should either assign @code{NULL} to any assigned pointers,
17127 or restore a valid location to the pointer before the command exits.
17129 Similar caution must be exercised with any structs, unions, and typedefs
17130 defined in @code{compile} command. Types defined in the @code{compile}
17131 command will no longer be available in the next @code{compile} command.
17132 Therefore, if you cast a variable to a type defined in the
17133 @code{compile} command, care must be taken to ensure that any future
17134 need to resolve the type can be achieved.
17137 (gdb) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
17138 (gdb) compile code printf ("%d\n", ((struct a *) argv)->a);
17139 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
17140 Compilation failed.
17141 (gdb) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
17145 Variables that have been optimized away by the compiler are not
17146 accessible to the code submitted to the @code{compile} command.
17147 Access to those variables will generate a compiler error which @value{GDBN}
17148 will print to the console.
17152 @chapter @value{GDBN} Files
17154 @value{GDBN} needs to know the file name of the program to be debugged,
17155 both in order to read its symbol table and in order to start your
17156 program. To debug a core dump of a previous run, you must also tell
17157 @value{GDBN} the name of the core dump file.
17160 * Files:: Commands to specify files
17161 * Separate Debug Files:: Debugging information in separate files
17162 * MiniDebugInfo:: Debugging information in a special section
17163 * Index Files:: Index files speed up GDB
17164 * Symbol Errors:: Errors reading symbol files
17165 * Data Files:: GDB data files
17169 @section Commands to Specify Files
17171 @cindex symbol table
17172 @cindex core dump file
17174 You may want to specify executable and core dump file names. The usual
17175 way to do this is at start-up time, using the arguments to
17176 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
17177 Out of @value{GDBN}}).
17179 Occasionally it is necessary to change to a different file during a
17180 @value{GDBN} session. Or you may run @value{GDBN} and forget to
17181 specify a file you want to use. Or you are debugging a remote target
17182 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
17183 Program}). In these situations the @value{GDBN} commands to specify
17184 new files are useful.
17187 @cindex executable file
17189 @item file @var{filename}
17190 Use @var{filename} as the program to be debugged. It is read for its
17191 symbols and for the contents of pure memory. It is also the program
17192 executed when you use the @code{run} command. If you do not specify a
17193 directory and the file is not found in the @value{GDBN} working directory,
17194 @value{GDBN} uses the environment variable @code{PATH} as a list of
17195 directories to search, just as the shell does when looking for a program
17196 to run. You can change the value of this variable, for both @value{GDBN}
17197 and your program, using the @code{path} command.
17199 @cindex unlinked object files
17200 @cindex patching object files
17201 You can load unlinked object @file{.o} files into @value{GDBN} using
17202 the @code{file} command. You will not be able to ``run'' an object
17203 file, but you can disassemble functions and inspect variables. Also,
17204 if the underlying BFD functionality supports it, you could use
17205 @kbd{gdb -write} to patch object files using this technique. Note
17206 that @value{GDBN} can neither interpret nor modify relocations in this
17207 case, so branches and some initialized variables will appear to go to
17208 the wrong place. But this feature is still handy from time to time.
17211 @code{file} with no argument makes @value{GDBN} discard any information it
17212 has on both executable file and the symbol table.
17215 @item exec-file @r{[} @var{filename} @r{]}
17216 Specify that the program to be run (but not the symbol table) is found
17217 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
17218 if necessary to locate your program. Omitting @var{filename} means to
17219 discard information on the executable file.
17221 @kindex symbol-file
17222 @item symbol-file @r{[} @var{filename} @r{]}
17223 Read symbol table information from file @var{filename}. @code{PATH} is
17224 searched when necessary. Use the @code{file} command to get both symbol
17225 table and program to run from the same file.
17227 @code{symbol-file} with no argument clears out @value{GDBN} information on your
17228 program's symbol table.
17230 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
17231 some breakpoints and auto-display expressions. This is because they may
17232 contain pointers to the internal data recording symbols and data types,
17233 which are part of the old symbol table data being discarded inside
17236 @code{symbol-file} does not repeat if you press @key{RET} again after
17239 When @value{GDBN} is configured for a particular environment, it
17240 understands debugging information in whatever format is the standard
17241 generated for that environment; you may use either a @sc{gnu} compiler, or
17242 other compilers that adhere to the local conventions.
17243 Best results are usually obtained from @sc{gnu} compilers; for example,
17244 using @code{@value{NGCC}} you can generate debugging information for
17247 For most kinds of object files, with the exception of old SVR3 systems
17248 using COFF, the @code{symbol-file} command does not normally read the
17249 symbol table in full right away. Instead, it scans the symbol table
17250 quickly to find which source files and which symbols are present. The
17251 details are read later, one source file at a time, as they are needed.
17253 The purpose of this two-stage reading strategy is to make @value{GDBN}
17254 start up faster. For the most part, it is invisible except for
17255 occasional pauses while the symbol table details for a particular source
17256 file are being read. (The @code{set verbose} command can turn these
17257 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
17258 Warnings and Messages}.)
17260 We have not implemented the two-stage strategy for COFF yet. When the
17261 symbol table is stored in COFF format, @code{symbol-file} reads the
17262 symbol table data in full right away. Note that ``stabs-in-COFF''
17263 still does the two-stage strategy, since the debug info is actually
17267 @cindex reading symbols immediately
17268 @cindex symbols, reading immediately
17269 @item symbol-file @r{[} -readnow @r{]} @var{filename}
17270 @itemx file @r{[} -readnow @r{]} @var{filename}
17271 You can override the @value{GDBN} two-stage strategy for reading symbol
17272 tables by using the @samp{-readnow} option with any of the commands that
17273 load symbol table information, if you want to be sure @value{GDBN} has the
17274 entire symbol table available.
17276 @c FIXME: for now no mention of directories, since this seems to be in
17277 @c flux. 13mar1992 status is that in theory GDB would look either in
17278 @c current dir or in same dir as myprog; but issues like competing
17279 @c GDB's, or clutter in system dirs, mean that in practice right now
17280 @c only current dir is used. FFish says maybe a special GDB hierarchy
17281 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
17285 @item core-file @r{[}@var{filename}@r{]}
17287 Specify the whereabouts of a core dump file to be used as the ``contents
17288 of memory''. Traditionally, core files contain only some parts of the
17289 address space of the process that generated them; @value{GDBN} can access the
17290 executable file itself for other parts.
17292 @code{core-file} with no argument specifies that no core file is
17295 Note that the core file is ignored when your program is actually running
17296 under @value{GDBN}. So, if you have been running your program and you
17297 wish to debug a core file instead, you must kill the subprocess in which
17298 the program is running. To do this, use the @code{kill} command
17299 (@pxref{Kill Process, ,Killing the Child Process}).
17301 @kindex add-symbol-file
17302 @cindex dynamic linking
17303 @item add-symbol-file @var{filename} @var{address}
17304 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
17305 @itemx add-symbol-file @var{filename} @var{address} -s @var{section} @var{address} @dots{}
17306 The @code{add-symbol-file} command reads additional symbol table
17307 information from the file @var{filename}. You would use this command
17308 when @var{filename} has been dynamically loaded (by some other means)
17309 into the program that is running. The @var{address} should give the memory
17310 address at which the file has been loaded; @value{GDBN} cannot figure
17311 this out for itself. You can additionally specify an arbitrary number
17312 of @samp{-s @var{section} @var{address}} pairs, to give an explicit
17313 section name and base address for that section. You can specify any
17314 @var{address} as an expression.
17316 The symbol table of the file @var{filename} is added to the symbol table
17317 originally read with the @code{symbol-file} command. You can use the
17318 @code{add-symbol-file} command any number of times; the new symbol data
17319 thus read is kept in addition to the old.
17321 Changes can be reverted using the command @code{remove-symbol-file}.
17323 @cindex relocatable object files, reading symbols from
17324 @cindex object files, relocatable, reading symbols from
17325 @cindex reading symbols from relocatable object files
17326 @cindex symbols, reading from relocatable object files
17327 @cindex @file{.o} files, reading symbols from
17328 Although @var{filename} is typically a shared library file, an
17329 executable file, or some other object file which has been fully
17330 relocated for loading into a process, you can also load symbolic
17331 information from relocatable @file{.o} files, as long as:
17335 the file's symbolic information refers only to linker symbols defined in
17336 that file, not to symbols defined by other object files,
17338 every section the file's symbolic information refers to has actually
17339 been loaded into the inferior, as it appears in the file, and
17341 you can determine the address at which every section was loaded, and
17342 provide these to the @code{add-symbol-file} command.
17346 Some embedded operating systems, like Sun Chorus and VxWorks, can load
17347 relocatable files into an already running program; such systems
17348 typically make the requirements above easy to meet. However, it's
17349 important to recognize that many native systems use complex link
17350 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
17351 assembly, for example) that make the requirements difficult to meet. In
17352 general, one cannot assume that using @code{add-symbol-file} to read a
17353 relocatable object file's symbolic information will have the same effect
17354 as linking the relocatable object file into the program in the normal
17357 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
17359 @kindex remove-symbol-file
17360 @item remove-symbol-file @var{filename}
17361 @item remove-symbol-file -a @var{address}
17362 Remove a symbol file added via the @code{add-symbol-file} command. The
17363 file to remove can be identified by its @var{filename} or by an @var{address}
17364 that lies within the boundaries of this symbol file in memory. Example:
17367 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
17368 add symbol table from file "/home/user/gdb/mylib.so" at
17369 .text_addr = 0x7ffff7ff9480
17371 Reading symbols from /home/user/gdb/mylib.so...done.
17372 (gdb) remove-symbol-file -a 0x7ffff7ff9480
17373 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
17378 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
17380 @kindex add-symbol-file-from-memory
17381 @cindex @code{syscall DSO}
17382 @cindex load symbols from memory
17383 @item add-symbol-file-from-memory @var{address}
17384 Load symbols from the given @var{address} in a dynamically loaded
17385 object file whose image is mapped directly into the inferior's memory.
17386 For example, the Linux kernel maps a @code{syscall DSO} into each
17387 process's address space; this DSO provides kernel-specific code for
17388 some system calls. The argument can be any expression whose
17389 evaluation yields the address of the file's shared object file header.
17390 For this command to work, you must have used @code{symbol-file} or
17391 @code{exec-file} commands in advance.
17394 @item section @var{section} @var{addr}
17395 The @code{section} command changes the base address of the named
17396 @var{section} of the exec file to @var{addr}. This can be used if the
17397 exec file does not contain section addresses, (such as in the
17398 @code{a.out} format), or when the addresses specified in the file
17399 itself are wrong. Each section must be changed separately. The
17400 @code{info files} command, described below, lists all the sections and
17404 @kindex info target
17407 @code{info files} and @code{info target} are synonymous; both print the
17408 current target (@pxref{Targets, ,Specifying a Debugging Target}),
17409 including the names of the executable and core dump files currently in
17410 use by @value{GDBN}, and the files from which symbols were loaded. The
17411 command @code{help target} lists all possible targets rather than
17414 @kindex maint info sections
17415 @item maint info sections
17416 Another command that can give you extra information about program sections
17417 is @code{maint info sections}. In addition to the section information
17418 displayed by @code{info files}, this command displays the flags and file
17419 offset of each section in the executable and core dump files. In addition,
17420 @code{maint info sections} provides the following command options (which
17421 may be arbitrarily combined):
17425 Display sections for all loaded object files, including shared libraries.
17426 @item @var{sections}
17427 Display info only for named @var{sections}.
17428 @item @var{section-flags}
17429 Display info only for sections for which @var{section-flags} are true.
17430 The section flags that @value{GDBN} currently knows about are:
17433 Section will have space allocated in the process when loaded.
17434 Set for all sections except those containing debug information.
17436 Section will be loaded from the file into the child process memory.
17437 Set for pre-initialized code and data, clear for @code{.bss} sections.
17439 Section needs to be relocated before loading.
17441 Section cannot be modified by the child process.
17443 Section contains executable code only.
17445 Section contains data only (no executable code).
17447 Section will reside in ROM.
17449 Section contains data for constructor/destructor lists.
17451 Section is not empty.
17453 An instruction to the linker to not output the section.
17454 @item COFF_SHARED_LIBRARY
17455 A notification to the linker that the section contains
17456 COFF shared library information.
17458 Section contains common symbols.
17461 @kindex set trust-readonly-sections
17462 @cindex read-only sections
17463 @item set trust-readonly-sections on
17464 Tell @value{GDBN} that readonly sections in your object file
17465 really are read-only (i.e.@: that their contents will not change).
17466 In that case, @value{GDBN} can fetch values from these sections
17467 out of the object file, rather than from the target program.
17468 For some targets (notably embedded ones), this can be a significant
17469 enhancement to debugging performance.
17471 The default is off.
17473 @item set trust-readonly-sections off
17474 Tell @value{GDBN} not to trust readonly sections. This means that
17475 the contents of the section might change while the program is running,
17476 and must therefore be fetched from the target when needed.
17478 @item show trust-readonly-sections
17479 Show the current setting of trusting readonly sections.
17482 All file-specifying commands allow both absolute and relative file names
17483 as arguments. @value{GDBN} always converts the file name to an absolute file
17484 name and remembers it that way.
17486 @cindex shared libraries
17487 @anchor{Shared Libraries}
17488 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
17489 and IBM RS/6000 AIX shared libraries.
17491 On MS-Windows @value{GDBN} must be linked with the Expat library to support
17492 shared libraries. @xref{Expat}.
17494 @value{GDBN} automatically loads symbol definitions from shared libraries
17495 when you use the @code{run} command, or when you examine a core file.
17496 (Before you issue the @code{run} command, @value{GDBN} does not understand
17497 references to a function in a shared library, however---unless you are
17498 debugging a core file).
17500 On HP-UX, if the program loads a library explicitly, @value{GDBN}
17501 automatically loads the symbols at the time of the @code{shl_load} call.
17503 @c FIXME: some @value{GDBN} release may permit some refs to undef
17504 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
17505 @c FIXME...lib; check this from time to time when updating manual
17507 There are times, however, when you may wish to not automatically load
17508 symbol definitions from shared libraries, such as when they are
17509 particularly large or there are many of them.
17511 To control the automatic loading of shared library symbols, use the
17515 @kindex set auto-solib-add
17516 @item set auto-solib-add @var{mode}
17517 If @var{mode} is @code{on}, symbols from all shared object libraries
17518 will be loaded automatically when the inferior begins execution, you
17519 attach to an independently started inferior, or when the dynamic linker
17520 informs @value{GDBN} that a new library has been loaded. If @var{mode}
17521 is @code{off}, symbols must be loaded manually, using the
17522 @code{sharedlibrary} command. The default value is @code{on}.
17524 @cindex memory used for symbol tables
17525 If your program uses lots of shared libraries with debug info that
17526 takes large amounts of memory, you can decrease the @value{GDBN}
17527 memory footprint by preventing it from automatically loading the
17528 symbols from shared libraries. To that end, type @kbd{set
17529 auto-solib-add off} before running the inferior, then load each
17530 library whose debug symbols you do need with @kbd{sharedlibrary
17531 @var{regexp}}, where @var{regexp} is a regular expression that matches
17532 the libraries whose symbols you want to be loaded.
17534 @kindex show auto-solib-add
17535 @item show auto-solib-add
17536 Display the current autoloading mode.
17539 @cindex load shared library
17540 To explicitly load shared library symbols, use the @code{sharedlibrary}
17544 @kindex info sharedlibrary
17546 @item info share @var{regex}
17547 @itemx info sharedlibrary @var{regex}
17548 Print the names of the shared libraries which are currently loaded
17549 that match @var{regex}. If @var{regex} is omitted then print
17550 all shared libraries that are loaded.
17552 @kindex sharedlibrary
17554 @item sharedlibrary @var{regex}
17555 @itemx share @var{regex}
17556 Load shared object library symbols for files matching a
17557 Unix regular expression.
17558 As with files loaded automatically, it only loads shared libraries
17559 required by your program for a core file or after typing @code{run}. If
17560 @var{regex} is omitted all shared libraries required by your program are
17563 @item nosharedlibrary
17564 @kindex nosharedlibrary
17565 @cindex unload symbols from shared libraries
17566 Unload all shared object library symbols. This discards all symbols
17567 that have been loaded from all shared libraries. Symbols from shared
17568 libraries that were loaded by explicit user requests are not
17572 Sometimes you may wish that @value{GDBN} stops and gives you control
17573 when any of shared library events happen. The best way to do this is
17574 to use @code{catch load} and @code{catch unload} (@pxref{Set
17577 @value{GDBN} also supports the the @code{set stop-on-solib-events}
17578 command for this. This command exists for historical reasons. It is
17579 less useful than setting a catchpoint, because it does not allow for
17580 conditions or commands as a catchpoint does.
17583 @item set stop-on-solib-events
17584 @kindex set stop-on-solib-events
17585 This command controls whether @value{GDBN} should give you control
17586 when the dynamic linker notifies it about some shared library event.
17587 The most common event of interest is loading or unloading of a new
17590 @item show stop-on-solib-events
17591 @kindex show stop-on-solib-events
17592 Show whether @value{GDBN} stops and gives you control when shared
17593 library events happen.
17596 Shared libraries are also supported in many cross or remote debugging
17597 configurations. @value{GDBN} needs to have access to the target's libraries;
17598 this can be accomplished either by providing copies of the libraries
17599 on the host system, or by asking @value{GDBN} to automatically retrieve the
17600 libraries from the target. If copies of the target libraries are
17601 provided, they need to be the same as the target libraries, although the
17602 copies on the target can be stripped as long as the copies on the host are
17605 @cindex where to look for shared libraries
17606 For remote debugging, you need to tell @value{GDBN} where the target
17607 libraries are, so that it can load the correct copies---otherwise, it
17608 may try to load the host's libraries. @value{GDBN} has two variables
17609 to specify the search directories for target libraries.
17612 @cindex prefix for shared library file names
17613 @cindex system root, alternate
17614 @kindex set solib-absolute-prefix
17615 @kindex set sysroot
17616 @item set sysroot @var{path}
17617 Use @var{path} as the system root for the program being debugged. Any
17618 absolute shared library paths will be prefixed with @var{path}; many
17619 runtime loaders store the absolute paths to the shared library in the
17620 target program's memory. If you use @code{set sysroot} to find shared
17621 libraries, they need to be laid out in the same way that they are on
17622 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
17625 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
17626 retrieve the target libraries from the remote system. This is only
17627 supported when using a remote target that supports the @code{remote get}
17628 command (@pxref{File Transfer,,Sending files to a remote system}).
17629 The part of @var{path} following the initial @file{remote:}
17630 (if present) is used as system root prefix on the remote file system.
17631 @footnote{If you want to specify a local system root using a directory
17632 that happens to be named @file{remote:}, you need to use some equivalent
17633 variant of the name like @file{./remote:}.}
17635 For targets with an MS-DOS based filesystem, such as MS-Windows and
17636 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
17637 absolute file name with @var{path}. But first, on Unix hosts,
17638 @value{GDBN} converts all backslash directory separators into forward
17639 slashes, because the backslash is not a directory separator on Unix:
17642 c:\foo\bar.dll @result{} c:/foo/bar.dll
17645 Then, @value{GDBN} attempts prefixing the target file name with
17646 @var{path}, and looks for the resulting file name in the host file
17650 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
17653 If that does not find the shared library, @value{GDBN} tries removing
17654 the @samp{:} character from the drive spec, both for convenience, and,
17655 for the case of the host file system not supporting file names with
17659 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
17662 This makes it possible to have a system root that mirrors a target
17663 with more than one drive. E.g., you may want to setup your local
17664 copies of the target system shared libraries like so (note @samp{c} vs
17668 @file{/path/to/sysroot/c/sys/bin/foo.dll}
17669 @file{/path/to/sysroot/c/sys/bin/bar.dll}
17670 @file{/path/to/sysroot/z/sys/bin/bar.dll}
17674 and point the system root at @file{/path/to/sysroot}, so that
17675 @value{GDBN} can find the correct copies of both
17676 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
17678 If that still does not find the shared library, @value{GDBN} tries
17679 removing the whole drive spec from the target file name:
17682 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
17685 This last lookup makes it possible to not care about the drive name,
17686 if you don't want or need to.
17688 The @code{set solib-absolute-prefix} command is an alias for @code{set
17691 @cindex default system root
17692 @cindex @samp{--with-sysroot}
17693 You can set the default system root by using the configure-time
17694 @samp{--with-sysroot} option. If the system root is inside
17695 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
17696 @samp{--exec-prefix}), then the default system root will be updated
17697 automatically if the installed @value{GDBN} is moved to a new
17700 @kindex show sysroot
17702 Display the current shared library prefix.
17704 @kindex set solib-search-path
17705 @item set solib-search-path @var{path}
17706 If this variable is set, @var{path} is a colon-separated list of
17707 directories to search for shared libraries. @samp{solib-search-path}
17708 is used after @samp{sysroot} fails to locate the library, or if the
17709 path to the library is relative instead of absolute. If you want to
17710 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
17711 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
17712 finding your host's libraries. @samp{sysroot} is preferred; setting
17713 it to a nonexistent directory may interfere with automatic loading
17714 of shared library symbols.
17716 @kindex show solib-search-path
17717 @item show solib-search-path
17718 Display the current shared library search path.
17720 @cindex DOS file-name semantics of file names.
17721 @kindex set target-file-system-kind (unix|dos-based|auto)
17722 @kindex show target-file-system-kind
17723 @item set target-file-system-kind @var{kind}
17724 Set assumed file system kind for target reported file names.
17726 Shared library file names as reported by the target system may not
17727 make sense as is on the system @value{GDBN} is running on. For
17728 example, when remote debugging a target that has MS-DOS based file
17729 system semantics, from a Unix host, the target may be reporting to
17730 @value{GDBN} a list of loaded shared libraries with file names such as
17731 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
17732 drive letters, so the @samp{c:\} prefix is not normally understood as
17733 indicating an absolute file name, and neither is the backslash
17734 normally considered a directory separator character. In that case,
17735 the native file system would interpret this whole absolute file name
17736 as a relative file name with no directory components. This would make
17737 it impossible to point @value{GDBN} at a copy of the remote target's
17738 shared libraries on the host using @code{set sysroot}, and impractical
17739 with @code{set solib-search-path}. Setting
17740 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
17741 to interpret such file names similarly to how the target would, and to
17742 map them to file names valid on @value{GDBN}'s native file system
17743 semantics. The value of @var{kind} can be @code{"auto"}, in addition
17744 to one of the supported file system kinds. In that case, @value{GDBN}
17745 tries to determine the appropriate file system variant based on the
17746 current target's operating system (@pxref{ABI, ,Configuring the
17747 Current ABI}). The supported file system settings are:
17751 Instruct @value{GDBN} to assume the target file system is of Unix
17752 kind. Only file names starting the forward slash (@samp{/}) character
17753 are considered absolute, and the directory separator character is also
17757 Instruct @value{GDBN} to assume the target file system is DOS based.
17758 File names starting with either a forward slash, or a drive letter
17759 followed by a colon (e.g., @samp{c:}), are considered absolute, and
17760 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
17761 considered directory separators.
17764 Instruct @value{GDBN} to use the file system kind associated with the
17765 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
17766 This is the default.
17770 @cindex file name canonicalization
17771 @cindex base name differences
17772 When processing file names provided by the user, @value{GDBN}
17773 frequently needs to compare them to the file names recorded in the
17774 program's debug info. Normally, @value{GDBN} compares just the
17775 @dfn{base names} of the files as strings, which is reasonably fast
17776 even for very large programs. (The base name of a file is the last
17777 portion of its name, after stripping all the leading directories.)
17778 This shortcut in comparison is based upon the assumption that files
17779 cannot have more than one base name. This is usually true, but
17780 references to files that use symlinks or similar filesystem
17781 facilities violate that assumption. If your program records files
17782 using such facilities, or if you provide file names to @value{GDBN}
17783 using symlinks etc., you can set @code{basenames-may-differ} to
17784 @code{true} to instruct @value{GDBN} to completely canonicalize each
17785 pair of file names it needs to compare. This will make file-name
17786 comparisons accurate, but at a price of a significant slowdown.
17789 @item set basenames-may-differ
17790 @kindex set basenames-may-differ
17791 Set whether a source file may have multiple base names.
17793 @item show basenames-may-differ
17794 @kindex show basenames-may-differ
17795 Show whether a source file may have multiple base names.
17798 @node Separate Debug Files
17799 @section Debugging Information in Separate Files
17800 @cindex separate debugging information files
17801 @cindex debugging information in separate files
17802 @cindex @file{.debug} subdirectories
17803 @cindex debugging information directory, global
17804 @cindex global debugging information directories
17805 @cindex build ID, and separate debugging files
17806 @cindex @file{.build-id} directory
17808 @value{GDBN} allows you to put a program's debugging information in a
17809 file separate from the executable itself, in a way that allows
17810 @value{GDBN} to find and load the debugging information automatically.
17811 Since debugging information can be very large---sometimes larger
17812 than the executable code itself---some systems distribute debugging
17813 information for their executables in separate files, which users can
17814 install only when they need to debug a problem.
17816 @value{GDBN} supports two ways of specifying the separate debug info
17821 The executable contains a @dfn{debug link} that specifies the name of
17822 the separate debug info file. The separate debug file's name is
17823 usually @file{@var{executable}.debug}, where @var{executable} is the
17824 name of the corresponding executable file without leading directories
17825 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
17826 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
17827 checksum for the debug file, which @value{GDBN} uses to validate that
17828 the executable and the debug file came from the same build.
17831 The executable contains a @dfn{build ID}, a unique bit string that is
17832 also present in the corresponding debug info file. (This is supported
17833 only on some operating systems, notably those which use the ELF format
17834 for binary files and the @sc{gnu} Binutils.) For more details about
17835 this feature, see the description of the @option{--build-id}
17836 command-line option in @ref{Options, , Command Line Options, ld.info,
17837 The GNU Linker}. The debug info file's name is not specified
17838 explicitly by the build ID, but can be computed from the build ID, see
17842 Depending on the way the debug info file is specified, @value{GDBN}
17843 uses two different methods of looking for the debug file:
17847 For the ``debug link'' method, @value{GDBN} looks up the named file in
17848 the directory of the executable file, then in a subdirectory of that
17849 directory named @file{.debug}, and finally under each one of the global debug
17850 directories, in a subdirectory whose name is identical to the leading
17851 directories of the executable's absolute file name.
17854 For the ``build ID'' method, @value{GDBN} looks in the
17855 @file{.build-id} subdirectory of each one of the global debug directories for
17856 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
17857 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
17858 are the rest of the bit string. (Real build ID strings are 32 or more
17859 hex characters, not 10.)
17862 So, for example, suppose you ask @value{GDBN} to debug
17863 @file{/usr/bin/ls}, which has a debug link that specifies the
17864 file @file{ls.debug}, and a build ID whose value in hex is
17865 @code{abcdef1234}. If the list of the global debug directories includes
17866 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
17867 debug information files, in the indicated order:
17871 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
17873 @file{/usr/bin/ls.debug}
17875 @file{/usr/bin/.debug/ls.debug}
17877 @file{/usr/lib/debug/usr/bin/ls.debug}.
17880 @anchor{debug-file-directory}
17881 Global debugging info directories default to what is set by @value{GDBN}
17882 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
17883 you can also set the global debugging info directories, and view the list
17884 @value{GDBN} is currently using.
17888 @kindex set debug-file-directory
17889 @item set debug-file-directory @var{directories}
17890 Set the directories which @value{GDBN} searches for separate debugging
17891 information files to @var{directory}. Multiple path components can be set
17892 concatenating them by a path separator.
17894 @kindex show debug-file-directory
17895 @item show debug-file-directory
17896 Show the directories @value{GDBN} searches for separate debugging
17901 @cindex @code{.gnu_debuglink} sections
17902 @cindex debug link sections
17903 A debug link is a special section of the executable file named
17904 @code{.gnu_debuglink}. The section must contain:
17908 A filename, with any leading directory components removed, followed by
17911 zero to three bytes of padding, as needed to reach the next four-byte
17912 boundary within the section, and
17914 a four-byte CRC checksum, stored in the same endianness used for the
17915 executable file itself. The checksum is computed on the debugging
17916 information file's full contents by the function given below, passing
17917 zero as the @var{crc} argument.
17920 Any executable file format can carry a debug link, as long as it can
17921 contain a section named @code{.gnu_debuglink} with the contents
17924 @cindex @code{.note.gnu.build-id} sections
17925 @cindex build ID sections
17926 The build ID is a special section in the executable file (and in other
17927 ELF binary files that @value{GDBN} may consider). This section is
17928 often named @code{.note.gnu.build-id}, but that name is not mandatory.
17929 It contains unique identification for the built files---the ID remains
17930 the same across multiple builds of the same build tree. The default
17931 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
17932 content for the build ID string. The same section with an identical
17933 value is present in the original built binary with symbols, in its
17934 stripped variant, and in the separate debugging information file.
17936 The debugging information file itself should be an ordinary
17937 executable, containing a full set of linker symbols, sections, and
17938 debugging information. The sections of the debugging information file
17939 should have the same names, addresses, and sizes as the original file,
17940 but they need not contain any data---much like a @code{.bss} section
17941 in an ordinary executable.
17943 The @sc{gnu} binary utilities (Binutils) package includes the
17944 @samp{objcopy} utility that can produce
17945 the separated executable / debugging information file pairs using the
17946 following commands:
17949 @kbd{objcopy --only-keep-debug foo foo.debug}
17954 These commands remove the debugging
17955 information from the executable file @file{foo} and place it in the file
17956 @file{foo.debug}. You can use the first, second or both methods to link the
17961 The debug link method needs the following additional command to also leave
17962 behind a debug link in @file{foo}:
17965 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
17968 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
17969 a version of the @code{strip} command such that the command @kbd{strip foo -f
17970 foo.debug} has the same functionality as the two @code{objcopy} commands and
17971 the @code{ln -s} command above, together.
17974 Build ID gets embedded into the main executable using @code{ld --build-id} or
17975 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
17976 compatibility fixes for debug files separation are present in @sc{gnu} binary
17977 utilities (Binutils) package since version 2.18.
17982 @cindex CRC algorithm definition
17983 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
17984 IEEE 802.3 using the polynomial:
17986 @c TexInfo requires naked braces for multi-digit exponents for Tex
17987 @c output, but this causes HTML output to barf. HTML has to be set using
17988 @c raw commands. So we end up having to specify this equation in 2
17993 <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>
17994 + <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
18000 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
18001 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
18005 The function is computed byte at a time, taking the least
18006 significant bit of each byte first. The initial pattern
18007 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
18008 the final result is inverted to ensure trailing zeros also affect the
18011 @emph{Note:} This is the same CRC polynomial as used in handling the
18012 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
18013 However in the case of the Remote Serial Protocol, the CRC is computed
18014 @emph{most} significant bit first, and the result is not inverted, so
18015 trailing zeros have no effect on the CRC value.
18017 To complete the description, we show below the code of the function
18018 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
18019 initially supplied @code{crc} argument means that an initial call to
18020 this function passing in zero will start computing the CRC using
18023 @kindex gnu_debuglink_crc32
18026 gnu_debuglink_crc32 (unsigned long crc,
18027 unsigned char *buf, size_t len)
18029 static const unsigned long crc32_table[256] =
18031 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
18032 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
18033 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
18034 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
18035 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
18036 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
18037 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
18038 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
18039 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
18040 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
18041 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
18042 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
18043 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
18044 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
18045 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
18046 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
18047 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
18048 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
18049 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
18050 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
18051 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
18052 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
18053 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
18054 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
18055 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
18056 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
18057 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
18058 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
18059 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
18060 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
18061 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
18062 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
18063 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
18064 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
18065 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
18066 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
18067 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
18068 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
18069 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
18070 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
18071 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
18072 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
18073 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
18074 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
18075 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
18076 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
18077 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
18078 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
18079 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
18080 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
18081 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
18084 unsigned char *end;
18086 crc = ~crc & 0xffffffff;
18087 for (end = buf + len; buf < end; ++buf)
18088 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
18089 return ~crc & 0xffffffff;
18094 This computation does not apply to the ``build ID'' method.
18096 @node MiniDebugInfo
18097 @section Debugging information in a special section
18098 @cindex separate debug sections
18099 @cindex @samp{.gnu_debugdata} section
18101 Some systems ship pre-built executables and libraries that have a
18102 special @samp{.gnu_debugdata} section. This feature is called
18103 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
18104 is used to supply extra symbols for backtraces.
18106 The intent of this section is to provide extra minimal debugging
18107 information for use in simple backtraces. It is not intended to be a
18108 replacement for full separate debugging information (@pxref{Separate
18109 Debug Files}). The example below shows the intended use; however,
18110 @value{GDBN} does not currently put restrictions on what sort of
18111 debugging information might be included in the section.
18113 @value{GDBN} has support for this extension. If the section exists,
18114 then it is used provided that no other source of debugging information
18115 can be found, and that @value{GDBN} was configured with LZMA support.
18117 This section can be easily created using @command{objcopy} and other
18118 standard utilities:
18121 # Extract the dynamic symbols from the main binary, there is no need
18122 # to also have these in the normal symbol table.
18123 nm -D @var{binary} --format=posix --defined-only \
18124 | awk '@{ print $1 @}' | sort > dynsyms
18126 # Extract all the text (i.e. function) symbols from the debuginfo.
18127 # (Note that we actually also accept "D" symbols, for the benefit
18128 # of platforms like PowerPC64 that use function descriptors.)
18129 nm @var{binary} --format=posix --defined-only \
18130 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
18133 # Keep all the function symbols not already in the dynamic symbol
18135 comm -13 dynsyms funcsyms > keep_symbols
18137 # Separate full debug info into debug binary.
18138 objcopy --only-keep-debug @var{binary} debug
18140 # Copy the full debuginfo, keeping only a minimal set of symbols and
18141 # removing some unnecessary sections.
18142 objcopy -S --remove-section .gdb_index --remove-section .comment \
18143 --keep-symbols=keep_symbols debug mini_debuginfo
18145 # Drop the full debug info from the original binary.
18146 strip --strip-all -R .comment @var{binary}
18148 # Inject the compressed data into the .gnu_debugdata section of the
18151 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
18155 @section Index Files Speed Up @value{GDBN}
18156 @cindex index files
18157 @cindex @samp{.gdb_index} section
18159 When @value{GDBN} finds a symbol file, it scans the symbols in the
18160 file in order to construct an internal symbol table. This lets most
18161 @value{GDBN} operations work quickly---at the cost of a delay early
18162 on. For large programs, this delay can be quite lengthy, so
18163 @value{GDBN} provides a way to build an index, which speeds up
18166 The index is stored as a section in the symbol file. @value{GDBN} can
18167 write the index to a file, then you can put it into the symbol file
18168 using @command{objcopy}.
18170 To create an index file, use the @code{save gdb-index} command:
18173 @item save gdb-index @var{directory}
18174 @kindex save gdb-index
18175 Create an index file for each symbol file currently known by
18176 @value{GDBN}. Each file is named after its corresponding symbol file,
18177 with @samp{.gdb-index} appended, and is written into the given
18181 Once you have created an index file you can merge it into your symbol
18182 file, here named @file{symfile}, using @command{objcopy}:
18185 $ objcopy --add-section .gdb_index=symfile.gdb-index \
18186 --set-section-flags .gdb_index=readonly symfile symfile
18189 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
18190 sections that have been deprecated. Usually they are deprecated because
18191 they are missing a new feature or have performance issues.
18192 To tell @value{GDBN} to use a deprecated index section anyway
18193 specify @code{set use-deprecated-index-sections on}.
18194 The default is @code{off}.
18195 This can speed up startup, but may result in some functionality being lost.
18196 @xref{Index Section Format}.
18198 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
18199 must be done before gdb reads the file. The following will not work:
18202 $ gdb -ex "set use-deprecated-index-sections on" <program>
18205 Instead you must do, for example,
18208 $ gdb -iex "set use-deprecated-index-sections on" <program>
18211 There are currently some limitation on indices. They only work when
18212 for DWARF debugging information, not stabs. And, they do not
18213 currently work for programs using Ada.
18215 @node Symbol Errors
18216 @section Errors Reading Symbol Files
18218 While reading a symbol file, @value{GDBN} occasionally encounters problems,
18219 such as symbol types it does not recognize, or known bugs in compiler
18220 output. By default, @value{GDBN} does not notify you of such problems, since
18221 they are relatively common and primarily of interest to people
18222 debugging compilers. If you are interested in seeing information
18223 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
18224 only one message about each such type of problem, no matter how many
18225 times the problem occurs; or you can ask @value{GDBN} to print more messages,
18226 to see how many times the problems occur, with the @code{set
18227 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
18230 The messages currently printed, and their meanings, include:
18233 @item inner block not inside outer block in @var{symbol}
18235 The symbol information shows where symbol scopes begin and end
18236 (such as at the start of a function or a block of statements). This
18237 error indicates that an inner scope block is not fully contained
18238 in its outer scope blocks.
18240 @value{GDBN} circumvents the problem by treating the inner block as if it had
18241 the same scope as the outer block. In the error message, @var{symbol}
18242 may be shown as ``@code{(don't know)}'' if the outer block is not a
18245 @item block at @var{address} out of order
18247 The symbol information for symbol scope blocks should occur in
18248 order of increasing addresses. This error indicates that it does not
18251 @value{GDBN} does not circumvent this problem, and has trouble
18252 locating symbols in the source file whose symbols it is reading. (You
18253 can often determine what source file is affected by specifying
18254 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
18257 @item bad block start address patched
18259 The symbol information for a symbol scope block has a start address
18260 smaller than the address of the preceding source line. This is known
18261 to occur in the SunOS 4.1.1 (and earlier) C compiler.
18263 @value{GDBN} circumvents the problem by treating the symbol scope block as
18264 starting on the previous source line.
18266 @item bad string table offset in symbol @var{n}
18269 Symbol number @var{n} contains a pointer into the string table which is
18270 larger than the size of the string table.
18272 @value{GDBN} circumvents the problem by considering the symbol to have the
18273 name @code{foo}, which may cause other problems if many symbols end up
18276 @item unknown symbol type @code{0x@var{nn}}
18278 The symbol information contains new data types that @value{GDBN} does
18279 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
18280 uncomprehended information, in hexadecimal.
18282 @value{GDBN} circumvents the error by ignoring this symbol information.
18283 This usually allows you to debug your program, though certain symbols
18284 are not accessible. If you encounter such a problem and feel like
18285 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
18286 on @code{complain}, then go up to the function @code{read_dbx_symtab}
18287 and examine @code{*bufp} to see the symbol.
18289 @item stub type has NULL name
18291 @value{GDBN} could not find the full definition for a struct or class.
18293 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
18294 The symbol information for a C@t{++} member function is missing some
18295 information that recent versions of the compiler should have output for
18298 @item info mismatch between compiler and debugger
18300 @value{GDBN} could not parse a type specification output by the compiler.
18305 @section GDB Data Files
18307 @cindex prefix for data files
18308 @value{GDBN} will sometimes read an auxiliary data file. These files
18309 are kept in a directory known as the @dfn{data directory}.
18311 You can set the data directory's name, and view the name @value{GDBN}
18312 is currently using.
18315 @kindex set data-directory
18316 @item set data-directory @var{directory}
18317 Set the directory which @value{GDBN} searches for auxiliary data files
18318 to @var{directory}.
18320 @kindex show data-directory
18321 @item show data-directory
18322 Show the directory @value{GDBN} searches for auxiliary data files.
18325 @cindex default data directory
18326 @cindex @samp{--with-gdb-datadir}
18327 You can set the default data directory by using the configure-time
18328 @samp{--with-gdb-datadir} option. If the data directory is inside
18329 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
18330 @samp{--exec-prefix}), then the default data directory will be updated
18331 automatically if the installed @value{GDBN} is moved to a new
18334 The data directory may also be specified with the
18335 @code{--data-directory} command line option.
18336 @xref{Mode Options}.
18339 @chapter Specifying a Debugging Target
18341 @cindex debugging target
18342 A @dfn{target} is the execution environment occupied by your program.
18344 Often, @value{GDBN} runs in the same host environment as your program;
18345 in that case, the debugging target is specified as a side effect when
18346 you use the @code{file} or @code{core} commands. When you need more
18347 flexibility---for example, running @value{GDBN} on a physically separate
18348 host, or controlling a standalone system over a serial port or a
18349 realtime system over a TCP/IP connection---you can use the @code{target}
18350 command to specify one of the target types configured for @value{GDBN}
18351 (@pxref{Target Commands, ,Commands for Managing Targets}).
18353 @cindex target architecture
18354 It is possible to build @value{GDBN} for several different @dfn{target
18355 architectures}. When @value{GDBN} is built like that, you can choose
18356 one of the available architectures with the @kbd{set architecture}
18360 @kindex set architecture
18361 @kindex show architecture
18362 @item set architecture @var{arch}
18363 This command sets the current target architecture to @var{arch}. The
18364 value of @var{arch} can be @code{"auto"}, in addition to one of the
18365 supported architectures.
18367 @item show architecture
18368 Show the current target architecture.
18370 @item set processor
18372 @kindex set processor
18373 @kindex show processor
18374 These are alias commands for, respectively, @code{set architecture}
18375 and @code{show architecture}.
18379 * Active Targets:: Active targets
18380 * Target Commands:: Commands for managing targets
18381 * Byte Order:: Choosing target byte order
18384 @node Active Targets
18385 @section Active Targets
18387 @cindex stacking targets
18388 @cindex active targets
18389 @cindex multiple targets
18391 There are multiple classes of targets such as: processes, executable files or
18392 recording sessions. Core files belong to the process class, making core file
18393 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
18394 on multiple active targets, one in each class. This allows you to (for
18395 example) start a process and inspect its activity, while still having access to
18396 the executable file after the process finishes. Or if you start process
18397 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
18398 presented a virtual layer of the recording target, while the process target
18399 remains stopped at the chronologically last point of the process execution.
18401 Use the @code{core-file} and @code{exec-file} commands to select a new core
18402 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
18403 specify as a target a process that is already running, use the @code{attach}
18404 command (@pxref{Attach, ,Debugging an Already-running Process}).
18406 @node Target Commands
18407 @section Commands for Managing Targets
18410 @item target @var{type} @var{parameters}
18411 Connects the @value{GDBN} host environment to a target machine or
18412 process. A target is typically a protocol for talking to debugging
18413 facilities. You use the argument @var{type} to specify the type or
18414 protocol of the target machine.
18416 Further @var{parameters} are interpreted by the target protocol, but
18417 typically include things like device names or host names to connect
18418 with, process numbers, and baud rates.
18420 The @code{target} command does not repeat if you press @key{RET} again
18421 after executing the command.
18423 @kindex help target
18425 Displays the names of all targets available. To display targets
18426 currently selected, use either @code{info target} or @code{info files}
18427 (@pxref{Files, ,Commands to Specify Files}).
18429 @item help target @var{name}
18430 Describe a particular target, including any parameters necessary to
18433 @kindex set gnutarget
18434 @item set gnutarget @var{args}
18435 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
18436 knows whether it is reading an @dfn{executable},
18437 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
18438 with the @code{set gnutarget} command. Unlike most @code{target} commands,
18439 with @code{gnutarget} the @code{target} refers to a program, not a machine.
18442 @emph{Warning:} To specify a file format with @code{set gnutarget},
18443 you must know the actual BFD name.
18447 @xref{Files, , Commands to Specify Files}.
18449 @kindex show gnutarget
18450 @item show gnutarget
18451 Use the @code{show gnutarget} command to display what file format
18452 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
18453 @value{GDBN} will determine the file format for each file automatically,
18454 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
18457 @cindex common targets
18458 Here are some common targets (available, or not, depending on the GDB
18463 @item target exec @var{program}
18464 @cindex executable file target
18465 An executable file. @samp{target exec @var{program}} is the same as
18466 @samp{exec-file @var{program}}.
18468 @item target core @var{filename}
18469 @cindex core dump file target
18470 A core dump file. @samp{target core @var{filename}} is the same as
18471 @samp{core-file @var{filename}}.
18473 @item target remote @var{medium}
18474 @cindex remote target
18475 A remote system connected to @value{GDBN} via a serial line or network
18476 connection. This command tells @value{GDBN} to use its own remote
18477 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
18479 For example, if you have a board connected to @file{/dev/ttya} on the
18480 machine running @value{GDBN}, you could say:
18483 target remote /dev/ttya
18486 @code{target remote} supports the @code{load} command. This is only
18487 useful if you have some other way of getting the stub to the target
18488 system, and you can put it somewhere in memory where it won't get
18489 clobbered by the download.
18491 @item target sim @r{[}@var{simargs}@r{]} @dots{}
18492 @cindex built-in simulator target
18493 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
18501 works; however, you cannot assume that a specific memory map, device
18502 drivers, or even basic I/O is available, although some simulators do
18503 provide these. For info about any processor-specific simulator details,
18504 see the appropriate section in @ref{Embedded Processors, ,Embedded
18507 @item target native
18508 @cindex native target
18509 Setup for local/native process debugging. Useful to make the
18510 @code{run} command spawn native processes (likewise @code{attach},
18511 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
18512 (@pxref{set auto-connect-native-target}).
18516 Different targets are available on different configurations of @value{GDBN};
18517 your configuration may have more or fewer targets.
18519 Many remote targets require you to download the executable's code once
18520 you've successfully established a connection. You may wish to control
18521 various aspects of this process.
18526 @kindex set hash@r{, for remote monitors}
18527 @cindex hash mark while downloading
18528 This command controls whether a hash mark @samp{#} is displayed while
18529 downloading a file to the remote monitor. If on, a hash mark is
18530 displayed after each S-record is successfully downloaded to the
18534 @kindex show hash@r{, for remote monitors}
18535 Show the current status of displaying the hash mark.
18537 @item set debug monitor
18538 @kindex set debug monitor
18539 @cindex display remote monitor communications
18540 Enable or disable display of communications messages between
18541 @value{GDBN} and the remote monitor.
18543 @item show debug monitor
18544 @kindex show debug monitor
18545 Show the current status of displaying communications between
18546 @value{GDBN} and the remote monitor.
18551 @kindex load @var{filename}
18552 @item load @var{filename}
18554 Depending on what remote debugging facilities are configured into
18555 @value{GDBN}, the @code{load} command may be available. Where it exists, it
18556 is meant to make @var{filename} (an executable) available for debugging
18557 on the remote system---by downloading, or dynamic linking, for example.
18558 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
18559 the @code{add-symbol-file} command.
18561 If your @value{GDBN} does not have a @code{load} command, attempting to
18562 execute it gets the error message ``@code{You can't do that when your
18563 target is @dots{}}''
18565 The file is loaded at whatever address is specified in the executable.
18566 For some object file formats, you can specify the load address when you
18567 link the program; for other formats, like a.out, the object file format
18568 specifies a fixed address.
18569 @c FIXME! This would be a good place for an xref to the GNU linker doc.
18571 Depending on the remote side capabilities, @value{GDBN} may be able to
18572 load programs into flash memory.
18574 @code{load} does not repeat if you press @key{RET} again after using it.
18578 @section Choosing Target Byte Order
18580 @cindex choosing target byte order
18581 @cindex target byte order
18583 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
18584 offer the ability to run either big-endian or little-endian byte
18585 orders. Usually the executable or symbol will include a bit to
18586 designate the endian-ness, and you will not need to worry about
18587 which to use. However, you may still find it useful to adjust
18588 @value{GDBN}'s idea of processor endian-ness manually.
18592 @item set endian big
18593 Instruct @value{GDBN} to assume the target is big-endian.
18595 @item set endian little
18596 Instruct @value{GDBN} to assume the target is little-endian.
18598 @item set endian auto
18599 Instruct @value{GDBN} to use the byte order associated with the
18603 Display @value{GDBN}'s current idea of the target byte order.
18607 Note that these commands merely adjust interpretation of symbolic
18608 data on the host, and that they have absolutely no effect on the
18612 @node Remote Debugging
18613 @chapter Debugging Remote Programs
18614 @cindex remote debugging
18616 If you are trying to debug a program running on a machine that cannot run
18617 @value{GDBN} in the usual way, it is often useful to use remote debugging.
18618 For example, you might use remote debugging on an operating system kernel,
18619 or on a small system which does not have a general purpose operating system
18620 powerful enough to run a full-featured debugger.
18622 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
18623 to make this work with particular debugging targets. In addition,
18624 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
18625 but not specific to any particular target system) which you can use if you
18626 write the remote stubs---the code that runs on the remote system to
18627 communicate with @value{GDBN}.
18629 Other remote targets may be available in your
18630 configuration of @value{GDBN}; use @code{help target} to list them.
18633 * Connecting:: Connecting to a remote target
18634 * File Transfer:: Sending files to a remote system
18635 * Server:: Using the gdbserver program
18636 * Remote Configuration:: Remote configuration
18637 * Remote Stub:: Implementing a remote stub
18641 @section Connecting to a Remote Target
18643 On the @value{GDBN} host machine, you will need an unstripped copy of
18644 your program, since @value{GDBN} needs symbol and debugging information.
18645 Start up @value{GDBN} as usual, using the name of the local copy of your
18646 program as the first argument.
18648 @cindex @code{target remote}
18649 @value{GDBN} can communicate with the target over a serial line, or
18650 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
18651 each case, @value{GDBN} uses the same protocol for debugging your
18652 program; only the medium carrying the debugging packets varies. The
18653 @code{target remote} command establishes a connection to the target.
18654 Its arguments indicate which medium to use:
18658 @item target remote @var{serial-device}
18659 @cindex serial line, @code{target remote}
18660 Use @var{serial-device} to communicate with the target. For example,
18661 to use a serial line connected to the device named @file{/dev/ttyb}:
18664 target remote /dev/ttyb
18667 If you're using a serial line, you may want to give @value{GDBN} the
18668 @samp{--baud} option, or use the @code{set serial baud} command
18669 (@pxref{Remote Configuration, set serial baud}) before the
18670 @code{target} command.
18672 @item target remote @code{@var{host}:@var{port}}
18673 @itemx target remote @code{tcp:@var{host}:@var{port}}
18674 @cindex @acronym{TCP} port, @code{target remote}
18675 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
18676 The @var{host} may be either a host name or a numeric @acronym{IP}
18677 address; @var{port} must be a decimal number. The @var{host} could be
18678 the target machine itself, if it is directly connected to the net, or
18679 it might be a terminal server which in turn has a serial line to the
18682 For example, to connect to port 2828 on a terminal server named
18686 target remote manyfarms:2828
18689 If your remote target is actually running on the same machine as your
18690 debugger session (e.g.@: a simulator for your target running on the
18691 same host), you can omit the hostname. For example, to connect to
18692 port 1234 on your local machine:
18695 target remote :1234
18699 Note that the colon is still required here.
18701 @item target remote @code{udp:@var{host}:@var{port}}
18702 @cindex @acronym{UDP} port, @code{target remote}
18703 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
18704 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
18707 target remote udp:manyfarms:2828
18710 When using a @acronym{UDP} connection for remote debugging, you should
18711 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
18712 can silently drop packets on busy or unreliable networks, which will
18713 cause havoc with your debugging session.
18715 @item target remote | @var{command}
18716 @cindex pipe, @code{target remote} to
18717 Run @var{command} in the background and communicate with it using a
18718 pipe. The @var{command} is a shell command, to be parsed and expanded
18719 by the system's command shell, @code{/bin/sh}; it should expect remote
18720 protocol packets on its standard input, and send replies on its
18721 standard output. You could use this to run a stand-alone simulator
18722 that speaks the remote debugging protocol, to make net connections
18723 using programs like @code{ssh}, or for other similar tricks.
18725 If @var{command} closes its standard output (perhaps by exiting),
18726 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
18727 program has already exited, this will have no effect.)
18731 Once the connection has been established, you can use all the usual
18732 commands to examine and change data. The remote program is already
18733 running; you can use @kbd{step} and @kbd{continue}, and you do not
18734 need to use @kbd{run}.
18736 @cindex interrupting remote programs
18737 @cindex remote programs, interrupting
18738 Whenever @value{GDBN} is waiting for the remote program, if you type the
18739 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
18740 program. This may or may not succeed, depending in part on the hardware
18741 and the serial drivers the remote system uses. If you type the
18742 interrupt character once again, @value{GDBN} displays this prompt:
18745 Interrupted while waiting for the program.
18746 Give up (and stop debugging it)? (y or n)
18749 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
18750 (If you decide you want to try again later, you can use @samp{target
18751 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
18752 goes back to waiting.
18755 @kindex detach (remote)
18757 When you have finished debugging the remote program, you can use the
18758 @code{detach} command to release it from @value{GDBN} control.
18759 Detaching from the target normally resumes its execution, but the results
18760 will depend on your particular remote stub. After the @code{detach}
18761 command, @value{GDBN} is free to connect to another target.
18765 The @code{disconnect} command behaves like @code{detach}, except that
18766 the target is generally not resumed. It will wait for @value{GDBN}
18767 (this instance or another one) to connect and continue debugging. After
18768 the @code{disconnect} command, @value{GDBN} is again free to connect to
18771 @cindex send command to remote monitor
18772 @cindex extend @value{GDBN} for remote targets
18773 @cindex add new commands for external monitor
18775 @item monitor @var{cmd}
18776 This command allows you to send arbitrary commands directly to the
18777 remote monitor. Since @value{GDBN} doesn't care about the commands it
18778 sends like this, this command is the way to extend @value{GDBN}---you
18779 can add new commands that only the external monitor will understand
18783 @node File Transfer
18784 @section Sending files to a remote system
18785 @cindex remote target, file transfer
18786 @cindex file transfer
18787 @cindex sending files to remote systems
18789 Some remote targets offer the ability to transfer files over the same
18790 connection used to communicate with @value{GDBN}. This is convenient
18791 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
18792 running @code{gdbserver} over a network interface. For other targets,
18793 e.g.@: embedded devices with only a single serial port, this may be
18794 the only way to upload or download files.
18796 Not all remote targets support these commands.
18800 @item remote put @var{hostfile} @var{targetfile}
18801 Copy file @var{hostfile} from the host system (the machine running
18802 @value{GDBN}) to @var{targetfile} on the target system.
18805 @item remote get @var{targetfile} @var{hostfile}
18806 Copy file @var{targetfile} from the target system to @var{hostfile}
18807 on the host system.
18809 @kindex remote delete
18810 @item remote delete @var{targetfile}
18811 Delete @var{targetfile} from the target system.
18816 @section Using the @code{gdbserver} Program
18819 @cindex remote connection without stubs
18820 @code{gdbserver} is a control program for Unix-like systems, which
18821 allows you to connect your program with a remote @value{GDBN} via
18822 @code{target remote}---but without linking in the usual debugging stub.
18824 @code{gdbserver} is not a complete replacement for the debugging stubs,
18825 because it requires essentially the same operating-system facilities
18826 that @value{GDBN} itself does. In fact, a system that can run
18827 @code{gdbserver} to connect to a remote @value{GDBN} could also run
18828 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
18829 because it is a much smaller program than @value{GDBN} itself. It is
18830 also easier to port than all of @value{GDBN}, so you may be able to get
18831 started more quickly on a new system by using @code{gdbserver}.
18832 Finally, if you develop code for real-time systems, you may find that
18833 the tradeoffs involved in real-time operation make it more convenient to
18834 do as much development work as possible on another system, for example
18835 by cross-compiling. You can use @code{gdbserver} to make a similar
18836 choice for debugging.
18838 @value{GDBN} and @code{gdbserver} communicate via either a serial line
18839 or a TCP connection, using the standard @value{GDBN} remote serial
18843 @emph{Warning:} @code{gdbserver} does not have any built-in security.
18844 Do not run @code{gdbserver} connected to any public network; a
18845 @value{GDBN} connection to @code{gdbserver} provides access to the
18846 target system with the same privileges as the user running
18850 @subsection Running @code{gdbserver}
18851 @cindex arguments, to @code{gdbserver}
18852 @cindex @code{gdbserver}, command-line arguments
18854 Run @code{gdbserver} on the target system. You need a copy of the
18855 program you want to debug, including any libraries it requires.
18856 @code{gdbserver} does not need your program's symbol table, so you can
18857 strip the program if necessary to save space. @value{GDBN} on the host
18858 system does all the symbol handling.
18860 To use the server, you must tell it how to communicate with @value{GDBN};
18861 the name of your program; and the arguments for your program. The usual
18865 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
18868 @var{comm} is either a device name (to use a serial line), or a TCP
18869 hostname and portnumber, or @code{-} or @code{stdio} to use
18870 stdin/stdout of @code{gdbserver}.
18871 For example, to debug Emacs with the argument
18872 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
18876 target> gdbserver /dev/com1 emacs foo.txt
18879 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
18882 To use a TCP connection instead of a serial line:
18885 target> gdbserver host:2345 emacs foo.txt
18888 The only difference from the previous example is the first argument,
18889 specifying that you are communicating with the host @value{GDBN} via
18890 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
18891 expect a TCP connection from machine @samp{host} to local TCP port 2345.
18892 (Currently, the @samp{host} part is ignored.) You can choose any number
18893 you want for the port number as long as it does not conflict with any
18894 TCP ports already in use on the target system (for example, @code{23} is
18895 reserved for @code{telnet}).@footnote{If you choose a port number that
18896 conflicts with another service, @code{gdbserver} prints an error message
18897 and exits.} You must use the same port number with the host @value{GDBN}
18898 @code{target remote} command.
18900 The @code{stdio} connection is useful when starting @code{gdbserver}
18904 (gdb) target remote | ssh -T hostname gdbserver - hello
18907 The @samp{-T} option to ssh is provided because we don't need a remote pty,
18908 and we don't want escape-character handling. Ssh does this by default when
18909 a command is provided, the flag is provided to make it explicit.
18910 You could elide it if you want to.
18912 Programs started with stdio-connected gdbserver have @file{/dev/null} for
18913 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
18914 display through a pipe connected to gdbserver.
18915 Both @code{stdout} and @code{stderr} use the same pipe.
18917 @subsubsection Attaching to a Running Program
18918 @cindex attach to a program, @code{gdbserver}
18919 @cindex @option{--attach}, @code{gdbserver} option
18921 On some targets, @code{gdbserver} can also attach to running programs.
18922 This is accomplished via the @code{--attach} argument. The syntax is:
18925 target> gdbserver --attach @var{comm} @var{pid}
18928 @var{pid} is the process ID of a currently running process. It isn't necessary
18929 to point @code{gdbserver} at a binary for the running process.
18932 You can debug processes by name instead of process ID if your target has the
18933 @code{pidof} utility:
18936 target> gdbserver --attach @var{comm} `pidof @var{program}`
18939 In case more than one copy of @var{program} is running, or @var{program}
18940 has multiple threads, most versions of @code{pidof} support the
18941 @code{-s} option to only return the first process ID.
18943 @subsubsection Multi-Process Mode for @code{gdbserver}
18944 @cindex @code{gdbserver}, multiple processes
18945 @cindex multiple processes with @code{gdbserver}
18947 When you connect to @code{gdbserver} using @code{target remote},
18948 @code{gdbserver} debugs the specified program only once. When the
18949 program exits, or you detach from it, @value{GDBN} closes the connection
18950 and @code{gdbserver} exits.
18952 If you connect using @kbd{target extended-remote}, @code{gdbserver}
18953 enters multi-process mode. When the debugged program exits, or you
18954 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
18955 though no program is running. The @code{run} and @code{attach}
18956 commands instruct @code{gdbserver} to run or attach to a new program.
18957 The @code{run} command uses @code{set remote exec-file} (@pxref{set
18958 remote exec-file}) to select the program to run. Command line
18959 arguments are supported, except for wildcard expansion and I/O
18960 redirection (@pxref{Arguments}).
18962 @cindex @option{--multi}, @code{gdbserver} option
18963 To start @code{gdbserver} without supplying an initial command to run
18964 or process ID to attach, use the @option{--multi} command line option.
18965 Then you can connect using @kbd{target extended-remote} and start
18966 the program you want to debug.
18968 In multi-process mode @code{gdbserver} does not automatically exit unless you
18969 use the option @option{--once}. You can terminate it by using
18970 @code{monitor exit} (@pxref{Monitor Commands for gdbserver}). Note that the
18971 conditions under which @code{gdbserver} terminates depend on how @value{GDBN}
18972 connects to it (@kbd{target remote} or @kbd{target extended-remote}). The
18973 @option{--multi} option to @code{gdbserver} has no influence on that.
18975 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
18977 This section applies only when @code{gdbserver} is run to listen on a TCP port.
18979 @code{gdbserver} normally terminates after all of its debugged processes have
18980 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
18981 extended-remote}, @code{gdbserver} stays running even with no processes left.
18982 @value{GDBN} normally terminates the spawned debugged process on its exit,
18983 which normally also terminates @code{gdbserver} in the @kbd{target remote}
18984 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
18985 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
18986 stays running even in the @kbd{target remote} mode.
18988 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
18989 Such reconnecting is useful for features like @ref{disconnected tracing}. For
18990 completeness, at most one @value{GDBN} can be connected at a time.
18992 @cindex @option{--once}, @code{gdbserver} option
18993 By default, @code{gdbserver} keeps the listening TCP port open, so that
18994 subsequent connections are possible. However, if you start @code{gdbserver}
18995 with the @option{--once} option, it will stop listening for any further
18996 connection attempts after connecting to the first @value{GDBN} session. This
18997 means no further connections to @code{gdbserver} will be possible after the
18998 first one. It also means @code{gdbserver} will terminate after the first
18999 connection with remote @value{GDBN} has closed, even for unexpectedly closed
19000 connections and even in the @kbd{target extended-remote} mode. The
19001 @option{--once} option allows reusing the same port number for connecting to
19002 multiple instances of @code{gdbserver} running on the same host, since each
19003 instance closes its port after the first connection.
19005 @anchor{Other Command-Line Arguments for gdbserver}
19006 @subsubsection Other Command-Line Arguments for @code{gdbserver}
19008 @cindex @option{--debug}, @code{gdbserver} option
19009 The @option{--debug} option tells @code{gdbserver} to display extra
19010 status information about the debugging process.
19011 @cindex @option{--remote-debug}, @code{gdbserver} option
19012 The @option{--remote-debug} option tells @code{gdbserver} to display
19013 remote protocol debug output. These options are intended for
19014 @code{gdbserver} development and for bug reports to the developers.
19016 @cindex @option{--debug-format}, @code{gdbserver} option
19017 The @option{--debug-format=option1[,option2,...]} option tells
19018 @code{gdbserver} to include additional information in each output.
19019 Possible options are:
19023 Turn off all extra information in debugging output.
19025 Turn on all extra information in debugging output.
19027 Include a timestamp in each line of debugging output.
19030 Options are processed in order. Thus, for example, if @option{none}
19031 appears last then no additional information is added to debugging output.
19033 @cindex @option{--wrapper}, @code{gdbserver} option
19034 The @option{--wrapper} option specifies a wrapper to launch programs
19035 for debugging. The option should be followed by the name of the
19036 wrapper, then any command-line arguments to pass to the wrapper, then
19037 @kbd{--} indicating the end of the wrapper arguments.
19039 @code{gdbserver} runs the specified wrapper program with a combined
19040 command line including the wrapper arguments, then the name of the
19041 program to debug, then any arguments to the program. The wrapper
19042 runs until it executes your program, and then @value{GDBN} gains control.
19044 You can use any program that eventually calls @code{execve} with
19045 its arguments as a wrapper. Several standard Unix utilities do
19046 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
19047 with @code{exec "$@@"} will also work.
19049 For example, you can use @code{env} to pass an environment variable to
19050 the debugged program, without setting the variable in @code{gdbserver}'s
19054 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
19057 @subsection Connecting to @code{gdbserver}
19059 Run @value{GDBN} on the host system.
19061 First make sure you have the necessary symbol files. Load symbols for
19062 your application using the @code{file} command before you connect. Use
19063 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
19064 was compiled with the correct sysroot using @code{--with-sysroot}).
19066 The symbol file and target libraries must exactly match the executable
19067 and libraries on the target, with one exception: the files on the host
19068 system should not be stripped, even if the files on the target system
19069 are. Mismatched or missing files will lead to confusing results
19070 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
19071 files may also prevent @code{gdbserver} from debugging multi-threaded
19074 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
19075 For TCP connections, you must start up @code{gdbserver} prior to using
19076 the @code{target remote} command. Otherwise you may get an error whose
19077 text depends on the host system, but which usually looks something like
19078 @samp{Connection refused}. Don't use the @code{load}
19079 command in @value{GDBN} when using @code{gdbserver}, since the program is
19080 already on the target.
19082 @subsection Monitor Commands for @code{gdbserver}
19083 @cindex monitor commands, for @code{gdbserver}
19084 @anchor{Monitor Commands for gdbserver}
19086 During a @value{GDBN} session using @code{gdbserver}, you can use the
19087 @code{monitor} command to send special requests to @code{gdbserver}.
19088 Here are the available commands.
19092 List the available monitor commands.
19094 @item monitor set debug 0
19095 @itemx monitor set debug 1
19096 Disable or enable general debugging messages.
19098 @item monitor set remote-debug 0
19099 @itemx monitor set remote-debug 1
19100 Disable or enable specific debugging messages associated with the remote
19101 protocol (@pxref{Remote Protocol}).
19103 @item monitor set debug-format option1@r{[},option2,...@r{]}
19104 Specify additional text to add to debugging messages.
19105 Possible options are:
19109 Turn off all extra information in debugging output.
19111 Turn on all extra information in debugging output.
19113 Include a timestamp in each line of debugging output.
19116 Options are processed in order. Thus, for example, if @option{none}
19117 appears last then no additional information is added to debugging output.
19119 @item monitor set libthread-db-search-path [PATH]
19120 @cindex gdbserver, search path for @code{libthread_db}
19121 When this command is issued, @var{path} is a colon-separated list of
19122 directories to search for @code{libthread_db} (@pxref{Threads,,set
19123 libthread-db-search-path}). If you omit @var{path},
19124 @samp{libthread-db-search-path} will be reset to its default value.
19126 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
19127 not supported in @code{gdbserver}.
19130 Tell gdbserver to exit immediately. This command should be followed by
19131 @code{disconnect} to close the debugging session. @code{gdbserver} will
19132 detach from any attached processes and kill any processes it created.
19133 Use @code{monitor exit} to terminate @code{gdbserver} at the end
19134 of a multi-process mode debug session.
19138 @subsection Tracepoints support in @code{gdbserver}
19139 @cindex tracepoints support in @code{gdbserver}
19141 On some targets, @code{gdbserver} supports tracepoints, fast
19142 tracepoints and static tracepoints.
19144 For fast or static tracepoints to work, a special library called the
19145 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
19146 This library is built and distributed as an integral part of
19147 @code{gdbserver}. In addition, support for static tracepoints
19148 requires building the in-process agent library with static tracepoints
19149 support. At present, the UST (LTTng Userspace Tracer,
19150 @url{http://lttng.org/ust}) tracing engine is supported. This support
19151 is automatically available if UST development headers are found in the
19152 standard include path when @code{gdbserver} is built, or if
19153 @code{gdbserver} was explicitly configured using @option{--with-ust}
19154 to point at such headers. You can explicitly disable the support
19155 using @option{--with-ust=no}.
19157 There are several ways to load the in-process agent in your program:
19160 @item Specifying it as dependency at link time
19162 You can link your program dynamically with the in-process agent
19163 library. On most systems, this is accomplished by adding
19164 @code{-linproctrace} to the link command.
19166 @item Using the system's preloading mechanisms
19168 You can force loading the in-process agent at startup time by using
19169 your system's support for preloading shared libraries. Many Unixes
19170 support the concept of preloading user defined libraries. In most
19171 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
19172 in the environment. See also the description of @code{gdbserver}'s
19173 @option{--wrapper} command line option.
19175 @item Using @value{GDBN} to force loading the agent at run time
19177 On some systems, you can force the inferior to load a shared library,
19178 by calling a dynamic loader function in the inferior that takes care
19179 of dynamically looking up and loading a shared library. On most Unix
19180 systems, the function is @code{dlopen}. You'll use the @code{call}
19181 command for that. For example:
19184 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
19187 Note that on most Unix systems, for the @code{dlopen} function to be
19188 available, the program needs to be linked with @code{-ldl}.
19191 On systems that have a userspace dynamic loader, like most Unix
19192 systems, when you connect to @code{gdbserver} using @code{target
19193 remote}, you'll find that the program is stopped at the dynamic
19194 loader's entry point, and no shared library has been loaded in the
19195 program's address space yet, including the in-process agent. In that
19196 case, before being able to use any of the fast or static tracepoints
19197 features, you need to let the loader run and load the shared
19198 libraries. The simplest way to do that is to run the program to the
19199 main procedure. E.g., if debugging a C or C@t{++} program, start
19200 @code{gdbserver} like so:
19203 $ gdbserver :9999 myprogram
19206 Start GDB and connect to @code{gdbserver} like so, and run to main:
19210 (@value{GDBP}) target remote myhost:9999
19211 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
19212 (@value{GDBP}) b main
19213 (@value{GDBP}) continue
19216 The in-process tracing agent library should now be loaded into the
19217 process; you can confirm it with the @code{info sharedlibrary}
19218 command, which will list @file{libinproctrace.so} as loaded in the
19219 process. You are now ready to install fast tracepoints, list static
19220 tracepoint markers, probe static tracepoints markers, and start
19223 @node Remote Configuration
19224 @section Remote Configuration
19227 @kindex show remote
19228 This section documents the configuration options available when
19229 debugging remote programs. For the options related to the File I/O
19230 extensions of the remote protocol, see @ref{system,
19231 system-call-allowed}.
19234 @item set remoteaddresssize @var{bits}
19235 @cindex address size for remote targets
19236 @cindex bits in remote address
19237 Set the maximum size of address in a memory packet to the specified
19238 number of bits. @value{GDBN} will mask off the address bits above
19239 that number, when it passes addresses to the remote target. The
19240 default value is the number of bits in the target's address.
19242 @item show remoteaddresssize
19243 Show the current value of remote address size in bits.
19245 @item set serial baud @var{n}
19246 @cindex baud rate for remote targets
19247 Set the baud rate for the remote serial I/O to @var{n} baud. The
19248 value is used to set the speed of the serial port used for debugging
19251 @item show serial baud
19252 Show the current speed of the remote connection.
19254 @item set remotebreak
19255 @cindex interrupt remote programs
19256 @cindex BREAK signal instead of Ctrl-C
19257 @anchor{set remotebreak}
19258 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
19259 when you type @kbd{Ctrl-c} to interrupt the program running
19260 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
19261 character instead. The default is off, since most remote systems
19262 expect to see @samp{Ctrl-C} as the interrupt signal.
19264 @item show remotebreak
19265 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
19266 interrupt the remote program.
19268 @item set remoteflow on
19269 @itemx set remoteflow off
19270 @kindex set remoteflow
19271 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
19272 on the serial port used to communicate to the remote target.
19274 @item show remoteflow
19275 @kindex show remoteflow
19276 Show the current setting of hardware flow control.
19278 @item set remotelogbase @var{base}
19279 Set the base (a.k.a.@: radix) of logging serial protocol
19280 communications to @var{base}. Supported values of @var{base} are:
19281 @code{ascii}, @code{octal}, and @code{hex}. The default is
19284 @item show remotelogbase
19285 Show the current setting of the radix for logging remote serial
19288 @item set remotelogfile @var{file}
19289 @cindex record serial communications on file
19290 Record remote serial communications on the named @var{file}. The
19291 default is not to record at all.
19293 @item show remotelogfile.
19294 Show the current setting of the file name on which to record the
19295 serial communications.
19297 @item set remotetimeout @var{num}
19298 @cindex timeout for serial communications
19299 @cindex remote timeout
19300 Set the timeout limit to wait for the remote target to respond to
19301 @var{num} seconds. The default is 2 seconds.
19303 @item show remotetimeout
19304 Show the current number of seconds to wait for the remote target
19307 @cindex limit hardware breakpoints and watchpoints
19308 @cindex remote target, limit break- and watchpoints
19309 @anchor{set remote hardware-watchpoint-limit}
19310 @anchor{set remote hardware-breakpoint-limit}
19311 @item set remote hardware-watchpoint-limit @var{limit}
19312 @itemx set remote hardware-breakpoint-limit @var{limit}
19313 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
19314 watchpoints. A limit of -1, the default, is treated as unlimited.
19316 @cindex limit hardware watchpoints length
19317 @cindex remote target, limit watchpoints length
19318 @anchor{set remote hardware-watchpoint-length-limit}
19319 @item set remote hardware-watchpoint-length-limit @var{limit}
19320 Restrict @value{GDBN} to using @var{limit} bytes for the maximum length of
19321 a remote hardware watchpoint. A limit of -1, the default, is treated
19324 @item show remote hardware-watchpoint-length-limit
19325 Show the current limit (in bytes) of the maximum length of
19326 a remote hardware watchpoint.
19328 @item set remote exec-file @var{filename}
19329 @itemx show remote exec-file
19330 @anchor{set remote exec-file}
19331 @cindex executable file, for remote target
19332 Select the file used for @code{run} with @code{target
19333 extended-remote}. This should be set to a filename valid on the
19334 target system. If it is not set, the target will use a default
19335 filename (e.g.@: the last program run).
19337 @item set remote interrupt-sequence
19338 @cindex interrupt remote programs
19339 @cindex select Ctrl-C, BREAK or BREAK-g
19340 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
19341 @samp{BREAK-g} as the
19342 sequence to the remote target in order to interrupt the execution.
19343 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
19344 is high level of serial line for some certain time.
19345 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
19346 It is @code{BREAK} signal followed by character @code{g}.
19348 @item show interrupt-sequence
19349 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
19350 is sent by @value{GDBN} to interrupt the remote program.
19351 @code{BREAK-g} is BREAK signal followed by @code{g} and
19352 also known as Magic SysRq g.
19354 @item set remote interrupt-on-connect
19355 @cindex send interrupt-sequence on start
19356 Specify whether interrupt-sequence is sent to remote target when
19357 @value{GDBN} connects to it. This is mostly needed when you debug
19358 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
19359 which is known as Magic SysRq g in order to connect @value{GDBN}.
19361 @item show interrupt-on-connect
19362 Show whether interrupt-sequence is sent
19363 to remote target when @value{GDBN} connects to it.
19367 @item set tcp auto-retry on
19368 @cindex auto-retry, for remote TCP target
19369 Enable auto-retry for remote TCP connections. This is useful if the remote
19370 debugging agent is launched in parallel with @value{GDBN}; there is a race
19371 condition because the agent may not become ready to accept the connection
19372 before @value{GDBN} attempts to connect. When auto-retry is
19373 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
19374 to establish the connection using the timeout specified by
19375 @code{set tcp connect-timeout}.
19377 @item set tcp auto-retry off
19378 Do not auto-retry failed TCP connections.
19380 @item show tcp auto-retry
19381 Show the current auto-retry setting.
19383 @item set tcp connect-timeout @var{seconds}
19384 @itemx set tcp connect-timeout unlimited
19385 @cindex connection timeout, for remote TCP target
19386 @cindex timeout, for remote target connection
19387 Set the timeout for establishing a TCP connection to the remote target to
19388 @var{seconds}. The timeout affects both polling to retry failed connections
19389 (enabled by @code{set tcp auto-retry on}) and waiting for connections
19390 that are merely slow to complete, and represents an approximate cumulative
19391 value. If @var{seconds} is @code{unlimited}, there is no timeout and
19392 @value{GDBN} will keep attempting to establish a connection forever,
19393 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
19395 @item show tcp connect-timeout
19396 Show the current connection timeout setting.
19399 @cindex remote packets, enabling and disabling
19400 The @value{GDBN} remote protocol autodetects the packets supported by
19401 your debugging stub. If you need to override the autodetection, you
19402 can use these commands to enable or disable individual packets. Each
19403 packet can be set to @samp{on} (the remote target supports this
19404 packet), @samp{off} (the remote target does not support this packet),
19405 or @samp{auto} (detect remote target support for this packet). They
19406 all default to @samp{auto}. For more information about each packet,
19407 see @ref{Remote Protocol}.
19409 During normal use, you should not have to use any of these commands.
19410 If you do, that may be a bug in your remote debugging stub, or a bug
19411 in @value{GDBN}. You may want to report the problem to the
19412 @value{GDBN} developers.
19414 For each packet @var{name}, the command to enable or disable the
19415 packet is @code{set remote @var{name}-packet}. The available settings
19418 @multitable @columnfractions 0.28 0.32 0.25
19421 @tab Related Features
19423 @item @code{fetch-register}
19425 @tab @code{info registers}
19427 @item @code{set-register}
19431 @item @code{binary-download}
19433 @tab @code{load}, @code{set}
19435 @item @code{read-aux-vector}
19436 @tab @code{qXfer:auxv:read}
19437 @tab @code{info auxv}
19439 @item @code{symbol-lookup}
19440 @tab @code{qSymbol}
19441 @tab Detecting multiple threads
19443 @item @code{attach}
19444 @tab @code{vAttach}
19447 @item @code{verbose-resume}
19449 @tab Stepping or resuming multiple threads
19455 @item @code{software-breakpoint}
19459 @item @code{hardware-breakpoint}
19463 @item @code{write-watchpoint}
19467 @item @code{read-watchpoint}
19471 @item @code{access-watchpoint}
19475 @item @code{target-features}
19476 @tab @code{qXfer:features:read}
19477 @tab @code{set architecture}
19479 @item @code{library-info}
19480 @tab @code{qXfer:libraries:read}
19481 @tab @code{info sharedlibrary}
19483 @item @code{memory-map}
19484 @tab @code{qXfer:memory-map:read}
19485 @tab @code{info mem}
19487 @item @code{read-sdata-object}
19488 @tab @code{qXfer:sdata:read}
19489 @tab @code{print $_sdata}
19491 @item @code{read-spu-object}
19492 @tab @code{qXfer:spu:read}
19493 @tab @code{info spu}
19495 @item @code{write-spu-object}
19496 @tab @code{qXfer:spu:write}
19497 @tab @code{info spu}
19499 @item @code{read-siginfo-object}
19500 @tab @code{qXfer:siginfo:read}
19501 @tab @code{print $_siginfo}
19503 @item @code{write-siginfo-object}
19504 @tab @code{qXfer:siginfo:write}
19505 @tab @code{set $_siginfo}
19507 @item @code{threads}
19508 @tab @code{qXfer:threads:read}
19509 @tab @code{info threads}
19511 @item @code{get-thread-local-@*storage-address}
19512 @tab @code{qGetTLSAddr}
19513 @tab Displaying @code{__thread} variables
19515 @item @code{get-thread-information-block-address}
19516 @tab @code{qGetTIBAddr}
19517 @tab Display MS-Windows Thread Information Block.
19519 @item @code{search-memory}
19520 @tab @code{qSearch:memory}
19523 @item @code{supported-packets}
19524 @tab @code{qSupported}
19525 @tab Remote communications parameters
19527 @item @code{pass-signals}
19528 @tab @code{QPassSignals}
19529 @tab @code{handle @var{signal}}
19531 @item @code{program-signals}
19532 @tab @code{QProgramSignals}
19533 @tab @code{handle @var{signal}}
19535 @item @code{hostio-close-packet}
19536 @tab @code{vFile:close}
19537 @tab @code{remote get}, @code{remote put}
19539 @item @code{hostio-open-packet}
19540 @tab @code{vFile:open}
19541 @tab @code{remote get}, @code{remote put}
19543 @item @code{hostio-pread-packet}
19544 @tab @code{vFile:pread}
19545 @tab @code{remote get}, @code{remote put}
19547 @item @code{hostio-pwrite-packet}
19548 @tab @code{vFile:pwrite}
19549 @tab @code{remote get}, @code{remote put}
19551 @item @code{hostio-unlink-packet}
19552 @tab @code{vFile:unlink}
19553 @tab @code{remote delete}
19555 @item @code{hostio-readlink-packet}
19556 @tab @code{vFile:readlink}
19559 @item @code{noack-packet}
19560 @tab @code{QStartNoAckMode}
19561 @tab Packet acknowledgment
19563 @item @code{osdata}
19564 @tab @code{qXfer:osdata:read}
19565 @tab @code{info os}
19567 @item @code{query-attached}
19568 @tab @code{qAttached}
19569 @tab Querying remote process attach state.
19571 @item @code{trace-buffer-size}
19572 @tab @code{QTBuffer:size}
19573 @tab @code{set trace-buffer-size}
19575 @item @code{trace-status}
19576 @tab @code{qTStatus}
19577 @tab @code{tstatus}
19579 @item @code{traceframe-info}
19580 @tab @code{qXfer:traceframe-info:read}
19581 @tab Traceframe info
19583 @item @code{install-in-trace}
19584 @tab @code{InstallInTrace}
19585 @tab Install tracepoint in tracing
19587 @item @code{disable-randomization}
19588 @tab @code{QDisableRandomization}
19589 @tab @code{set disable-randomization}
19591 @item @code{conditional-breakpoints-packet}
19592 @tab @code{Z0 and Z1}
19593 @tab @code{Support for target-side breakpoint condition evaluation}
19597 @section Implementing a Remote Stub
19599 @cindex debugging stub, example
19600 @cindex remote stub, example
19601 @cindex stub example, remote debugging
19602 The stub files provided with @value{GDBN} implement the target side of the
19603 communication protocol, and the @value{GDBN} side is implemented in the
19604 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
19605 these subroutines to communicate, and ignore the details. (If you're
19606 implementing your own stub file, you can still ignore the details: start
19607 with one of the existing stub files. @file{sparc-stub.c} is the best
19608 organized, and therefore the easiest to read.)
19610 @cindex remote serial debugging, overview
19611 To debug a program running on another machine (the debugging
19612 @dfn{target} machine), you must first arrange for all the usual
19613 prerequisites for the program to run by itself. For example, for a C
19618 A startup routine to set up the C runtime environment; these usually
19619 have a name like @file{crt0}. The startup routine may be supplied by
19620 your hardware supplier, or you may have to write your own.
19623 A C subroutine library to support your program's
19624 subroutine calls, notably managing input and output.
19627 A way of getting your program to the other machine---for example, a
19628 download program. These are often supplied by the hardware
19629 manufacturer, but you may have to write your own from hardware
19633 The next step is to arrange for your program to use a serial port to
19634 communicate with the machine where @value{GDBN} is running (the @dfn{host}
19635 machine). In general terms, the scheme looks like this:
19639 @value{GDBN} already understands how to use this protocol; when everything
19640 else is set up, you can simply use the @samp{target remote} command
19641 (@pxref{Targets,,Specifying a Debugging Target}).
19643 @item On the target,
19644 you must link with your program a few special-purpose subroutines that
19645 implement the @value{GDBN} remote serial protocol. The file containing these
19646 subroutines is called a @dfn{debugging stub}.
19648 On certain remote targets, you can use an auxiliary program
19649 @code{gdbserver} instead of linking a stub into your program.
19650 @xref{Server,,Using the @code{gdbserver} Program}, for details.
19653 The debugging stub is specific to the architecture of the remote
19654 machine; for example, use @file{sparc-stub.c} to debug programs on
19657 @cindex remote serial stub list
19658 These working remote stubs are distributed with @value{GDBN}:
19663 @cindex @file{i386-stub.c}
19666 For Intel 386 and compatible architectures.
19669 @cindex @file{m68k-stub.c}
19670 @cindex Motorola 680x0
19672 For Motorola 680x0 architectures.
19675 @cindex @file{sh-stub.c}
19678 For Renesas SH architectures.
19681 @cindex @file{sparc-stub.c}
19683 For @sc{sparc} architectures.
19685 @item sparcl-stub.c
19686 @cindex @file{sparcl-stub.c}
19689 For Fujitsu @sc{sparclite} architectures.
19693 The @file{README} file in the @value{GDBN} distribution may list other
19694 recently added stubs.
19697 * Stub Contents:: What the stub can do for you
19698 * Bootstrapping:: What you must do for the stub
19699 * Debug Session:: Putting it all together
19702 @node Stub Contents
19703 @subsection What the Stub Can Do for You
19705 @cindex remote serial stub
19706 The debugging stub for your architecture supplies these three
19710 @item set_debug_traps
19711 @findex set_debug_traps
19712 @cindex remote serial stub, initialization
19713 This routine arranges for @code{handle_exception} to run when your
19714 program stops. You must call this subroutine explicitly in your
19715 program's startup code.
19717 @item handle_exception
19718 @findex handle_exception
19719 @cindex remote serial stub, main routine
19720 This is the central workhorse, but your program never calls it
19721 explicitly---the setup code arranges for @code{handle_exception} to
19722 run when a trap is triggered.
19724 @code{handle_exception} takes control when your program stops during
19725 execution (for example, on a breakpoint), and mediates communications
19726 with @value{GDBN} on the host machine. This is where the communications
19727 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
19728 representative on the target machine. It begins by sending summary
19729 information on the state of your program, then continues to execute,
19730 retrieving and transmitting any information @value{GDBN} needs, until you
19731 execute a @value{GDBN} command that makes your program resume; at that point,
19732 @code{handle_exception} returns control to your own code on the target
19736 @cindex @code{breakpoint} subroutine, remote
19737 Use this auxiliary subroutine to make your program contain a
19738 breakpoint. Depending on the particular situation, this may be the only
19739 way for @value{GDBN} to get control. For instance, if your target
19740 machine has some sort of interrupt button, you won't need to call this;
19741 pressing the interrupt button transfers control to
19742 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
19743 simply receiving characters on the serial port may also trigger a trap;
19744 again, in that situation, you don't need to call @code{breakpoint} from
19745 your own program---simply running @samp{target remote} from the host
19746 @value{GDBN} session gets control.
19748 Call @code{breakpoint} if none of these is true, or if you simply want
19749 to make certain your program stops at a predetermined point for the
19750 start of your debugging session.
19753 @node Bootstrapping
19754 @subsection What You Must Do for the Stub
19756 @cindex remote stub, support routines
19757 The debugging stubs that come with @value{GDBN} are set up for a particular
19758 chip architecture, but they have no information about the rest of your
19759 debugging target machine.
19761 First of all you need to tell the stub how to communicate with the
19765 @item int getDebugChar()
19766 @findex getDebugChar
19767 Write this subroutine to read a single character from the serial port.
19768 It may be identical to @code{getchar} for your target system; a
19769 different name is used to allow you to distinguish the two if you wish.
19771 @item void putDebugChar(int)
19772 @findex putDebugChar
19773 Write this subroutine to write a single character to the serial port.
19774 It may be identical to @code{putchar} for your target system; a
19775 different name is used to allow you to distinguish the two if you wish.
19778 @cindex control C, and remote debugging
19779 @cindex interrupting remote targets
19780 If you want @value{GDBN} to be able to stop your program while it is
19781 running, you need to use an interrupt-driven serial driver, and arrange
19782 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
19783 character). That is the character which @value{GDBN} uses to tell the
19784 remote system to stop.
19786 Getting the debugging target to return the proper status to @value{GDBN}
19787 probably requires changes to the standard stub; one quick and dirty way
19788 is to just execute a breakpoint instruction (the ``dirty'' part is that
19789 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
19791 Other routines you need to supply are:
19794 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
19795 @findex exceptionHandler
19796 Write this function to install @var{exception_address} in the exception
19797 handling tables. You need to do this because the stub does not have any
19798 way of knowing what the exception handling tables on your target system
19799 are like (for example, the processor's table might be in @sc{rom},
19800 containing entries which point to a table in @sc{ram}).
19801 The @var{exception_number} specifies the exception which should be changed;
19802 its meaning is architecture-dependent (for example, different numbers
19803 might represent divide by zero, misaligned access, etc). When this
19804 exception occurs, control should be transferred directly to
19805 @var{exception_address}, and the processor state (stack, registers,
19806 and so on) should be just as it is when a processor exception occurs. So if
19807 you want to use a jump instruction to reach @var{exception_address}, it
19808 should be a simple jump, not a jump to subroutine.
19810 For the 386, @var{exception_address} should be installed as an interrupt
19811 gate so that interrupts are masked while the handler runs. The gate
19812 should be at privilege level 0 (the most privileged level). The
19813 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
19814 help from @code{exceptionHandler}.
19816 @item void flush_i_cache()
19817 @findex flush_i_cache
19818 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
19819 instruction cache, if any, on your target machine. If there is no
19820 instruction cache, this subroutine may be a no-op.
19822 On target machines that have instruction caches, @value{GDBN} requires this
19823 function to make certain that the state of your program is stable.
19827 You must also make sure this library routine is available:
19830 @item void *memset(void *, int, int)
19832 This is the standard library function @code{memset} that sets an area of
19833 memory to a known value. If you have one of the free versions of
19834 @code{libc.a}, @code{memset} can be found there; otherwise, you must
19835 either obtain it from your hardware manufacturer, or write your own.
19838 If you do not use the GNU C compiler, you may need other standard
19839 library subroutines as well; this varies from one stub to another,
19840 but in general the stubs are likely to use any of the common library
19841 subroutines which @code{@value{NGCC}} generates as inline code.
19844 @node Debug Session
19845 @subsection Putting it All Together
19847 @cindex remote serial debugging summary
19848 In summary, when your program is ready to debug, you must follow these
19853 Make sure you have defined the supporting low-level routines
19854 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
19856 @code{getDebugChar}, @code{putDebugChar},
19857 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
19861 Insert these lines in your program's startup code, before the main
19862 procedure is called:
19869 On some machines, when a breakpoint trap is raised, the hardware
19870 automatically makes the PC point to the instruction after the
19871 breakpoint. If your machine doesn't do that, you may need to adjust
19872 @code{handle_exception} to arrange for it to return to the instruction
19873 after the breakpoint on this first invocation, so that your program
19874 doesn't keep hitting the initial breakpoint instead of making
19878 For the 680x0 stub only, you need to provide a variable called
19879 @code{exceptionHook}. Normally you just use:
19882 void (*exceptionHook)() = 0;
19886 but if before calling @code{set_debug_traps}, you set it to point to a
19887 function in your program, that function is called when
19888 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
19889 error). The function indicated by @code{exceptionHook} is called with
19890 one parameter: an @code{int} which is the exception number.
19893 Compile and link together: your program, the @value{GDBN} debugging stub for
19894 your target architecture, and the supporting subroutines.
19897 Make sure you have a serial connection between your target machine and
19898 the @value{GDBN} host, and identify the serial port on the host.
19901 @c The "remote" target now provides a `load' command, so we should
19902 @c document that. FIXME.
19903 Download your program to your target machine (or get it there by
19904 whatever means the manufacturer provides), and start it.
19907 Start @value{GDBN} on the host, and connect to the target
19908 (@pxref{Connecting,,Connecting to a Remote Target}).
19912 @node Configurations
19913 @chapter Configuration-Specific Information
19915 While nearly all @value{GDBN} commands are available for all native and
19916 cross versions of the debugger, there are some exceptions. This chapter
19917 describes things that are only available in certain configurations.
19919 There are three major categories of configurations: native
19920 configurations, where the host and target are the same, embedded
19921 operating system configurations, which are usually the same for several
19922 different processor architectures, and bare embedded processors, which
19923 are quite different from each other.
19928 * Embedded Processors::
19935 This section describes details specific to particular native
19940 * BSD libkvm Interface:: Debugging BSD kernel memory images
19941 * SVR4 Process Information:: SVR4 process information
19942 * DJGPP Native:: Features specific to the DJGPP port
19943 * Cygwin Native:: Features specific to the Cygwin port
19944 * Hurd Native:: Features specific to @sc{gnu} Hurd
19945 * Darwin:: Features specific to Darwin
19951 On HP-UX systems, if you refer to a function or variable name that
19952 begins with a dollar sign, @value{GDBN} searches for a user or system
19953 name first, before it searches for a convenience variable.
19956 @node BSD libkvm Interface
19957 @subsection BSD libkvm Interface
19960 @cindex kernel memory image
19961 @cindex kernel crash dump
19963 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
19964 interface that provides a uniform interface for accessing kernel virtual
19965 memory images, including live systems and crash dumps. @value{GDBN}
19966 uses this interface to allow you to debug live kernels and kernel crash
19967 dumps on many native BSD configurations. This is implemented as a
19968 special @code{kvm} debugging target. For debugging a live system, load
19969 the currently running kernel into @value{GDBN} and connect to the
19973 (@value{GDBP}) @b{target kvm}
19976 For debugging crash dumps, provide the file name of the crash dump as an
19980 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
19983 Once connected to the @code{kvm} target, the following commands are
19989 Set current context from the @dfn{Process Control Block} (PCB) address.
19992 Set current context from proc address. This command isn't available on
19993 modern FreeBSD systems.
19996 @node SVR4 Process Information
19997 @subsection SVR4 Process Information
19999 @cindex examine process image
20000 @cindex process info via @file{/proc}
20002 Many versions of SVR4 and compatible systems provide a facility called
20003 @samp{/proc} that can be used to examine the image of a running
20004 process using file-system subroutines.
20006 If @value{GDBN} is configured for an operating system with this
20007 facility, the command @code{info proc} is available to report
20008 information about the process running your program, or about any
20009 process running on your system. This includes, as of this writing,
20010 @sc{gnu}/Linux and Solaris, but not HP-UX, for example.
20012 This command may also work on core files that were created on a system
20013 that has the @samp{/proc} facility.
20019 @itemx info proc @var{process-id}
20020 Summarize available information about any running process. If a
20021 process ID is specified by @var{process-id}, display information about
20022 that process; otherwise display information about the program being
20023 debugged. The summary includes the debugged process ID, the command
20024 line used to invoke it, its current working directory, and its
20025 executable file's absolute file name.
20027 On some systems, @var{process-id} can be of the form
20028 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
20029 within a process. If the optional @var{pid} part is missing, it means
20030 a thread from the process being debugged (the leading @samp{/} still
20031 needs to be present, or else @value{GDBN} will interpret the number as
20032 a process ID rather than a thread ID).
20034 @item info proc cmdline
20035 @cindex info proc cmdline
20036 Show the original command line of the process. This command is
20037 specific to @sc{gnu}/Linux.
20039 @item info proc cwd
20040 @cindex info proc cwd
20041 Show the current working directory of the process. This command is
20042 specific to @sc{gnu}/Linux.
20044 @item info proc exe
20045 @cindex info proc exe
20046 Show the name of executable of the process. This command is specific
20049 @item info proc mappings
20050 @cindex memory address space mappings
20051 Report the memory address space ranges accessible in the program, with
20052 information on whether the process has read, write, or execute access
20053 rights to each range. On @sc{gnu}/Linux systems, each memory range
20054 includes the object file which is mapped to that range, instead of the
20055 memory access rights to that range.
20057 @item info proc stat
20058 @itemx info proc status
20059 @cindex process detailed status information
20060 These subcommands are specific to @sc{gnu}/Linux systems. They show
20061 the process-related information, including the user ID and group ID;
20062 how many threads are there in the process; its virtual memory usage;
20063 the signals that are pending, blocked, and ignored; its TTY; its
20064 consumption of system and user time; its stack size; its @samp{nice}
20065 value; etc. For more information, see the @samp{proc} man page
20066 (type @kbd{man 5 proc} from your shell prompt).
20068 @item info proc all
20069 Show all the information about the process described under all of the
20070 above @code{info proc} subcommands.
20073 @comment These sub-options of 'info proc' were not included when
20074 @comment procfs.c was re-written. Keep their descriptions around
20075 @comment against the day when someone finds the time to put them back in.
20076 @kindex info proc times
20077 @item info proc times
20078 Starting time, user CPU time, and system CPU time for your program and
20081 @kindex info proc id
20083 Report on the process IDs related to your program: its own process ID,
20084 the ID of its parent, the process group ID, and the session ID.
20087 @item set procfs-trace
20088 @kindex set procfs-trace
20089 @cindex @code{procfs} API calls
20090 This command enables and disables tracing of @code{procfs} API calls.
20092 @item show procfs-trace
20093 @kindex show procfs-trace
20094 Show the current state of @code{procfs} API call tracing.
20096 @item set procfs-file @var{file}
20097 @kindex set procfs-file
20098 Tell @value{GDBN} to write @code{procfs} API trace to the named
20099 @var{file}. @value{GDBN} appends the trace info to the previous
20100 contents of the file. The default is to display the trace on the
20103 @item show procfs-file
20104 @kindex show procfs-file
20105 Show the file to which @code{procfs} API trace is written.
20107 @item proc-trace-entry
20108 @itemx proc-trace-exit
20109 @itemx proc-untrace-entry
20110 @itemx proc-untrace-exit
20111 @kindex proc-trace-entry
20112 @kindex proc-trace-exit
20113 @kindex proc-untrace-entry
20114 @kindex proc-untrace-exit
20115 These commands enable and disable tracing of entries into and exits
20116 from the @code{syscall} interface.
20119 @kindex info pidlist
20120 @cindex process list, QNX Neutrino
20121 For QNX Neutrino only, this command displays the list of all the
20122 processes and all the threads within each process.
20125 @kindex info meminfo
20126 @cindex mapinfo list, QNX Neutrino
20127 For QNX Neutrino only, this command displays the list of all mapinfos.
20131 @subsection Features for Debugging @sc{djgpp} Programs
20132 @cindex @sc{djgpp} debugging
20133 @cindex native @sc{djgpp} debugging
20134 @cindex MS-DOS-specific commands
20137 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
20138 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
20139 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
20140 top of real-mode DOS systems and their emulations.
20142 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
20143 defines a few commands specific to the @sc{djgpp} port. This
20144 subsection describes those commands.
20149 This is a prefix of @sc{djgpp}-specific commands which print
20150 information about the target system and important OS structures.
20153 @cindex MS-DOS system info
20154 @cindex free memory information (MS-DOS)
20155 @item info dos sysinfo
20156 This command displays assorted information about the underlying
20157 platform: the CPU type and features, the OS version and flavor, the
20158 DPMI version, and the available conventional and DPMI memory.
20163 @cindex segment descriptor tables
20164 @cindex descriptor tables display
20166 @itemx info dos ldt
20167 @itemx info dos idt
20168 These 3 commands display entries from, respectively, Global, Local,
20169 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
20170 tables are data structures which store a descriptor for each segment
20171 that is currently in use. The segment's selector is an index into a
20172 descriptor table; the table entry for that index holds the
20173 descriptor's base address and limit, and its attributes and access
20176 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
20177 segment (used for both data and the stack), and a DOS segment (which
20178 allows access to DOS/BIOS data structures and absolute addresses in
20179 conventional memory). However, the DPMI host will usually define
20180 additional segments in order to support the DPMI environment.
20182 @cindex garbled pointers
20183 These commands allow to display entries from the descriptor tables.
20184 Without an argument, all entries from the specified table are
20185 displayed. An argument, which should be an integer expression, means
20186 display a single entry whose index is given by the argument. For
20187 example, here's a convenient way to display information about the
20188 debugged program's data segment:
20191 @exdent @code{(@value{GDBP}) info dos ldt $ds}
20192 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
20196 This comes in handy when you want to see whether a pointer is outside
20197 the data segment's limit (i.e.@: @dfn{garbled}).
20199 @cindex page tables display (MS-DOS)
20201 @itemx info dos pte
20202 These two commands display entries from, respectively, the Page
20203 Directory and the Page Tables. Page Directories and Page Tables are
20204 data structures which control how virtual memory addresses are mapped
20205 into physical addresses. A Page Table includes an entry for every
20206 page of memory that is mapped into the program's address space; there
20207 may be several Page Tables, each one holding up to 4096 entries. A
20208 Page Directory has up to 4096 entries, one each for every Page Table
20209 that is currently in use.
20211 Without an argument, @kbd{info dos pde} displays the entire Page
20212 Directory, and @kbd{info dos pte} displays all the entries in all of
20213 the Page Tables. An argument, an integer expression, given to the
20214 @kbd{info dos pde} command means display only that entry from the Page
20215 Directory table. An argument given to the @kbd{info dos pte} command
20216 means display entries from a single Page Table, the one pointed to by
20217 the specified entry in the Page Directory.
20219 @cindex direct memory access (DMA) on MS-DOS
20220 These commands are useful when your program uses @dfn{DMA} (Direct
20221 Memory Access), which needs physical addresses to program the DMA
20224 These commands are supported only with some DPMI servers.
20226 @cindex physical address from linear address
20227 @item info dos address-pte @var{addr}
20228 This command displays the Page Table entry for a specified linear
20229 address. The argument @var{addr} is a linear address which should
20230 already have the appropriate segment's base address added to it,
20231 because this command accepts addresses which may belong to @emph{any}
20232 segment. For example, here's how to display the Page Table entry for
20233 the page where a variable @code{i} is stored:
20236 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
20237 @exdent @code{Page Table entry for address 0x11a00d30:}
20238 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
20242 This says that @code{i} is stored at offset @code{0xd30} from the page
20243 whose physical base address is @code{0x02698000}, and shows all the
20244 attributes of that page.
20246 Note that you must cast the addresses of variables to a @code{char *},
20247 since otherwise the value of @code{__djgpp_base_address}, the base
20248 address of all variables and functions in a @sc{djgpp} program, will
20249 be added using the rules of C pointer arithmetics: if @code{i} is
20250 declared an @code{int}, @value{GDBN} will add 4 times the value of
20251 @code{__djgpp_base_address} to the address of @code{i}.
20253 Here's another example, it displays the Page Table entry for the
20257 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
20258 @exdent @code{Page Table entry for address 0x29110:}
20259 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
20263 (The @code{+ 3} offset is because the transfer buffer's address is the
20264 3rd member of the @code{_go32_info_block} structure.) The output
20265 clearly shows that this DPMI server maps the addresses in conventional
20266 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
20267 linear (@code{0x29110}) addresses are identical.
20269 This command is supported only with some DPMI servers.
20272 @cindex DOS serial data link, remote debugging
20273 In addition to native debugging, the DJGPP port supports remote
20274 debugging via a serial data link. The following commands are specific
20275 to remote serial debugging in the DJGPP port of @value{GDBN}.
20278 @kindex set com1base
20279 @kindex set com1irq
20280 @kindex set com2base
20281 @kindex set com2irq
20282 @kindex set com3base
20283 @kindex set com3irq
20284 @kindex set com4base
20285 @kindex set com4irq
20286 @item set com1base @var{addr}
20287 This command sets the base I/O port address of the @file{COM1} serial
20290 @item set com1irq @var{irq}
20291 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
20292 for the @file{COM1} serial port.
20294 There are similar commands @samp{set com2base}, @samp{set com3irq},
20295 etc.@: for setting the port address and the @code{IRQ} lines for the
20298 @kindex show com1base
20299 @kindex show com1irq
20300 @kindex show com2base
20301 @kindex show com2irq
20302 @kindex show com3base
20303 @kindex show com3irq
20304 @kindex show com4base
20305 @kindex show com4irq
20306 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
20307 display the current settings of the base address and the @code{IRQ}
20308 lines used by the COM ports.
20311 @kindex info serial
20312 @cindex DOS serial port status
20313 This command prints the status of the 4 DOS serial ports. For each
20314 port, it prints whether it's active or not, its I/O base address and
20315 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
20316 counts of various errors encountered so far.
20320 @node Cygwin Native
20321 @subsection Features for Debugging MS Windows PE Executables
20322 @cindex MS Windows debugging
20323 @cindex native Cygwin debugging
20324 @cindex Cygwin-specific commands
20326 @value{GDBN} supports native debugging of MS Windows programs, including
20327 DLLs with and without symbolic debugging information.
20329 @cindex Ctrl-BREAK, MS-Windows
20330 @cindex interrupt debuggee on MS-Windows
20331 MS-Windows programs that call @code{SetConsoleMode} to switch off the
20332 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
20333 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
20334 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
20335 sequence, which can be used to interrupt the debuggee even if it
20338 There are various additional Cygwin-specific commands, described in
20339 this section. Working with DLLs that have no debugging symbols is
20340 described in @ref{Non-debug DLL Symbols}.
20345 This is a prefix of MS Windows-specific commands which print
20346 information about the target system and important OS structures.
20348 @item info w32 selector
20349 This command displays information returned by
20350 the Win32 API @code{GetThreadSelectorEntry} function.
20351 It takes an optional argument that is evaluated to
20352 a long value to give the information about this given selector.
20353 Without argument, this command displays information
20354 about the six segment registers.
20356 @item info w32 thread-information-block
20357 This command displays thread specific information stored in the
20358 Thread Information Block (readable on the X86 CPU family using @code{$fs}
20359 selector for 32-bit programs and @code{$gs} for 64-bit programs).
20363 This is a Cygwin-specific alias of @code{info shared}.
20365 @kindex set cygwin-exceptions
20366 @cindex debugging the Cygwin DLL
20367 @cindex Cygwin DLL, debugging
20368 @item set cygwin-exceptions @var{mode}
20369 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
20370 happen inside the Cygwin DLL. If @var{mode} is @code{off},
20371 @value{GDBN} will delay recognition of exceptions, and may ignore some
20372 exceptions which seem to be caused by internal Cygwin DLL
20373 ``bookkeeping''. This option is meant primarily for debugging the
20374 Cygwin DLL itself; the default value is @code{off} to avoid annoying
20375 @value{GDBN} users with false @code{SIGSEGV} signals.
20377 @kindex show cygwin-exceptions
20378 @item show cygwin-exceptions
20379 Displays whether @value{GDBN} will break on exceptions that happen
20380 inside the Cygwin DLL itself.
20382 @kindex set new-console
20383 @item set new-console @var{mode}
20384 If @var{mode} is @code{on} the debuggee will
20385 be started in a new console on next start.
20386 If @var{mode} is @code{off}, the debuggee will
20387 be started in the same console as the debugger.
20389 @kindex show new-console
20390 @item show new-console
20391 Displays whether a new console is used
20392 when the debuggee is started.
20394 @kindex set new-group
20395 @item set new-group @var{mode}
20396 This boolean value controls whether the debuggee should
20397 start a new group or stay in the same group as the debugger.
20398 This affects the way the Windows OS handles
20401 @kindex show new-group
20402 @item show new-group
20403 Displays current value of new-group boolean.
20405 @kindex set debugevents
20406 @item set debugevents
20407 This boolean value adds debug output concerning kernel events related
20408 to the debuggee seen by the debugger. This includes events that
20409 signal thread and process creation and exit, DLL loading and
20410 unloading, console interrupts, and debugging messages produced by the
20411 Windows @code{OutputDebugString} API call.
20413 @kindex set debugexec
20414 @item set debugexec
20415 This boolean value adds debug output concerning execute events
20416 (such as resume thread) seen by the debugger.
20418 @kindex set debugexceptions
20419 @item set debugexceptions
20420 This boolean value adds debug output concerning exceptions in the
20421 debuggee seen by the debugger.
20423 @kindex set debugmemory
20424 @item set debugmemory
20425 This boolean value adds debug output concerning debuggee memory reads
20426 and writes by the debugger.
20430 This boolean values specifies whether the debuggee is called
20431 via a shell or directly (default value is on).
20435 Displays if the debuggee will be started with a shell.
20440 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
20443 @node Non-debug DLL Symbols
20444 @subsubsection Support for DLLs without Debugging Symbols
20445 @cindex DLLs with no debugging symbols
20446 @cindex Minimal symbols and DLLs
20448 Very often on windows, some of the DLLs that your program relies on do
20449 not include symbolic debugging information (for example,
20450 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
20451 symbols in a DLL, it relies on the minimal amount of symbolic
20452 information contained in the DLL's export table. This section
20453 describes working with such symbols, known internally to @value{GDBN} as
20454 ``minimal symbols''.
20456 Note that before the debugged program has started execution, no DLLs
20457 will have been loaded. The easiest way around this problem is simply to
20458 start the program --- either by setting a breakpoint or letting the
20459 program run once to completion.
20461 @subsubsection DLL Name Prefixes
20463 In keeping with the naming conventions used by the Microsoft debugging
20464 tools, DLL export symbols are made available with a prefix based on the
20465 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
20466 also entered into the symbol table, so @code{CreateFileA} is often
20467 sufficient. In some cases there will be name clashes within a program
20468 (particularly if the executable itself includes full debugging symbols)
20469 necessitating the use of the fully qualified name when referring to the
20470 contents of the DLL. Use single-quotes around the name to avoid the
20471 exclamation mark (``!'') being interpreted as a language operator.
20473 Note that the internal name of the DLL may be all upper-case, even
20474 though the file name of the DLL is lower-case, or vice-versa. Since
20475 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
20476 some confusion. If in doubt, try the @code{info functions} and
20477 @code{info variables} commands or even @code{maint print msymbols}
20478 (@pxref{Symbols}). Here's an example:
20481 (@value{GDBP}) info function CreateFileA
20482 All functions matching regular expression "CreateFileA":
20484 Non-debugging symbols:
20485 0x77e885f4 CreateFileA
20486 0x77e885f4 KERNEL32!CreateFileA
20490 (@value{GDBP}) info function !
20491 All functions matching regular expression "!":
20493 Non-debugging symbols:
20494 0x6100114c cygwin1!__assert
20495 0x61004034 cygwin1!_dll_crt0@@0
20496 0x61004240 cygwin1!dll_crt0(per_process *)
20500 @subsubsection Working with Minimal Symbols
20502 Symbols extracted from a DLL's export table do not contain very much
20503 type information. All that @value{GDBN} can do is guess whether a symbol
20504 refers to a function or variable depending on the linker section that
20505 contains the symbol. Also note that the actual contents of the memory
20506 contained in a DLL are not available unless the program is running. This
20507 means that you cannot examine the contents of a variable or disassemble
20508 a function within a DLL without a running program.
20510 Variables are generally treated as pointers and dereferenced
20511 automatically. For this reason, it is often necessary to prefix a
20512 variable name with the address-of operator (``&'') and provide explicit
20513 type information in the command. Here's an example of the type of
20517 (@value{GDBP}) print 'cygwin1!__argv'
20522 (@value{GDBP}) x 'cygwin1!__argv'
20523 0x10021610: "\230y\""
20526 And two possible solutions:
20529 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
20530 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
20534 (@value{GDBP}) x/2x &'cygwin1!__argv'
20535 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
20536 (@value{GDBP}) x/x 0x10021608
20537 0x10021608: 0x0022fd98
20538 (@value{GDBP}) x/s 0x0022fd98
20539 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
20542 Setting a break point within a DLL is possible even before the program
20543 starts execution. However, under these circumstances, @value{GDBN} can't
20544 examine the initial instructions of the function in order to skip the
20545 function's frame set-up code. You can work around this by using ``*&''
20546 to set the breakpoint at a raw memory address:
20549 (@value{GDBP}) break *&'python22!PyOS_Readline'
20550 Breakpoint 1 at 0x1e04eff0
20553 The author of these extensions is not entirely convinced that setting a
20554 break point within a shared DLL like @file{kernel32.dll} is completely
20558 @subsection Commands Specific to @sc{gnu} Hurd Systems
20559 @cindex @sc{gnu} Hurd debugging
20561 This subsection describes @value{GDBN} commands specific to the
20562 @sc{gnu} Hurd native debugging.
20567 @kindex set signals@r{, Hurd command}
20568 @kindex set sigs@r{, Hurd command}
20569 This command toggles the state of inferior signal interception by
20570 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
20571 affected by this command. @code{sigs} is a shorthand alias for
20576 @kindex show signals@r{, Hurd command}
20577 @kindex show sigs@r{, Hurd command}
20578 Show the current state of intercepting inferior's signals.
20580 @item set signal-thread
20581 @itemx set sigthread
20582 @kindex set signal-thread
20583 @kindex set sigthread
20584 This command tells @value{GDBN} which thread is the @code{libc} signal
20585 thread. That thread is run when a signal is delivered to a running
20586 process. @code{set sigthread} is the shorthand alias of @code{set
20589 @item show signal-thread
20590 @itemx show sigthread
20591 @kindex show signal-thread
20592 @kindex show sigthread
20593 These two commands show which thread will run when the inferior is
20594 delivered a signal.
20597 @kindex set stopped@r{, Hurd command}
20598 This commands tells @value{GDBN} that the inferior process is stopped,
20599 as with the @code{SIGSTOP} signal. The stopped process can be
20600 continued by delivering a signal to it.
20603 @kindex show stopped@r{, Hurd command}
20604 This command shows whether @value{GDBN} thinks the debuggee is
20607 @item set exceptions
20608 @kindex set exceptions@r{, Hurd command}
20609 Use this command to turn off trapping of exceptions in the inferior.
20610 When exception trapping is off, neither breakpoints nor
20611 single-stepping will work. To restore the default, set exception
20614 @item show exceptions
20615 @kindex show exceptions@r{, Hurd command}
20616 Show the current state of trapping exceptions in the inferior.
20618 @item set task pause
20619 @kindex set task@r{, Hurd commands}
20620 @cindex task attributes (@sc{gnu} Hurd)
20621 @cindex pause current task (@sc{gnu} Hurd)
20622 This command toggles task suspension when @value{GDBN} has control.
20623 Setting it to on takes effect immediately, and the task is suspended
20624 whenever @value{GDBN} gets control. Setting it to off will take
20625 effect the next time the inferior is continued. If this option is set
20626 to off, you can use @code{set thread default pause on} or @code{set
20627 thread pause on} (see below) to pause individual threads.
20629 @item show task pause
20630 @kindex show task@r{, Hurd commands}
20631 Show the current state of task suspension.
20633 @item set task detach-suspend-count
20634 @cindex task suspend count
20635 @cindex detach from task, @sc{gnu} Hurd
20636 This command sets the suspend count the task will be left with when
20637 @value{GDBN} detaches from it.
20639 @item show task detach-suspend-count
20640 Show the suspend count the task will be left with when detaching.
20642 @item set task exception-port
20643 @itemx set task excp
20644 @cindex task exception port, @sc{gnu} Hurd
20645 This command sets the task exception port to which @value{GDBN} will
20646 forward exceptions. The argument should be the value of the @dfn{send
20647 rights} of the task. @code{set task excp} is a shorthand alias.
20649 @item set noninvasive
20650 @cindex noninvasive task options
20651 This command switches @value{GDBN} to a mode that is the least
20652 invasive as far as interfering with the inferior is concerned. This
20653 is the same as using @code{set task pause}, @code{set exceptions}, and
20654 @code{set signals} to values opposite to the defaults.
20656 @item info send-rights
20657 @itemx info receive-rights
20658 @itemx info port-rights
20659 @itemx info port-sets
20660 @itemx info dead-names
20663 @cindex send rights, @sc{gnu} Hurd
20664 @cindex receive rights, @sc{gnu} Hurd
20665 @cindex port rights, @sc{gnu} Hurd
20666 @cindex port sets, @sc{gnu} Hurd
20667 @cindex dead names, @sc{gnu} Hurd
20668 These commands display information about, respectively, send rights,
20669 receive rights, port rights, port sets, and dead names of a task.
20670 There are also shorthand aliases: @code{info ports} for @code{info
20671 port-rights} and @code{info psets} for @code{info port-sets}.
20673 @item set thread pause
20674 @kindex set thread@r{, Hurd command}
20675 @cindex thread properties, @sc{gnu} Hurd
20676 @cindex pause current thread (@sc{gnu} Hurd)
20677 This command toggles current thread suspension when @value{GDBN} has
20678 control. Setting it to on takes effect immediately, and the current
20679 thread is suspended whenever @value{GDBN} gets control. Setting it to
20680 off will take effect the next time the inferior is continued.
20681 Normally, this command has no effect, since when @value{GDBN} has
20682 control, the whole task is suspended. However, if you used @code{set
20683 task pause off} (see above), this command comes in handy to suspend
20684 only the current thread.
20686 @item show thread pause
20687 @kindex show thread@r{, Hurd command}
20688 This command shows the state of current thread suspension.
20690 @item set thread run
20691 This command sets whether the current thread is allowed to run.
20693 @item show thread run
20694 Show whether the current thread is allowed to run.
20696 @item set thread detach-suspend-count
20697 @cindex thread suspend count, @sc{gnu} Hurd
20698 @cindex detach from thread, @sc{gnu} Hurd
20699 This command sets the suspend count @value{GDBN} will leave on a
20700 thread when detaching. This number is relative to the suspend count
20701 found by @value{GDBN} when it notices the thread; use @code{set thread
20702 takeover-suspend-count} to force it to an absolute value.
20704 @item show thread detach-suspend-count
20705 Show the suspend count @value{GDBN} will leave on the thread when
20708 @item set thread exception-port
20709 @itemx set thread excp
20710 Set the thread exception port to which to forward exceptions. This
20711 overrides the port set by @code{set task exception-port} (see above).
20712 @code{set thread excp} is the shorthand alias.
20714 @item set thread takeover-suspend-count
20715 Normally, @value{GDBN}'s thread suspend counts are relative to the
20716 value @value{GDBN} finds when it notices each thread. This command
20717 changes the suspend counts to be absolute instead.
20719 @item set thread default
20720 @itemx show thread default
20721 @cindex thread default settings, @sc{gnu} Hurd
20722 Each of the above @code{set thread} commands has a @code{set thread
20723 default} counterpart (e.g., @code{set thread default pause}, @code{set
20724 thread default exception-port}, etc.). The @code{thread default}
20725 variety of commands sets the default thread properties for all
20726 threads; you can then change the properties of individual threads with
20727 the non-default commands.
20734 @value{GDBN} provides the following commands specific to the Darwin target:
20737 @item set debug darwin @var{num}
20738 @kindex set debug darwin
20739 When set to a non zero value, enables debugging messages specific to
20740 the Darwin support. Higher values produce more verbose output.
20742 @item show debug darwin
20743 @kindex show debug darwin
20744 Show the current state of Darwin messages.
20746 @item set debug mach-o @var{num}
20747 @kindex set debug mach-o
20748 When set to a non zero value, enables debugging messages while
20749 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
20750 file format used on Darwin for object and executable files.) Higher
20751 values produce more verbose output. This is a command to diagnose
20752 problems internal to @value{GDBN} and should not be needed in normal
20755 @item show debug mach-o
20756 @kindex show debug mach-o
20757 Show the current state of Mach-O file messages.
20759 @item set mach-exceptions on
20760 @itemx set mach-exceptions off
20761 @kindex set mach-exceptions
20762 On Darwin, faults are first reported as a Mach exception and are then
20763 mapped to a Posix signal. Use this command to turn on trapping of
20764 Mach exceptions in the inferior. This might be sometimes useful to
20765 better understand the cause of a fault. The default is off.
20767 @item show mach-exceptions
20768 @kindex show mach-exceptions
20769 Show the current state of exceptions trapping.
20774 @section Embedded Operating Systems
20776 This section describes configurations involving the debugging of
20777 embedded operating systems that are available for several different
20780 @value{GDBN} includes the ability to debug programs running on
20781 various real-time operating systems.
20783 @node Embedded Processors
20784 @section Embedded Processors
20786 This section goes into details specific to particular embedded
20789 @cindex send command to simulator
20790 Whenever a specific embedded processor has a simulator, @value{GDBN}
20791 allows to send an arbitrary command to the simulator.
20794 @item sim @var{command}
20795 @kindex sim@r{, a command}
20796 Send an arbitrary @var{command} string to the simulator. Consult the
20797 documentation for the specific simulator in use for information about
20798 acceptable commands.
20804 * M32R/D:: Renesas M32R/D
20805 * M68K:: Motorola M68K
20806 * MicroBlaze:: Xilinx MicroBlaze
20807 * MIPS Embedded:: MIPS Embedded
20808 * PowerPC Embedded:: PowerPC Embedded
20809 * PA:: HP PA Embedded
20810 * Sparclet:: Tsqware Sparclet
20811 * Sparclite:: Fujitsu Sparclite
20812 * Z8000:: Zilog Z8000
20815 * Super-H:: Renesas Super-H
20824 @item target rdi @var{dev}
20825 ARM Angel monitor, via RDI library interface to ADP protocol. You may
20826 use this target to communicate with both boards running the Angel
20827 monitor, or with the EmbeddedICE JTAG debug device.
20830 @item target rdp @var{dev}
20835 @value{GDBN} provides the following ARM-specific commands:
20838 @item set arm disassembler
20840 This commands selects from a list of disassembly styles. The
20841 @code{"std"} style is the standard style.
20843 @item show arm disassembler
20845 Show the current disassembly style.
20847 @item set arm apcs32
20848 @cindex ARM 32-bit mode
20849 This command toggles ARM operation mode between 32-bit and 26-bit.
20851 @item show arm apcs32
20852 Display the current usage of the ARM 32-bit mode.
20854 @item set arm fpu @var{fputype}
20855 This command sets the ARM floating-point unit (FPU) type. The
20856 argument @var{fputype} can be one of these:
20860 Determine the FPU type by querying the OS ABI.
20862 Software FPU, with mixed-endian doubles on little-endian ARM
20865 GCC-compiled FPA co-processor.
20867 Software FPU with pure-endian doubles.
20873 Show the current type of the FPU.
20876 This command forces @value{GDBN} to use the specified ABI.
20879 Show the currently used ABI.
20881 @item set arm fallback-mode (arm|thumb|auto)
20882 @value{GDBN} uses the symbol table, when available, to determine
20883 whether instructions are ARM or Thumb. This command controls
20884 @value{GDBN}'s default behavior when the symbol table is not
20885 available. The default is @samp{auto}, which causes @value{GDBN} to
20886 use the current execution mode (from the @code{T} bit in the @code{CPSR}
20889 @item show arm fallback-mode
20890 Show the current fallback instruction mode.
20892 @item set arm force-mode (arm|thumb|auto)
20893 This command overrides use of the symbol table to determine whether
20894 instructions are ARM or Thumb. The default is @samp{auto}, which
20895 causes @value{GDBN} to use the symbol table and then the setting
20896 of @samp{set arm fallback-mode}.
20898 @item show arm force-mode
20899 Show the current forced instruction mode.
20901 @item set debug arm
20902 Toggle whether to display ARM-specific debugging messages from the ARM
20903 target support subsystem.
20905 @item show debug arm
20906 Show whether ARM-specific debugging messages are enabled.
20909 The following commands are available when an ARM target is debugged
20910 using the RDI interface:
20913 @item rdilogfile @r{[}@var{file}@r{]}
20915 @cindex ADP (Angel Debugger Protocol) logging
20916 Set the filename for the ADP (Angel Debugger Protocol) packet log.
20917 With an argument, sets the log file to the specified @var{file}. With
20918 no argument, show the current log file name. The default log file is
20921 @item rdilogenable @r{[}@var{arg}@r{]}
20922 @kindex rdilogenable
20923 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
20924 enables logging, with an argument 0 or @code{"no"} disables it. With
20925 no arguments displays the current setting. When logging is enabled,
20926 ADP packets exchanged between @value{GDBN} and the RDI target device
20927 are logged to a file.
20929 @item set rdiromatzero
20930 @kindex set rdiromatzero
20931 @cindex ROM at zero address, RDI
20932 Tell @value{GDBN} whether the target has ROM at address 0. If on,
20933 vector catching is disabled, so that zero address can be used. If off
20934 (the default), vector catching is enabled. For this command to take
20935 effect, it needs to be invoked prior to the @code{target rdi} command.
20937 @item show rdiromatzero
20938 @kindex show rdiromatzero
20939 Show the current setting of ROM at zero address.
20941 @item set rdiheartbeat
20942 @kindex set rdiheartbeat
20943 @cindex RDI heartbeat
20944 Enable or disable RDI heartbeat packets. It is not recommended to
20945 turn on this option, since it confuses ARM and EPI JTAG interface, as
20946 well as the Angel monitor.
20948 @item show rdiheartbeat
20949 @kindex show rdiheartbeat
20950 Show the setting of RDI heartbeat packets.
20954 @item target sim @r{[}@var{simargs}@r{]} @dots{}
20955 The @value{GDBN} ARM simulator accepts the following optional arguments.
20958 @item --swi-support=@var{type}
20959 Tell the simulator which SWI interfaces to support. The argument
20960 @var{type} may be a comma separated list of the following values.
20961 The default value is @code{all}.
20974 @subsection Renesas M32R/D and M32R/SDI
20977 @kindex target m32r
20978 @item target m32r @var{dev}
20979 Renesas M32R/D ROM monitor.
20981 @kindex target m32rsdi
20982 @item target m32rsdi @var{dev}
20983 Renesas M32R SDI server, connected via parallel port to the board.
20986 The following @value{GDBN} commands are specific to the M32R monitor:
20989 @item set download-path @var{path}
20990 @kindex set download-path
20991 @cindex find downloadable @sc{srec} files (M32R)
20992 Set the default path for finding downloadable @sc{srec} files.
20994 @item show download-path
20995 @kindex show download-path
20996 Show the default path for downloadable @sc{srec} files.
20998 @item set board-address @var{addr}
20999 @kindex set board-address
21000 @cindex M32-EVA target board address
21001 Set the IP address for the M32R-EVA target board.
21003 @item show board-address
21004 @kindex show board-address
21005 Show the current IP address of the target board.
21007 @item set server-address @var{addr}
21008 @kindex set server-address
21009 @cindex download server address (M32R)
21010 Set the IP address for the download server, which is the @value{GDBN}'s
21013 @item show server-address
21014 @kindex show server-address
21015 Display the IP address of the download server.
21017 @item upload @r{[}@var{file}@r{]}
21018 @kindex upload@r{, M32R}
21019 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
21020 upload capability. If no @var{file} argument is given, the current
21021 executable file is uploaded.
21023 @item tload @r{[}@var{file}@r{]}
21024 @kindex tload@r{, M32R}
21025 Test the @code{upload} command.
21028 The following commands are available for M32R/SDI:
21033 @cindex reset SDI connection, M32R
21034 This command resets the SDI connection.
21038 This command shows the SDI connection status.
21041 @kindex debug_chaos
21042 @cindex M32R/Chaos debugging
21043 Instructs the remote that M32R/Chaos debugging is to be used.
21045 @item use_debug_dma
21046 @kindex use_debug_dma
21047 Instructs the remote to use the DEBUG_DMA method of accessing memory.
21050 @kindex use_mon_code
21051 Instructs the remote to use the MON_CODE method of accessing memory.
21054 @kindex use_ib_break
21055 Instructs the remote to set breakpoints by IB break.
21057 @item use_dbt_break
21058 @kindex use_dbt_break
21059 Instructs the remote to set breakpoints by DBT.
21065 The Motorola m68k configuration includes ColdFire support, and a
21066 target command for the following ROM monitor.
21070 @kindex target dbug
21071 @item target dbug @var{dev}
21072 dBUG ROM monitor for Motorola ColdFire.
21077 @subsection MicroBlaze
21078 @cindex Xilinx MicroBlaze
21079 @cindex XMD, Xilinx Microprocessor Debugger
21081 The MicroBlaze is a soft-core processor supported on various Xilinx
21082 FPGAs, such as Spartan or Virtex series. Boards with these processors
21083 usually have JTAG ports which connect to a host system running the Xilinx
21084 Embedded Development Kit (EDK) or Software Development Kit (SDK).
21085 This host system is used to download the configuration bitstream to
21086 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
21087 communicates with the target board using the JTAG interface and
21088 presents a @code{gdbserver} interface to the board. By default
21089 @code{xmd} uses port @code{1234}. (While it is possible to change
21090 this default port, it requires the use of undocumented @code{xmd}
21091 commands. Contact Xilinx support if you need to do this.)
21093 Use these GDB commands to connect to the MicroBlaze target processor.
21096 @item target remote :1234
21097 Use this command to connect to the target if you are running @value{GDBN}
21098 on the same system as @code{xmd}.
21100 @item target remote @var{xmd-host}:1234
21101 Use this command to connect to the target if it is connected to @code{xmd}
21102 running on a different system named @var{xmd-host}.
21105 Use this command to download a program to the MicroBlaze target.
21107 @item set debug microblaze @var{n}
21108 Enable MicroBlaze-specific debugging messages if non-zero.
21110 @item show debug microblaze @var{n}
21111 Show MicroBlaze-specific debugging level.
21114 @node MIPS Embedded
21115 @subsection @acronym{MIPS} Embedded
21117 @cindex @acronym{MIPS} boards
21118 @value{GDBN} can use the @acronym{MIPS} remote debugging protocol to talk to a
21119 @acronym{MIPS} board attached to a serial line. This is available when
21120 you configure @value{GDBN} with @samp{--target=mips-elf}.
21123 Use these @value{GDBN} commands to specify the connection to your target board:
21126 @item target mips @var{port}
21127 @kindex target mips @var{port}
21128 To run a program on the board, start up @code{@value{GDBP}} with the
21129 name of your program as the argument. To connect to the board, use the
21130 command @samp{target mips @var{port}}, where @var{port} is the name of
21131 the serial port connected to the board. If the program has not already
21132 been downloaded to the board, you may use the @code{load} command to
21133 download it. You can then use all the usual @value{GDBN} commands.
21135 For example, this sequence connects to the target board through a serial
21136 port, and loads and runs a program called @var{prog} through the
21140 host$ @value{GDBP} @var{prog}
21141 @value{GDBN} is free software and @dots{}
21142 (@value{GDBP}) target mips /dev/ttyb
21143 (@value{GDBP}) load @var{prog}
21147 @item target mips @var{hostname}:@var{portnumber}
21148 On some @value{GDBN} host configurations, you can specify a TCP
21149 connection (for instance, to a serial line managed by a terminal
21150 concentrator) instead of a serial port, using the syntax
21151 @samp{@var{hostname}:@var{portnumber}}.
21153 @item target pmon @var{port}
21154 @kindex target pmon @var{port}
21157 @item target ddb @var{port}
21158 @kindex target ddb @var{port}
21159 NEC's DDB variant of PMON for Vr4300.
21161 @item target lsi @var{port}
21162 @kindex target lsi @var{port}
21163 LSI variant of PMON.
21165 @kindex target r3900
21166 @item target r3900 @var{dev}
21167 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
21169 @kindex target array
21170 @item target array @var{dev}
21171 Array Tech LSI33K RAID controller board.
21177 @value{GDBN} also supports these special commands for @acronym{MIPS} targets:
21180 @item set mipsfpu double
21181 @itemx set mipsfpu single
21182 @itemx set mipsfpu none
21183 @itemx set mipsfpu auto
21184 @itemx show mipsfpu
21185 @kindex set mipsfpu
21186 @kindex show mipsfpu
21187 @cindex @acronym{MIPS} remote floating point
21188 @cindex floating point, @acronym{MIPS} remote
21189 If your target board does not support the @acronym{MIPS} floating point
21190 coprocessor, you should use the command @samp{set mipsfpu none} (if you
21191 need this, you may wish to put the command in your @value{GDBN} init
21192 file). This tells @value{GDBN} how to find the return value of
21193 functions which return floating point values. It also allows
21194 @value{GDBN} to avoid saving the floating point registers when calling
21195 functions on the board. If you are using a floating point coprocessor
21196 with only single precision floating point support, as on the @sc{r4650}
21197 processor, use the command @samp{set mipsfpu single}. The default
21198 double precision floating point coprocessor may be selected using
21199 @samp{set mipsfpu double}.
21201 In previous versions the only choices were double precision or no
21202 floating point, so @samp{set mipsfpu on} will select double precision
21203 and @samp{set mipsfpu off} will select no floating point.
21205 As usual, you can inquire about the @code{mipsfpu} variable with
21206 @samp{show mipsfpu}.
21208 @item set timeout @var{seconds}
21209 @itemx set retransmit-timeout @var{seconds}
21210 @itemx show timeout
21211 @itemx show retransmit-timeout
21212 @cindex @code{timeout}, @acronym{MIPS} protocol
21213 @cindex @code{retransmit-timeout}, @acronym{MIPS} protocol
21214 @kindex set timeout
21215 @kindex show timeout
21216 @kindex set retransmit-timeout
21217 @kindex show retransmit-timeout
21218 You can control the timeout used while waiting for a packet, in the @acronym{MIPS}
21219 remote protocol, with the @code{set timeout @var{seconds}} command. The
21220 default is 5 seconds. Similarly, you can control the timeout used while
21221 waiting for an acknowledgment of a packet with the @code{set
21222 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
21223 You can inspect both values with @code{show timeout} and @code{show
21224 retransmit-timeout}. (These commands are @emph{only} available when
21225 @value{GDBN} is configured for @samp{--target=mips-elf}.)
21227 The timeout set by @code{set timeout} does not apply when @value{GDBN}
21228 is waiting for your program to stop. In that case, @value{GDBN} waits
21229 forever because it has no way of knowing how long the program is going
21230 to run before stopping.
21232 @item set syn-garbage-limit @var{num}
21233 @kindex set syn-garbage-limit@r{, @acronym{MIPS} remote}
21234 @cindex synchronize with remote @acronym{MIPS} target
21235 Limit the maximum number of characters @value{GDBN} should ignore when
21236 it tries to synchronize with the remote target. The default is 10
21237 characters. Setting the limit to -1 means there's no limit.
21239 @item show syn-garbage-limit
21240 @kindex show syn-garbage-limit@r{, @acronym{MIPS} remote}
21241 Show the current limit on the number of characters to ignore when
21242 trying to synchronize with the remote system.
21244 @item set monitor-prompt @var{prompt}
21245 @kindex set monitor-prompt@r{, @acronym{MIPS} remote}
21246 @cindex remote monitor prompt
21247 Tell @value{GDBN} to expect the specified @var{prompt} string from the
21248 remote monitor. The default depends on the target:
21258 @item show monitor-prompt
21259 @kindex show monitor-prompt@r{, @acronym{MIPS} remote}
21260 Show the current strings @value{GDBN} expects as the prompt from the
21263 @item set monitor-warnings
21264 @kindex set monitor-warnings@r{, @acronym{MIPS} remote}
21265 Enable or disable monitor warnings about hardware breakpoints. This
21266 has effect only for the @code{lsi} target. When on, @value{GDBN} will
21267 display warning messages whose codes are returned by the @code{lsi}
21268 PMON monitor for breakpoint commands.
21270 @item show monitor-warnings
21271 @kindex show monitor-warnings@r{, @acronym{MIPS} remote}
21272 Show the current setting of printing monitor warnings.
21274 @item pmon @var{command}
21275 @kindex pmon@r{, @acronym{MIPS} remote}
21276 @cindex send PMON command
21277 This command allows sending an arbitrary @var{command} string to the
21278 monitor. The monitor must be in debug mode for this to work.
21281 @node PowerPC Embedded
21282 @subsection PowerPC Embedded
21284 @cindex DVC register
21285 @value{GDBN} supports using the DVC (Data Value Compare) register to
21286 implement in hardware simple hardware watchpoint conditions of the form:
21289 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
21290 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
21293 The DVC register will be automatically used when @value{GDBN} detects
21294 such pattern in a condition expression, and the created watchpoint uses one
21295 debug register (either the @code{exact-watchpoints} option is on and the
21296 variable is scalar, or the variable has a length of one byte). This feature
21297 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
21300 When running on PowerPC embedded processors, @value{GDBN} automatically uses
21301 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
21302 in which case watchpoints using only one debug register are created when
21303 watching variables of scalar types.
21305 You can create an artificial array to watch an arbitrary memory
21306 region using one of the following commands (@pxref{Expressions}):
21309 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
21310 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
21313 PowerPC embedded processors support masked watchpoints. See the discussion
21314 about the @code{mask} argument in @ref{Set Watchpoints}.
21316 @cindex ranged breakpoint
21317 PowerPC embedded processors support hardware accelerated
21318 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
21319 the inferior whenever it executes an instruction at any address within
21320 the range it specifies. To set a ranged breakpoint in @value{GDBN},
21321 use the @code{break-range} command.
21323 @value{GDBN} provides the following PowerPC-specific commands:
21326 @kindex break-range
21327 @item break-range @var{start-location}, @var{end-location}
21328 Set a breakpoint for an address range given by
21329 @var{start-location} and @var{end-location}, which can specify a function name,
21330 a line number, an offset of lines from the current line or from the start
21331 location, or an address of an instruction (see @ref{Specify Location},
21332 for a list of all the possible ways to specify a @var{location}.)
21333 The breakpoint will stop execution of the inferior whenever it
21334 executes an instruction at any address within the specified range,
21335 (including @var{start-location} and @var{end-location}.)
21337 @kindex set powerpc
21338 @item set powerpc soft-float
21339 @itemx show powerpc soft-float
21340 Force @value{GDBN} to use (or not use) a software floating point calling
21341 convention. By default, @value{GDBN} selects the calling convention based
21342 on the selected architecture and the provided executable file.
21344 @item set powerpc vector-abi
21345 @itemx show powerpc vector-abi
21346 Force @value{GDBN} to use the specified calling convention for vector
21347 arguments and return values. The valid options are @samp{auto};
21348 @samp{generic}, to avoid vector registers even if they are present;
21349 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
21350 registers. By default, @value{GDBN} selects the calling convention
21351 based on the selected architecture and the provided executable file.
21353 @item set powerpc exact-watchpoints
21354 @itemx show powerpc exact-watchpoints
21355 Allow @value{GDBN} to use only one debug register when watching a variable
21356 of scalar type, thus assuming that the variable is accessed through the
21357 address of its first byte.
21359 @kindex target dink32
21360 @item target dink32 @var{dev}
21361 DINK32 ROM monitor.
21363 @kindex target ppcbug
21364 @item target ppcbug @var{dev}
21365 @kindex target ppcbug1
21366 @item target ppcbug1 @var{dev}
21367 PPCBUG ROM monitor for PowerPC.
21370 @item target sds @var{dev}
21371 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
21374 @cindex SDS protocol
21375 The following commands specific to the SDS protocol are supported
21379 @item set sdstimeout @var{nsec}
21380 @kindex set sdstimeout
21381 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
21382 default is 2 seconds.
21384 @item show sdstimeout
21385 @kindex show sdstimeout
21386 Show the current value of the SDS timeout.
21388 @item sds @var{command}
21389 @kindex sds@r{, a command}
21390 Send the specified @var{command} string to the SDS monitor.
21395 @subsection HP PA Embedded
21399 @kindex target op50n
21400 @item target op50n @var{dev}
21401 OP50N monitor, running on an OKI HPPA board.
21403 @kindex target w89k
21404 @item target w89k @var{dev}
21405 W89K monitor, running on a Winbond HPPA board.
21410 @subsection Tsqware Sparclet
21414 @value{GDBN} enables developers to debug tasks running on
21415 Sparclet targets from a Unix host.
21416 @value{GDBN} uses code that runs on
21417 both the Unix host and on the Sparclet target. The program
21418 @code{@value{GDBP}} is installed and executed on the Unix host.
21421 @item remotetimeout @var{args}
21422 @kindex remotetimeout
21423 @value{GDBN} supports the option @code{remotetimeout}.
21424 This option is set by the user, and @var{args} represents the number of
21425 seconds @value{GDBN} waits for responses.
21428 @cindex compiling, on Sparclet
21429 When compiling for debugging, include the options @samp{-g} to get debug
21430 information and @samp{-Ttext} to relocate the program to where you wish to
21431 load it on the target. You may also want to add the options @samp{-n} or
21432 @samp{-N} in order to reduce the size of the sections. Example:
21435 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
21438 You can use @code{objdump} to verify that the addresses are what you intended:
21441 sparclet-aout-objdump --headers --syms prog
21444 @cindex running, on Sparclet
21446 your Unix execution search path to find @value{GDBN}, you are ready to
21447 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
21448 (or @code{sparclet-aout-gdb}, depending on your installation).
21450 @value{GDBN} comes up showing the prompt:
21457 * Sparclet File:: Setting the file to debug
21458 * Sparclet Connection:: Connecting to Sparclet
21459 * Sparclet Download:: Sparclet download
21460 * Sparclet Execution:: Running and debugging
21463 @node Sparclet File
21464 @subsubsection Setting File to Debug
21466 The @value{GDBN} command @code{file} lets you choose with program to debug.
21469 (gdbslet) file prog
21473 @value{GDBN} then attempts to read the symbol table of @file{prog}.
21474 @value{GDBN} locates
21475 the file by searching the directories listed in the command search
21477 If the file was compiled with debug information (option @samp{-g}), source
21478 files will be searched as well.
21479 @value{GDBN} locates
21480 the source files by searching the directories listed in the directory search
21481 path (@pxref{Environment, ,Your Program's Environment}).
21483 to find a file, it displays a message such as:
21486 prog: No such file or directory.
21489 When this happens, add the appropriate directories to the search paths with
21490 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
21491 @code{target} command again.
21493 @node Sparclet Connection
21494 @subsubsection Connecting to Sparclet
21496 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
21497 To connect to a target on serial port ``@code{ttya}'', type:
21500 (gdbslet) target sparclet /dev/ttya
21501 Remote target sparclet connected to /dev/ttya
21502 main () at ../prog.c:3
21506 @value{GDBN} displays messages like these:
21512 @node Sparclet Download
21513 @subsubsection Sparclet Download
21515 @cindex download to Sparclet
21516 Once connected to the Sparclet target,
21517 you can use the @value{GDBN}
21518 @code{load} command to download the file from the host to the target.
21519 The file name and load offset should be given as arguments to the @code{load}
21521 Since the file format is aout, the program must be loaded to the starting
21522 address. You can use @code{objdump} to find out what this value is. The load
21523 offset is an offset which is added to the VMA (virtual memory address)
21524 of each of the file's sections.
21525 For instance, if the program
21526 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
21527 and bss at 0x12010170, in @value{GDBN}, type:
21530 (gdbslet) load prog 0x12010000
21531 Loading section .text, size 0xdb0 vma 0x12010000
21534 If the code is loaded at a different address then what the program was linked
21535 to, you may need to use the @code{section} and @code{add-symbol-file} commands
21536 to tell @value{GDBN} where to map the symbol table.
21538 @node Sparclet Execution
21539 @subsubsection Running and Debugging
21541 @cindex running and debugging Sparclet programs
21542 You can now begin debugging the task using @value{GDBN}'s execution control
21543 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
21544 manual for the list of commands.
21548 Breakpoint 1 at 0x12010000: file prog.c, line 3.
21550 Starting program: prog
21551 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
21552 3 char *symarg = 0;
21554 4 char *execarg = "hello!";
21559 @subsection Fujitsu Sparclite
21563 @kindex target sparclite
21564 @item target sparclite @var{dev}
21565 Fujitsu sparclite boards, used only for the purpose of loading.
21566 You must use an additional command to debug the program.
21567 For example: target remote @var{dev} using @value{GDBN} standard
21573 @subsection Zilog Z8000
21576 @cindex simulator, Z8000
21577 @cindex Zilog Z8000 simulator
21579 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
21582 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
21583 unsegmented variant of the Z8000 architecture) or the Z8001 (the
21584 segmented variant). The simulator recognizes which architecture is
21585 appropriate by inspecting the object code.
21588 @item target sim @var{args}
21590 @kindex target sim@r{, with Z8000}
21591 Debug programs on a simulated CPU. If the simulator supports setup
21592 options, specify them via @var{args}.
21596 After specifying this target, you can debug programs for the simulated
21597 CPU in the same style as programs for your host computer; use the
21598 @code{file} command to load a new program image, the @code{run} command
21599 to run your program, and so on.
21601 As well as making available all the usual machine registers
21602 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
21603 additional items of information as specially named registers:
21608 Counts clock-ticks in the simulator.
21611 Counts instructions run in the simulator.
21614 Execution time in 60ths of a second.
21618 You can refer to these values in @value{GDBN} expressions with the usual
21619 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
21620 conditional breakpoint that suspends only after at least 5000
21621 simulated clock ticks.
21624 @subsection Atmel AVR
21627 When configured for debugging the Atmel AVR, @value{GDBN} supports the
21628 following AVR-specific commands:
21631 @item info io_registers
21632 @kindex info io_registers@r{, AVR}
21633 @cindex I/O registers (Atmel AVR)
21634 This command displays information about the AVR I/O registers. For
21635 each register, @value{GDBN} prints its number and value.
21642 When configured for debugging CRIS, @value{GDBN} provides the
21643 following CRIS-specific commands:
21646 @item set cris-version @var{ver}
21647 @cindex CRIS version
21648 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
21649 The CRIS version affects register names and sizes. This command is useful in
21650 case autodetection of the CRIS version fails.
21652 @item show cris-version
21653 Show the current CRIS version.
21655 @item set cris-dwarf2-cfi
21656 @cindex DWARF-2 CFI and CRIS
21657 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
21658 Change to @samp{off} when using @code{gcc-cris} whose version is below
21661 @item show cris-dwarf2-cfi
21662 Show the current state of using DWARF-2 CFI.
21664 @item set cris-mode @var{mode}
21666 Set the current CRIS mode to @var{mode}. It should only be changed when
21667 debugging in guru mode, in which case it should be set to
21668 @samp{guru} (the default is @samp{normal}).
21670 @item show cris-mode
21671 Show the current CRIS mode.
21675 @subsection Renesas Super-H
21678 For the Renesas Super-H processor, @value{GDBN} provides these
21682 @item set sh calling-convention @var{convention}
21683 @kindex set sh calling-convention
21684 Set the calling-convention used when calling functions from @value{GDBN}.
21685 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
21686 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
21687 convention. If the DWARF-2 information of the called function specifies
21688 that the function follows the Renesas calling convention, the function
21689 is called using the Renesas calling convention. If the calling convention
21690 is set to @samp{renesas}, the Renesas calling convention is always used,
21691 regardless of the DWARF-2 information. This can be used to override the
21692 default of @samp{gcc} if debug information is missing, or the compiler
21693 does not emit the DWARF-2 calling convention entry for a function.
21695 @item show sh calling-convention
21696 @kindex show sh calling-convention
21697 Show the current calling convention setting.
21702 @node Architectures
21703 @section Architectures
21705 This section describes characteristics of architectures that affect
21706 all uses of @value{GDBN} with the architecture, both native and cross.
21713 * HPPA:: HP PA architecture
21714 * SPU:: Cell Broadband Engine SPU architecture
21720 @subsection AArch64
21721 @cindex AArch64 support
21723 When @value{GDBN} is debugging the AArch64 architecture, it provides the
21724 following special commands:
21727 @item set debug aarch64
21728 @kindex set debug aarch64
21729 This command determines whether AArch64 architecture-specific debugging
21730 messages are to be displayed.
21732 @item show debug aarch64
21733 Show whether AArch64 debugging messages are displayed.
21738 @subsection x86 Architecture-specific Issues
21741 @item set struct-convention @var{mode}
21742 @kindex set struct-convention
21743 @cindex struct return convention
21744 @cindex struct/union returned in registers
21745 Set the convention used by the inferior to return @code{struct}s and
21746 @code{union}s from functions to @var{mode}. Possible values of
21747 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
21748 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
21749 are returned on the stack, while @code{"reg"} means that a
21750 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
21751 be returned in a register.
21753 @item show struct-convention
21754 @kindex show struct-convention
21755 Show the current setting of the convention to return @code{struct}s
21759 @subsubsection Intel(R) @dfn{Memory Protection Extensions} (MPX).
21760 @cindex Intel(R) Memory Protection Extensions (MPX).
21762 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
21763 @footnote{The register named with capital letters represent the architecture
21764 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
21765 which are the lower bound and upper bound. Bounds are effective addresses or
21766 memory locations. The upper bounds are architecturally represented in 1's
21767 complement form. A bound having lower bound = 0, and upper bound = 0
21768 (1's complement of all bits set) will allow access to the entire address space.
21770 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
21771 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
21772 display the upper bound performing the complement of one operation on the
21773 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
21774 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
21775 can also be noted that the upper bounds are inclusive.
21777 As an example, assume that the register BND0 holds bounds for a pointer having
21778 access allowed for the range between 0x32 and 0x71. The values present on
21779 bnd0raw and bnd registers are presented as follows:
21782 bnd0raw = @{0x32, 0xffffffff8e@}
21783 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
21786 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
21787 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
21788 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
21789 Python, the display includes the memory size, in bits, accessible to
21795 See the following section.
21798 @subsection @acronym{MIPS}
21800 @cindex stack on Alpha
21801 @cindex stack on @acronym{MIPS}
21802 @cindex Alpha stack
21803 @cindex @acronym{MIPS} stack
21804 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
21805 sometimes requires @value{GDBN} to search backward in the object code to
21806 find the beginning of a function.
21808 @cindex response time, @acronym{MIPS} debugging
21809 To improve response time (especially for embedded applications, where
21810 @value{GDBN} may be restricted to a slow serial line for this search)
21811 you may want to limit the size of this search, using one of these
21815 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
21816 @item set heuristic-fence-post @var{limit}
21817 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
21818 search for the beginning of a function. A value of @var{0} (the
21819 default) means there is no limit. However, except for @var{0}, the
21820 larger the limit the more bytes @code{heuristic-fence-post} must search
21821 and therefore the longer it takes to run. You should only need to use
21822 this command when debugging a stripped executable.
21824 @item show heuristic-fence-post
21825 Display the current limit.
21829 These commands are available @emph{only} when @value{GDBN} is configured
21830 for debugging programs on Alpha or @acronym{MIPS} processors.
21832 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
21836 @item set mips abi @var{arg}
21837 @kindex set mips abi
21838 @cindex set ABI for @acronym{MIPS}
21839 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
21840 values of @var{arg} are:
21844 The default ABI associated with the current binary (this is the
21854 @item show mips abi
21855 @kindex show mips abi
21856 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
21858 @item set mips compression @var{arg}
21859 @kindex set mips compression
21860 @cindex code compression, @acronym{MIPS}
21861 Tell @value{GDBN} which @acronym{MIPS} compressed
21862 @acronym{ISA, Instruction Set Architecture} encoding is used by the
21863 inferior. @value{GDBN} uses this for code disassembly and other
21864 internal interpretation purposes. This setting is only referred to
21865 when no executable has been associated with the debugging session or
21866 the executable does not provide information about the encoding it uses.
21867 Otherwise this setting is automatically updated from information
21868 provided by the executable.
21870 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
21871 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
21872 executables containing @acronym{MIPS16} code frequently are not
21873 identified as such.
21875 This setting is ``sticky''; that is, it retains its value across
21876 debugging sessions until reset either explicitly with this command or
21877 implicitly from an executable.
21879 The compiler and/or assembler typically add symbol table annotations to
21880 identify functions compiled for the @acronym{MIPS16} or
21881 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
21882 are present, @value{GDBN} uses them in preference to the global
21883 compressed @acronym{ISA} encoding setting.
21885 @item show mips compression
21886 @kindex show mips compression
21887 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
21888 @value{GDBN} to debug the inferior.
21891 @itemx show mipsfpu
21892 @xref{MIPS Embedded, set mipsfpu}.
21894 @item set mips mask-address @var{arg}
21895 @kindex set mips mask-address
21896 @cindex @acronym{MIPS} addresses, masking
21897 This command determines whether the most-significant 32 bits of 64-bit
21898 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
21899 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
21900 setting, which lets @value{GDBN} determine the correct value.
21902 @item show mips mask-address
21903 @kindex show mips mask-address
21904 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
21907 @item set remote-mips64-transfers-32bit-regs
21908 @kindex set remote-mips64-transfers-32bit-regs
21909 This command controls compatibility with 64-bit @acronym{MIPS} targets that
21910 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
21911 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
21912 and 64 bits for other registers, set this option to @samp{on}.
21914 @item show remote-mips64-transfers-32bit-regs
21915 @kindex show remote-mips64-transfers-32bit-regs
21916 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
21918 @item set debug mips
21919 @kindex set debug mips
21920 This command turns on and off debugging messages for the @acronym{MIPS}-specific
21921 target code in @value{GDBN}.
21923 @item show debug mips
21924 @kindex show debug mips
21925 Show the current setting of @acronym{MIPS} debugging messages.
21931 @cindex HPPA support
21933 When @value{GDBN} is debugging the HP PA architecture, it provides the
21934 following special commands:
21937 @item set debug hppa
21938 @kindex set debug hppa
21939 This command determines whether HPPA architecture-specific debugging
21940 messages are to be displayed.
21942 @item show debug hppa
21943 Show whether HPPA debugging messages are displayed.
21945 @item maint print unwind @var{address}
21946 @kindex maint print unwind@r{, HPPA}
21947 This command displays the contents of the unwind table entry at the
21948 given @var{address}.
21954 @subsection Cell Broadband Engine SPU architecture
21955 @cindex Cell Broadband Engine
21958 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
21959 it provides the following special commands:
21962 @item info spu event
21964 Display SPU event facility status. Shows current event mask
21965 and pending event status.
21967 @item info spu signal
21968 Display SPU signal notification facility status. Shows pending
21969 signal-control word and signal notification mode of both signal
21970 notification channels.
21972 @item info spu mailbox
21973 Display SPU mailbox facility status. Shows all pending entries,
21974 in order of processing, in each of the SPU Write Outbound,
21975 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
21978 Display MFC DMA status. Shows all pending commands in the MFC
21979 DMA queue. For each entry, opcode, tag, class IDs, effective
21980 and local store addresses and transfer size are shown.
21982 @item info spu proxydma
21983 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
21984 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
21985 and local store addresses and transfer size are shown.
21989 When @value{GDBN} is debugging a combined PowerPC/SPU application
21990 on the Cell Broadband Engine, it provides in addition the following
21994 @item set spu stop-on-load @var{arg}
21996 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
21997 will give control to the user when a new SPE thread enters its @code{main}
21998 function. The default is @code{off}.
22000 @item show spu stop-on-load
22002 Show whether to stop for new SPE threads.
22004 @item set spu auto-flush-cache @var{arg}
22005 Set whether to automatically flush the software-managed cache. When set to
22006 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
22007 cache to be flushed whenever SPE execution stops. This provides a consistent
22008 view of PowerPC memory that is accessed via the cache. If an application
22009 does not use the software-managed cache, this option has no effect.
22011 @item show spu auto-flush-cache
22012 Show whether to automatically flush the software-managed cache.
22017 @subsection PowerPC
22018 @cindex PowerPC architecture
22020 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
22021 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
22022 numbers stored in the floating point registers. These values must be stored
22023 in two consecutive registers, always starting at an even register like
22024 @code{f0} or @code{f2}.
22026 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
22027 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
22028 @code{f2} and @code{f3} for @code{$dl1} and so on.
22030 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
22031 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
22034 @subsection Nios II
22035 @cindex Nios II architecture
22037 When @value{GDBN} is debugging the Nios II architecture,
22038 it provides the following special commands:
22042 @item set debug nios2
22043 @kindex set debug nios2
22044 This command turns on and off debugging messages for the Nios II
22045 target code in @value{GDBN}.
22047 @item show debug nios2
22048 @kindex show debug nios2
22049 Show the current setting of Nios II debugging messages.
22052 @node Controlling GDB
22053 @chapter Controlling @value{GDBN}
22055 You can alter the way @value{GDBN} interacts with you by using the
22056 @code{set} command. For commands controlling how @value{GDBN} displays
22057 data, see @ref{Print Settings, ,Print Settings}. Other settings are
22062 * Editing:: Command editing
22063 * Command History:: Command history
22064 * Screen Size:: Screen size
22065 * Numbers:: Numbers
22066 * ABI:: Configuring the current ABI
22067 * Auto-loading:: Automatically loading associated files
22068 * Messages/Warnings:: Optional warnings and messages
22069 * Debugging Output:: Optional messages about internal happenings
22070 * Other Misc Settings:: Other Miscellaneous Settings
22078 @value{GDBN} indicates its readiness to read a command by printing a string
22079 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
22080 can change the prompt string with the @code{set prompt} command. For
22081 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
22082 the prompt in one of the @value{GDBN} sessions so that you can always tell
22083 which one you are talking to.
22085 @emph{Note:} @code{set prompt} does not add a space for you after the
22086 prompt you set. This allows you to set a prompt which ends in a space
22087 or a prompt that does not.
22091 @item set prompt @var{newprompt}
22092 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
22094 @kindex show prompt
22096 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
22099 Versions of @value{GDBN} that ship with Python scripting enabled have
22100 prompt extensions. The commands for interacting with these extensions
22104 @kindex set extended-prompt
22105 @item set extended-prompt @var{prompt}
22106 Set an extended prompt that allows for substitutions.
22107 @xref{gdb.prompt}, for a list of escape sequences that can be used for
22108 substitution. Any escape sequences specified as part of the prompt
22109 string are replaced with the corresponding strings each time the prompt
22115 set extended-prompt Current working directory: \w (gdb)
22118 Note that when an extended-prompt is set, it takes control of the
22119 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
22121 @kindex show extended-prompt
22122 @item show extended-prompt
22123 Prints the extended prompt. Any escape sequences specified as part of
22124 the prompt string with @code{set extended-prompt}, are replaced with the
22125 corresponding strings each time the prompt is displayed.
22129 @section Command Editing
22131 @cindex command line editing
22133 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
22134 @sc{gnu} library provides consistent behavior for programs which provide a
22135 command line interface to the user. Advantages are @sc{gnu} Emacs-style
22136 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
22137 substitution, and a storage and recall of command history across
22138 debugging sessions.
22140 You may control the behavior of command line editing in @value{GDBN} with the
22141 command @code{set}.
22144 @kindex set editing
22147 @itemx set editing on
22148 Enable command line editing (enabled by default).
22150 @item set editing off
22151 Disable command line editing.
22153 @kindex show editing
22155 Show whether command line editing is enabled.
22158 @ifset SYSTEM_READLINE
22159 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
22161 @ifclear SYSTEM_READLINE
22162 @xref{Command Line Editing},
22164 for more details about the Readline
22165 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
22166 encouraged to read that chapter.
22168 @node Command History
22169 @section Command History
22170 @cindex command history
22172 @value{GDBN} can keep track of the commands you type during your
22173 debugging sessions, so that you can be certain of precisely what
22174 happened. Use these commands to manage the @value{GDBN} command
22177 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
22178 package, to provide the history facility.
22179 @ifset SYSTEM_READLINE
22180 @xref{Using History Interactively, , , history, GNU History Library},
22182 @ifclear SYSTEM_READLINE
22183 @xref{Using History Interactively},
22185 for the detailed description of the History library.
22187 To issue a command to @value{GDBN} without affecting certain aspects of
22188 the state which is seen by users, prefix it with @samp{server }
22189 (@pxref{Server Prefix}). This
22190 means that this command will not affect the command history, nor will it
22191 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
22192 pressed on a line by itself.
22194 @cindex @code{server}, command prefix
22195 The server prefix does not affect the recording of values into the value
22196 history; to print a value without recording it into the value history,
22197 use the @code{output} command instead of the @code{print} command.
22199 Here is the description of @value{GDBN} commands related to command
22203 @cindex history substitution
22204 @cindex history file
22205 @kindex set history filename
22206 @cindex @env{GDBHISTFILE}, environment variable
22207 @item set history filename @var{fname}
22208 Set the name of the @value{GDBN} command history file to @var{fname}.
22209 This is the file where @value{GDBN} reads an initial command history
22210 list, and where it writes the command history from this session when it
22211 exits. You can access this list through history expansion or through
22212 the history command editing characters listed below. This file defaults
22213 to the value of the environment variable @code{GDBHISTFILE}, or to
22214 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
22217 @cindex save command history
22218 @kindex set history save
22219 @item set history save
22220 @itemx set history save on
22221 Record command history in a file, whose name may be specified with the
22222 @code{set history filename} command. By default, this option is disabled.
22224 @item set history save off
22225 Stop recording command history in a file.
22227 @cindex history size
22228 @kindex set history size
22229 @cindex @env{HISTSIZE}, environment variable
22230 @item set history size @var{size}
22231 @itemx set history size unlimited
22232 Set the number of commands which @value{GDBN} keeps in its history list.
22233 This defaults to the value of the environment variable
22234 @code{HISTSIZE}, or to 256 if this variable is not set. If @var{size}
22235 is @code{unlimited}, the number of commands @value{GDBN} keeps in the
22236 history list is unlimited.
22239 History expansion assigns special meaning to the character @kbd{!}.
22240 @ifset SYSTEM_READLINE
22241 @xref{Event Designators, , , history, GNU History Library},
22243 @ifclear SYSTEM_READLINE
22244 @xref{Event Designators},
22248 @cindex history expansion, turn on/off
22249 Since @kbd{!} is also the logical not operator in C, history expansion
22250 is off by default. If you decide to enable history expansion with the
22251 @code{set history expansion on} command, you may sometimes need to
22252 follow @kbd{!} (when it is used as logical not, in an expression) with
22253 a space or a tab to prevent it from being expanded. The readline
22254 history facilities do not attempt substitution on the strings
22255 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
22257 The commands to control history expansion are:
22260 @item set history expansion on
22261 @itemx set history expansion
22262 @kindex set history expansion
22263 Enable history expansion. History expansion is off by default.
22265 @item set history expansion off
22266 Disable history expansion.
22269 @kindex show history
22271 @itemx show history filename
22272 @itemx show history save
22273 @itemx show history size
22274 @itemx show history expansion
22275 These commands display the state of the @value{GDBN} history parameters.
22276 @code{show history} by itself displays all four states.
22281 @kindex show commands
22282 @cindex show last commands
22283 @cindex display command history
22284 @item show commands
22285 Display the last ten commands in the command history.
22287 @item show commands @var{n}
22288 Print ten commands centered on command number @var{n}.
22290 @item show commands +
22291 Print ten commands just after the commands last printed.
22295 @section Screen Size
22296 @cindex size of screen
22297 @cindex screen size
22300 @cindex pauses in output
22302 Certain commands to @value{GDBN} may produce large amounts of
22303 information output to the screen. To help you read all of it,
22304 @value{GDBN} pauses and asks you for input at the end of each page of
22305 output. Type @key{RET} when you want to continue the output, or @kbd{q}
22306 to discard the remaining output. Also, the screen width setting
22307 determines when to wrap lines of output. Depending on what is being
22308 printed, @value{GDBN} tries to break the line at a readable place,
22309 rather than simply letting it overflow onto the following line.
22311 Normally @value{GDBN} knows the size of the screen from the terminal
22312 driver software. For example, on Unix @value{GDBN} uses the termcap data base
22313 together with the value of the @code{TERM} environment variable and the
22314 @code{stty rows} and @code{stty cols} settings. If this is not correct,
22315 you can override it with the @code{set height} and @code{set
22322 @kindex show height
22323 @item set height @var{lpp}
22324 @itemx set height unlimited
22326 @itemx set width @var{cpl}
22327 @itemx set width unlimited
22329 These @code{set} commands specify a screen height of @var{lpp} lines and
22330 a screen width of @var{cpl} characters. The associated @code{show}
22331 commands display the current settings.
22333 If you specify a height of either @code{unlimited} or zero lines,
22334 @value{GDBN} does not pause during output no matter how long the
22335 output is. This is useful if output is to a file or to an editor
22338 Likewise, you can specify @samp{set width unlimited} or @samp{set
22339 width 0} to prevent @value{GDBN} from wrapping its output.
22341 @item set pagination on
22342 @itemx set pagination off
22343 @kindex set pagination
22344 Turn the output pagination on or off; the default is on. Turning
22345 pagination off is the alternative to @code{set height unlimited}. Note that
22346 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
22347 Options, -batch}) also automatically disables pagination.
22349 @item show pagination
22350 @kindex show pagination
22351 Show the current pagination mode.
22356 @cindex number representation
22357 @cindex entering numbers
22359 You can always enter numbers in octal, decimal, or hexadecimal in
22360 @value{GDBN} by the usual conventions: octal numbers begin with
22361 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
22362 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
22363 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
22364 10; likewise, the default display for numbers---when no particular
22365 format is specified---is base 10. You can change the default base for
22366 both input and output with the commands described below.
22369 @kindex set input-radix
22370 @item set input-radix @var{base}
22371 Set the default base for numeric input. Supported choices
22372 for @var{base} are decimal 8, 10, or 16. The base must itself be
22373 specified either unambiguously or using the current input radix; for
22377 set input-radix 012
22378 set input-radix 10.
22379 set input-radix 0xa
22383 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
22384 leaves the input radix unchanged, no matter what it was, since
22385 @samp{10}, being without any leading or trailing signs of its base, is
22386 interpreted in the current radix. Thus, if the current radix is 16,
22387 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
22390 @kindex set output-radix
22391 @item set output-radix @var{base}
22392 Set the default base for numeric display. Supported choices
22393 for @var{base} are decimal 8, 10, or 16. The base must itself be
22394 specified either unambiguously or using the current input radix.
22396 @kindex show input-radix
22397 @item show input-radix
22398 Display the current default base for numeric input.
22400 @kindex show output-radix
22401 @item show output-radix
22402 Display the current default base for numeric display.
22404 @item set radix @r{[}@var{base}@r{]}
22408 These commands set and show the default base for both input and output
22409 of numbers. @code{set radix} sets the radix of input and output to
22410 the same base; without an argument, it resets the radix back to its
22411 default value of 10.
22416 @section Configuring the Current ABI
22418 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
22419 application automatically. However, sometimes you need to override its
22420 conclusions. Use these commands to manage @value{GDBN}'s view of the
22426 @cindex Newlib OS ABI and its influence on the longjmp handling
22428 One @value{GDBN} configuration can debug binaries for multiple operating
22429 system targets, either via remote debugging or native emulation.
22430 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
22431 but you can override its conclusion using the @code{set osabi} command.
22432 One example where this is useful is in debugging of binaries which use
22433 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
22434 not have the same identifying marks that the standard C library for your
22437 When @value{GDBN} is debugging the AArch64 architecture, it provides a
22438 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
22439 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
22440 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
22444 Show the OS ABI currently in use.
22447 With no argument, show the list of registered available OS ABI's.
22449 @item set osabi @var{abi}
22450 Set the current OS ABI to @var{abi}.
22453 @cindex float promotion
22455 Generally, the way that an argument of type @code{float} is passed to a
22456 function depends on whether the function is prototyped. For a prototyped
22457 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
22458 according to the architecture's convention for @code{float}. For unprototyped
22459 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
22460 @code{double} and then passed.
22462 Unfortunately, some forms of debug information do not reliably indicate whether
22463 a function is prototyped. If @value{GDBN} calls a function that is not marked
22464 as prototyped, it consults @kbd{set coerce-float-to-double}.
22467 @kindex set coerce-float-to-double
22468 @item set coerce-float-to-double
22469 @itemx set coerce-float-to-double on
22470 Arguments of type @code{float} will be promoted to @code{double} when passed
22471 to an unprototyped function. This is the default setting.
22473 @item set coerce-float-to-double off
22474 Arguments of type @code{float} will be passed directly to unprototyped
22477 @kindex show coerce-float-to-double
22478 @item show coerce-float-to-double
22479 Show the current setting of promoting @code{float} to @code{double}.
22483 @kindex show cp-abi
22484 @value{GDBN} needs to know the ABI used for your program's C@t{++}
22485 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
22486 used to build your application. @value{GDBN} only fully supports
22487 programs with a single C@t{++} ABI; if your program contains code using
22488 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
22489 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
22490 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
22491 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
22492 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
22493 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
22498 Show the C@t{++} ABI currently in use.
22501 With no argument, show the list of supported C@t{++} ABI's.
22503 @item set cp-abi @var{abi}
22504 @itemx set cp-abi auto
22505 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
22509 @section Automatically loading associated files
22510 @cindex auto-loading
22512 @value{GDBN} sometimes reads files with commands and settings automatically,
22513 without being explicitly told so by the user. We call this feature
22514 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
22515 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
22516 results or introduce security risks (e.g., if the file comes from untrusted
22520 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
22521 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
22523 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
22524 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
22527 There are various kinds of files @value{GDBN} can automatically load.
22528 In addition to these files, @value{GDBN} supports auto-loading code written
22529 in various extension languages. @xref{Auto-loading extensions}.
22531 Note that loading of these associated files (including the local @file{.gdbinit}
22532 file) requires accordingly configured @code{auto-load safe-path}
22533 (@pxref{Auto-loading safe path}).
22535 For these reasons, @value{GDBN} includes commands and options to let you
22536 control when to auto-load files and which files should be auto-loaded.
22539 @anchor{set auto-load off}
22540 @kindex set auto-load off
22541 @item set auto-load off
22542 Globally disable loading of all auto-loaded files.
22543 You may want to use this command with the @samp{-iex} option
22544 (@pxref{Option -init-eval-command}) such as:
22546 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
22549 Be aware that system init file (@pxref{System-wide configuration})
22550 and init files from your home directory (@pxref{Home Directory Init File})
22551 still get read (as they come from generally trusted directories).
22552 To prevent @value{GDBN} from auto-loading even those init files, use the
22553 @option{-nx} option (@pxref{Mode Options}), in addition to
22554 @code{set auto-load no}.
22556 @anchor{show auto-load}
22557 @kindex show auto-load
22558 @item show auto-load
22559 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
22563 (gdb) show auto-load
22564 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
22565 libthread-db: Auto-loading of inferior specific libthread_db is on.
22566 local-gdbinit: Auto-loading of .gdbinit script from current directory
22568 python-scripts: Auto-loading of Python scripts is on.
22569 safe-path: List of directories from which it is safe to auto-load files
22570 is $debugdir:$datadir/auto-load.
22571 scripts-directory: List of directories from which to load auto-loaded scripts
22572 is $debugdir:$datadir/auto-load.
22575 @anchor{info auto-load}
22576 @kindex info auto-load
22577 @item info auto-load
22578 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
22582 (gdb) info auto-load
22585 Yes /home/user/gdb/gdb-gdb.gdb
22586 libthread-db: No auto-loaded libthread-db.
22587 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
22591 Yes /home/user/gdb/gdb-gdb.py
22595 These are @value{GDBN} control commands for the auto-loading:
22597 @multitable @columnfractions .5 .5
22598 @item @xref{set auto-load off}.
22599 @tab Disable auto-loading globally.
22600 @item @xref{show auto-load}.
22601 @tab Show setting of all kinds of files.
22602 @item @xref{info auto-load}.
22603 @tab Show state of all kinds of files.
22604 @item @xref{set auto-load gdb-scripts}.
22605 @tab Control for @value{GDBN} command scripts.
22606 @item @xref{show auto-load gdb-scripts}.
22607 @tab Show setting of @value{GDBN} command scripts.
22608 @item @xref{info auto-load gdb-scripts}.
22609 @tab Show state of @value{GDBN} command scripts.
22610 @item @xref{set auto-load python-scripts}.
22611 @tab Control for @value{GDBN} Python scripts.
22612 @item @xref{show auto-load python-scripts}.
22613 @tab Show setting of @value{GDBN} Python scripts.
22614 @item @xref{info auto-load python-scripts}.
22615 @tab Show state of @value{GDBN} Python scripts.
22616 @item @xref{set auto-load guile-scripts}.
22617 @tab Control for @value{GDBN} Guile scripts.
22618 @item @xref{show auto-load guile-scripts}.
22619 @tab Show setting of @value{GDBN} Guile scripts.
22620 @item @xref{info auto-load guile-scripts}.
22621 @tab Show state of @value{GDBN} Guile scripts.
22622 @item @xref{set auto-load scripts-directory}.
22623 @tab Control for @value{GDBN} auto-loaded scripts location.
22624 @item @xref{show auto-load scripts-directory}.
22625 @tab Show @value{GDBN} auto-loaded scripts location.
22626 @item @xref{add-auto-load-scripts-directory}.
22627 @tab Add directory for auto-loaded scripts location list.
22628 @item @xref{set auto-load local-gdbinit}.
22629 @tab Control for init file in the current directory.
22630 @item @xref{show auto-load local-gdbinit}.
22631 @tab Show setting of init file in the current directory.
22632 @item @xref{info auto-load local-gdbinit}.
22633 @tab Show state of init file in the current directory.
22634 @item @xref{set auto-load libthread-db}.
22635 @tab Control for thread debugging library.
22636 @item @xref{show auto-load libthread-db}.
22637 @tab Show setting of thread debugging library.
22638 @item @xref{info auto-load libthread-db}.
22639 @tab Show state of thread debugging library.
22640 @item @xref{set auto-load safe-path}.
22641 @tab Control directories trusted for automatic loading.
22642 @item @xref{show auto-load safe-path}.
22643 @tab Show directories trusted for automatic loading.
22644 @item @xref{add-auto-load-safe-path}.
22645 @tab Add directory trusted for automatic loading.
22648 @node Init File in the Current Directory
22649 @subsection Automatically loading init file in the current directory
22650 @cindex auto-loading init file in the current directory
22652 By default, @value{GDBN} reads and executes the canned sequences of commands
22653 from init file (if any) in the current working directory,
22654 see @ref{Init File in the Current Directory during Startup}.
22656 Note that loading of this local @file{.gdbinit} file also requires accordingly
22657 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
22660 @anchor{set auto-load local-gdbinit}
22661 @kindex set auto-load local-gdbinit
22662 @item set auto-load local-gdbinit [on|off]
22663 Enable or disable the auto-loading of canned sequences of commands
22664 (@pxref{Sequences}) found in init file in the current directory.
22666 @anchor{show auto-load local-gdbinit}
22667 @kindex show auto-load local-gdbinit
22668 @item show auto-load local-gdbinit
22669 Show whether auto-loading of canned sequences of commands from init file in the
22670 current directory is enabled or disabled.
22672 @anchor{info auto-load local-gdbinit}
22673 @kindex info auto-load local-gdbinit
22674 @item info auto-load local-gdbinit
22675 Print whether canned sequences of commands from init file in the
22676 current directory have been auto-loaded.
22679 @node libthread_db.so.1 file
22680 @subsection Automatically loading thread debugging library
22681 @cindex auto-loading libthread_db.so.1
22683 This feature is currently present only on @sc{gnu}/Linux native hosts.
22685 @value{GDBN} reads in some cases thread debugging library from places specific
22686 to the inferior (@pxref{set libthread-db-search-path}).
22688 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
22689 without checking this @samp{set auto-load libthread-db} switch as system
22690 libraries have to be trusted in general. In all other cases of
22691 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
22692 auto-load libthread-db} is enabled before trying to open such thread debugging
22695 Note that loading of this debugging library also requires accordingly configured
22696 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
22699 @anchor{set auto-load libthread-db}
22700 @kindex set auto-load libthread-db
22701 @item set auto-load libthread-db [on|off]
22702 Enable or disable the auto-loading of inferior specific thread debugging library.
22704 @anchor{show auto-load libthread-db}
22705 @kindex show auto-load libthread-db
22706 @item show auto-load libthread-db
22707 Show whether auto-loading of inferior specific thread debugging library is
22708 enabled or disabled.
22710 @anchor{info auto-load libthread-db}
22711 @kindex info auto-load libthread-db
22712 @item info auto-load libthread-db
22713 Print the list of all loaded inferior specific thread debugging libraries and
22714 for each such library print list of inferior @var{pid}s using it.
22717 @node Auto-loading safe path
22718 @subsection Security restriction for auto-loading
22719 @cindex auto-loading safe-path
22721 As the files of inferior can come from untrusted source (such as submitted by
22722 an application user) @value{GDBN} does not always load any files automatically.
22723 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
22724 directories trusted for loading files not explicitly requested by user.
22725 Each directory can also be a shell wildcard pattern.
22727 If the path is not set properly you will see a warning and the file will not
22732 Reading symbols from /home/user/gdb/gdb...done.
22733 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
22734 declined by your `auto-load safe-path' set
22735 to "$debugdir:$datadir/auto-load".
22736 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
22737 declined by your `auto-load safe-path' set
22738 to "$debugdir:$datadir/auto-load".
22742 To instruct @value{GDBN} to go ahead and use the init files anyway,
22743 invoke @value{GDBN} like this:
22746 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
22749 The list of trusted directories is controlled by the following commands:
22752 @anchor{set auto-load safe-path}
22753 @kindex set auto-load safe-path
22754 @item set auto-load safe-path @r{[}@var{directories}@r{]}
22755 Set the list of directories (and their subdirectories) trusted for automatic
22756 loading and execution of scripts. You can also enter a specific trusted file.
22757 Each directory can also be a shell wildcard pattern; wildcards do not match
22758 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
22759 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
22760 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
22761 its default value as specified during @value{GDBN} compilation.
22763 The list of directories uses path separator (@samp{:} on GNU and Unix
22764 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
22765 to the @env{PATH} environment variable.
22767 @anchor{show auto-load safe-path}
22768 @kindex show auto-load safe-path
22769 @item show auto-load safe-path
22770 Show the list of directories trusted for automatic loading and execution of
22773 @anchor{add-auto-load-safe-path}
22774 @kindex add-auto-load-safe-path
22775 @item add-auto-load-safe-path
22776 Add an entry (or list of entries) to the list of directories trusted for
22777 automatic loading and execution of scripts. Multiple entries may be delimited
22778 by the host platform path separator in use.
22781 This variable defaults to what @code{--with-auto-load-dir} has been configured
22782 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
22783 substitution applies the same as for @ref{set auto-load scripts-directory}.
22784 The default @code{set auto-load safe-path} value can be also overriden by
22785 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
22787 Setting this variable to @file{/} disables this security protection,
22788 corresponding @value{GDBN} configuration option is
22789 @option{--without-auto-load-safe-path}.
22790 This variable is supposed to be set to the system directories writable by the
22791 system superuser only. Users can add their source directories in init files in
22792 their home directories (@pxref{Home Directory Init File}). See also deprecated
22793 init file in the current directory
22794 (@pxref{Init File in the Current Directory during Startup}).
22796 To force @value{GDBN} to load the files it declined to load in the previous
22797 example, you could use one of the following ways:
22800 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
22801 Specify this trusted directory (or a file) as additional component of the list.
22802 You have to specify also any existing directories displayed by
22803 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
22805 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
22806 Specify this directory as in the previous case but just for a single
22807 @value{GDBN} session.
22809 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
22810 Disable auto-loading safety for a single @value{GDBN} session.
22811 This assumes all the files you debug during this @value{GDBN} session will come
22812 from trusted sources.
22814 @item @kbd{./configure --without-auto-load-safe-path}
22815 During compilation of @value{GDBN} you may disable any auto-loading safety.
22816 This assumes all the files you will ever debug with this @value{GDBN} come from
22820 On the other hand you can also explicitly forbid automatic files loading which
22821 also suppresses any such warning messages:
22824 @item @kbd{gdb -iex "set auto-load no" @dots{}}
22825 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
22827 @item @file{~/.gdbinit}: @samp{set auto-load no}
22828 Disable auto-loading globally for the user
22829 (@pxref{Home Directory Init File}). While it is improbable, you could also
22830 use system init file instead (@pxref{System-wide configuration}).
22833 This setting applies to the file names as entered by user. If no entry matches
22834 @value{GDBN} tries as a last resort to also resolve all the file names into
22835 their canonical form (typically resolving symbolic links) and compare the
22836 entries again. @value{GDBN} already canonicalizes most of the filenames on its
22837 own before starting the comparison so a canonical form of directories is
22838 recommended to be entered.
22840 @node Auto-loading verbose mode
22841 @subsection Displaying files tried for auto-load
22842 @cindex auto-loading verbose mode
22844 For better visibility of all the file locations where you can place scripts to
22845 be auto-loaded with inferior --- or to protect yourself against accidental
22846 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
22847 all the files attempted to be loaded. Both existing and non-existing files may
22850 For example the list of directories from which it is safe to auto-load files
22851 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
22852 may not be too obvious while setting it up.
22855 (gdb) set debug auto-load on
22856 (gdb) file ~/src/t/true
22857 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
22858 for objfile "/tmp/true".
22859 auto-load: Updating directories of "/usr:/opt".
22860 auto-load: Using directory "/usr".
22861 auto-load: Using directory "/opt".
22862 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
22863 by your `auto-load safe-path' set to "/usr:/opt".
22867 @anchor{set debug auto-load}
22868 @kindex set debug auto-load
22869 @item set debug auto-load [on|off]
22870 Set whether to print the filenames attempted to be auto-loaded.
22872 @anchor{show debug auto-load}
22873 @kindex show debug auto-load
22874 @item show debug auto-load
22875 Show whether printing of the filenames attempted to be auto-loaded is turned
22879 @node Messages/Warnings
22880 @section Optional Warnings and Messages
22882 @cindex verbose operation
22883 @cindex optional warnings
22884 By default, @value{GDBN} is silent about its inner workings. If you are
22885 running on a slow machine, you may want to use the @code{set verbose}
22886 command. This makes @value{GDBN} tell you when it does a lengthy
22887 internal operation, so you will not think it has crashed.
22889 Currently, the messages controlled by @code{set verbose} are those
22890 which announce that the symbol table for a source file is being read;
22891 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
22894 @kindex set verbose
22895 @item set verbose on
22896 Enables @value{GDBN} output of certain informational messages.
22898 @item set verbose off
22899 Disables @value{GDBN} output of certain informational messages.
22901 @kindex show verbose
22903 Displays whether @code{set verbose} is on or off.
22906 By default, if @value{GDBN} encounters bugs in the symbol table of an
22907 object file, it is silent; but if you are debugging a compiler, you may
22908 find this information useful (@pxref{Symbol Errors, ,Errors Reading
22913 @kindex set complaints
22914 @item set complaints @var{limit}
22915 Permits @value{GDBN} to output @var{limit} complaints about each type of
22916 unusual symbols before becoming silent about the problem. Set
22917 @var{limit} to zero to suppress all complaints; set it to a large number
22918 to prevent complaints from being suppressed.
22920 @kindex show complaints
22921 @item show complaints
22922 Displays how many symbol complaints @value{GDBN} is permitted to produce.
22926 @anchor{confirmation requests}
22927 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
22928 lot of stupid questions to confirm certain commands. For example, if
22929 you try to run a program which is already running:
22933 The program being debugged has been started already.
22934 Start it from the beginning? (y or n)
22937 If you are willing to unflinchingly face the consequences of your own
22938 commands, you can disable this ``feature'':
22942 @kindex set confirm
22944 @cindex confirmation
22945 @cindex stupid questions
22946 @item set confirm off
22947 Disables confirmation requests. Note that running @value{GDBN} with
22948 the @option{--batch} option (@pxref{Mode Options, -batch}) also
22949 automatically disables confirmation requests.
22951 @item set confirm on
22952 Enables confirmation requests (the default).
22954 @kindex show confirm
22956 Displays state of confirmation requests.
22960 @cindex command tracing
22961 If you need to debug user-defined commands or sourced files you may find it
22962 useful to enable @dfn{command tracing}. In this mode each command will be
22963 printed as it is executed, prefixed with one or more @samp{+} symbols, the
22964 quantity denoting the call depth of each command.
22967 @kindex set trace-commands
22968 @cindex command scripts, debugging
22969 @item set trace-commands on
22970 Enable command tracing.
22971 @item set trace-commands off
22972 Disable command tracing.
22973 @item show trace-commands
22974 Display the current state of command tracing.
22977 @node Debugging Output
22978 @section Optional Messages about Internal Happenings
22979 @cindex optional debugging messages
22981 @value{GDBN} has commands that enable optional debugging messages from
22982 various @value{GDBN} subsystems; normally these commands are of
22983 interest to @value{GDBN} maintainers, or when reporting a bug. This
22984 section documents those commands.
22987 @kindex set exec-done-display
22988 @item set exec-done-display
22989 Turns on or off the notification of asynchronous commands'
22990 completion. When on, @value{GDBN} will print a message when an
22991 asynchronous command finishes its execution. The default is off.
22992 @kindex show exec-done-display
22993 @item show exec-done-display
22994 Displays the current setting of asynchronous command completion
22997 @cindex ARM AArch64
22998 @item set debug aarch64
22999 Turns on or off display of debugging messages related to ARM AArch64.
23000 The default is off.
23002 @item show debug aarch64
23003 Displays the current state of displaying debugging messages related to
23005 @cindex gdbarch debugging info
23006 @cindex architecture debugging info
23007 @item set debug arch
23008 Turns on or off display of gdbarch debugging info. The default is off
23009 @item show debug arch
23010 Displays the current state of displaying gdbarch debugging info.
23011 @item set debug aix-solib
23012 @cindex AIX shared library debugging
23013 Control display of debugging messages from the AIX shared library
23014 support module. The default is off.
23015 @item show debug aix-thread
23016 Show the current state of displaying AIX shared library debugging messages.
23017 @item set debug aix-thread
23018 @cindex AIX threads
23019 Display debugging messages about inner workings of the AIX thread
23021 @item show debug aix-thread
23022 Show the current state of AIX thread debugging info display.
23023 @item set debug check-physname
23025 Check the results of the ``physname'' computation. When reading DWARF
23026 debugging information for C@t{++}, @value{GDBN} attempts to compute
23027 each entity's name. @value{GDBN} can do this computation in two
23028 different ways, depending on exactly what information is present.
23029 When enabled, this setting causes @value{GDBN} to compute the names
23030 both ways and display any discrepancies.
23031 @item show debug check-physname
23032 Show the current state of ``physname'' checking.
23033 @item set debug coff-pe-read
23034 @cindex COFF/PE exported symbols
23035 Control display of debugging messages related to reading of COFF/PE
23036 exported symbols. The default is off.
23037 @item show debug coff-pe-read
23038 Displays the current state of displaying debugging messages related to
23039 reading of COFF/PE exported symbols.
23040 @item set debug dwarf2-die
23041 @cindex DWARF2 DIEs
23042 Dump DWARF2 DIEs after they are read in.
23043 The value is the number of nesting levels to print.
23044 A value of zero turns off the display.
23045 @item show debug dwarf2-die
23046 Show the current state of DWARF2 DIE debugging.
23047 @item set debug dwarf2-read
23048 @cindex DWARF2 Reading
23049 Turns on or off display of debugging messages related to reading
23050 DWARF debug info. The default is 0 (off).
23051 A value of 1 provides basic information.
23052 A value greater than 1 provides more verbose information.
23053 @item show debug dwarf2-read
23054 Show the current state of DWARF2 reader debugging.
23055 @item set debug displaced
23056 @cindex displaced stepping debugging info
23057 Turns on or off display of @value{GDBN} debugging info for the
23058 displaced stepping support. The default is off.
23059 @item show debug displaced
23060 Displays the current state of displaying @value{GDBN} debugging info
23061 related to displaced stepping.
23062 @item set debug event
23063 @cindex event debugging info
23064 Turns on or off display of @value{GDBN} event debugging info. The
23066 @item show debug event
23067 Displays the current state of displaying @value{GDBN} event debugging
23069 @item set debug expression
23070 @cindex expression debugging info
23071 Turns on or off display of debugging info about @value{GDBN}
23072 expression parsing. The default is off.
23073 @item show debug expression
23074 Displays the current state of displaying debugging info about
23075 @value{GDBN} expression parsing.
23076 @item set debug frame
23077 @cindex frame debugging info
23078 Turns on or off display of @value{GDBN} frame debugging info. The
23080 @item show debug frame
23081 Displays the current state of displaying @value{GDBN} frame debugging
23083 @item set debug gnu-nat
23084 @cindex @sc{gnu}/Hurd debug messages
23085 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
23086 @item show debug gnu-nat
23087 Show the current state of @sc{gnu}/Hurd debugging messages.
23088 @item set debug infrun
23089 @cindex inferior debugging info
23090 Turns on or off display of @value{GDBN} debugging info for running the inferior.
23091 The default is off. @file{infrun.c} contains GDB's runtime state machine used
23092 for implementing operations such as single-stepping the inferior.
23093 @item show debug infrun
23094 Displays the current state of @value{GDBN} inferior debugging.
23095 @item set debug jit
23096 @cindex just-in-time compilation, debugging messages
23097 Turns on or off debugging messages from JIT debug support.
23098 @item show debug jit
23099 Displays the current state of @value{GDBN} JIT debugging.
23100 @item set debug lin-lwp
23101 @cindex @sc{gnu}/Linux LWP debug messages
23102 @cindex Linux lightweight processes
23103 Turns on or off debugging messages from the Linux LWP debug support.
23104 @item show debug lin-lwp
23105 Show the current state of Linux LWP debugging messages.
23106 @item set debug mach-o
23107 @cindex Mach-O symbols processing
23108 Control display of debugging messages related to Mach-O symbols
23109 processing. The default is off.
23110 @item show debug mach-o
23111 Displays the current state of displaying debugging messages related to
23112 reading of COFF/PE exported symbols.
23113 @item set debug notification
23114 @cindex remote async notification debugging info
23115 Turns on or off debugging messages about remote async notification.
23116 The default is off.
23117 @item show debug notification
23118 Displays the current state of remote async notification debugging messages.
23119 @item set debug observer
23120 @cindex observer debugging info
23121 Turns on or off display of @value{GDBN} observer debugging. This
23122 includes info such as the notification of observable events.
23123 @item show debug observer
23124 Displays the current state of observer debugging.
23125 @item set debug overload
23126 @cindex C@t{++} overload debugging info
23127 Turns on or off display of @value{GDBN} C@t{++} overload debugging
23128 info. This includes info such as ranking of functions, etc. The default
23130 @item show debug overload
23131 Displays the current state of displaying @value{GDBN} C@t{++} overload
23133 @cindex expression parser, debugging info
23134 @cindex debug expression parser
23135 @item set debug parser
23136 Turns on or off the display of expression parser debugging output.
23137 Internally, this sets the @code{yydebug} variable in the expression
23138 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
23139 details. The default is off.
23140 @item show debug parser
23141 Show the current state of expression parser debugging.
23142 @cindex packets, reporting on stdout
23143 @cindex serial connections, debugging
23144 @cindex debug remote protocol
23145 @cindex remote protocol debugging
23146 @cindex display remote packets
23147 @item set debug remote
23148 Turns on or off display of reports on all packets sent back and forth across
23149 the serial line to the remote machine. The info is printed on the
23150 @value{GDBN} standard output stream. The default is off.
23151 @item show debug remote
23152 Displays the state of display of remote packets.
23153 @item set debug serial
23154 Turns on or off display of @value{GDBN} serial debugging info. The
23156 @item show debug serial
23157 Displays the current state of displaying @value{GDBN} serial debugging
23159 @item set debug solib-frv
23160 @cindex FR-V shared-library debugging
23161 Turns on or off debugging messages for FR-V shared-library code.
23162 @item show debug solib-frv
23163 Display the current state of FR-V shared-library code debugging
23165 @item set debug symbol-lookup
23166 @cindex symbol lookup
23167 Turns on or off display of debugging messages related to symbol lookup.
23168 The default is 0 (off).
23169 A value of 1 provides basic information.
23170 A value greater than 1 provides more verbose information.
23171 @item show debug symbol-lookup
23172 Show the current state of symbol lookup debugging messages.
23173 @item set debug symfile
23174 @cindex symbol file functions
23175 Turns on or off display of debugging messages related to symbol file functions.
23176 The default is off. @xref{Files}.
23177 @item show debug symfile
23178 Show the current state of symbol file debugging messages.
23179 @item set debug symtab-create
23180 @cindex symbol table creation
23181 Turns on or off display of debugging messages related to symbol table creation.
23182 The default is 0 (off).
23183 A value of 1 provides basic information.
23184 A value greater than 1 provides more verbose information.
23185 @item show debug symtab-create
23186 Show the current state of symbol table creation debugging.
23187 @item set debug target
23188 @cindex target debugging info
23189 Turns on or off display of @value{GDBN} target debugging info. This info
23190 includes what is going on at the target level of GDB, as it happens. The
23191 default is 0. Set it to 1 to track events, and to 2 to also track the
23192 value of large memory transfers.
23193 @item show debug target
23194 Displays the current state of displaying @value{GDBN} target debugging
23196 @item set debug timestamp
23197 @cindex timestampping debugging info
23198 Turns on or off display of timestamps with @value{GDBN} debugging info.
23199 When enabled, seconds and microseconds are displayed before each debugging
23201 @item show debug timestamp
23202 Displays the current state of displaying timestamps with @value{GDBN}
23204 @item set debug varobj
23205 @cindex variable object debugging info
23206 Turns on or off display of @value{GDBN} variable object debugging
23207 info. The default is off.
23208 @item show debug varobj
23209 Displays the current state of displaying @value{GDBN} variable object
23211 @item set debug xml
23212 @cindex XML parser debugging
23213 Turns on or off debugging messages for built-in XML parsers.
23214 @item show debug xml
23215 Displays the current state of XML debugging messages.
23218 @node Other Misc Settings
23219 @section Other Miscellaneous Settings
23220 @cindex miscellaneous settings
23223 @kindex set interactive-mode
23224 @item set interactive-mode
23225 If @code{on}, forces @value{GDBN} to assume that GDB was started
23226 in a terminal. In practice, this means that @value{GDBN} should wait
23227 for the user to answer queries generated by commands entered at
23228 the command prompt. If @code{off}, forces @value{GDBN} to operate
23229 in the opposite mode, and it uses the default answers to all queries.
23230 If @code{auto} (the default), @value{GDBN} tries to determine whether
23231 its standard input is a terminal, and works in interactive-mode if it
23232 is, non-interactively otherwise.
23234 In the vast majority of cases, the debugger should be able to guess
23235 correctly which mode should be used. But this setting can be useful
23236 in certain specific cases, such as running a MinGW @value{GDBN}
23237 inside a cygwin window.
23239 @kindex show interactive-mode
23240 @item show interactive-mode
23241 Displays whether the debugger is operating in interactive mode or not.
23244 @node Extending GDB
23245 @chapter Extending @value{GDBN}
23246 @cindex extending GDB
23248 @value{GDBN} provides several mechanisms for extension.
23249 @value{GDBN} also provides the ability to automatically load
23250 extensions when it reads a file for debugging. This allows the
23251 user to automatically customize @value{GDBN} for the program
23255 * Sequences:: Canned Sequences of @value{GDBN} Commands
23256 * Python:: Extending @value{GDBN} using Python
23257 * Guile:: Extending @value{GDBN} using Guile
23258 * Auto-loading extensions:: Automatically loading extensions
23259 * Multiple Extension Languages:: Working with multiple extension languages
23260 * Aliases:: Creating new spellings of existing commands
23263 To facilitate the use of extension languages, @value{GDBN} is capable
23264 of evaluating the contents of a file. When doing so, @value{GDBN}
23265 can recognize which extension language is being used by looking at
23266 the filename extension. Files with an unrecognized filename extension
23267 are always treated as a @value{GDBN} Command Files.
23268 @xref{Command Files,, Command files}.
23270 You can control how @value{GDBN} evaluates these files with the following
23274 @kindex set script-extension
23275 @kindex show script-extension
23276 @item set script-extension off
23277 All scripts are always evaluated as @value{GDBN} Command Files.
23279 @item set script-extension soft
23280 The debugger determines the scripting language based on filename
23281 extension. If this scripting language is supported, @value{GDBN}
23282 evaluates the script using that language. Otherwise, it evaluates
23283 the file as a @value{GDBN} Command File.
23285 @item set script-extension strict
23286 The debugger determines the scripting language based on filename
23287 extension, and evaluates the script using that language. If the
23288 language is not supported, then the evaluation fails.
23290 @item show script-extension
23291 Display the current value of the @code{script-extension} option.
23296 @section Canned Sequences of Commands
23298 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
23299 Command Lists}), @value{GDBN} provides two ways to store sequences of
23300 commands for execution as a unit: user-defined commands and command
23304 * Define:: How to define your own commands
23305 * Hooks:: Hooks for user-defined commands
23306 * Command Files:: How to write scripts of commands to be stored in a file
23307 * Output:: Commands for controlled output
23308 * Auto-loading sequences:: Controlling auto-loaded command files
23312 @subsection User-defined Commands
23314 @cindex user-defined command
23315 @cindex arguments, to user-defined commands
23316 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
23317 which you assign a new name as a command. This is done with the
23318 @code{define} command. User commands may accept up to 10 arguments
23319 separated by whitespace. Arguments are accessed within the user command
23320 via @code{$arg0@dots{}$arg9}. A trivial example:
23324 print $arg0 + $arg1 + $arg2
23329 To execute the command use:
23336 This defines the command @code{adder}, which prints the sum of
23337 its three arguments. Note the arguments are text substitutions, so they may
23338 reference variables, use complex expressions, or even perform inferior
23341 @cindex argument count in user-defined commands
23342 @cindex how many arguments (user-defined commands)
23343 In addition, @code{$argc} may be used to find out how many arguments have
23344 been passed. This expands to a number in the range 0@dots{}10.
23349 print $arg0 + $arg1
23352 print $arg0 + $arg1 + $arg2
23360 @item define @var{commandname}
23361 Define a command named @var{commandname}. If there is already a command
23362 by that name, you are asked to confirm that you want to redefine it.
23363 The argument @var{commandname} may be a bare command name consisting of letters,
23364 numbers, dashes, and underscores. It may also start with any predefined
23365 prefix command. For example, @samp{define target my-target} creates
23366 a user-defined @samp{target my-target} command.
23368 The definition of the command is made up of other @value{GDBN} command lines,
23369 which are given following the @code{define} command. The end of these
23370 commands is marked by a line containing @code{end}.
23373 @kindex end@r{ (user-defined commands)}
23374 @item document @var{commandname}
23375 Document the user-defined command @var{commandname}, so that it can be
23376 accessed by @code{help}. The command @var{commandname} must already be
23377 defined. This command reads lines of documentation just as @code{define}
23378 reads the lines of the command definition, ending with @code{end}.
23379 After the @code{document} command is finished, @code{help} on command
23380 @var{commandname} displays the documentation you have written.
23382 You may use the @code{document} command again to change the
23383 documentation of a command. Redefining the command with @code{define}
23384 does not change the documentation.
23386 @kindex dont-repeat
23387 @cindex don't repeat command
23389 Used inside a user-defined command, this tells @value{GDBN} that this
23390 command should not be repeated when the user hits @key{RET}
23391 (@pxref{Command Syntax, repeat last command}).
23393 @kindex help user-defined
23394 @item help user-defined
23395 List all user-defined commands and all python commands defined in class
23396 COMAND_USER. The first line of the documentation or docstring is
23401 @itemx show user @var{commandname}
23402 Display the @value{GDBN} commands used to define @var{commandname} (but
23403 not its documentation). If no @var{commandname} is given, display the
23404 definitions for all user-defined commands.
23405 This does not work for user-defined python commands.
23407 @cindex infinite recursion in user-defined commands
23408 @kindex show max-user-call-depth
23409 @kindex set max-user-call-depth
23410 @item show max-user-call-depth
23411 @itemx set max-user-call-depth
23412 The value of @code{max-user-call-depth} controls how many recursion
23413 levels are allowed in user-defined commands before @value{GDBN} suspects an
23414 infinite recursion and aborts the command.
23415 This does not apply to user-defined python commands.
23418 In addition to the above commands, user-defined commands frequently
23419 use control flow commands, described in @ref{Command Files}.
23421 When user-defined commands are executed, the
23422 commands of the definition are not printed. An error in any command
23423 stops execution of the user-defined command.
23425 If used interactively, commands that would ask for confirmation proceed
23426 without asking when used inside a user-defined command. Many @value{GDBN}
23427 commands that normally print messages to say what they are doing omit the
23428 messages when used in a user-defined command.
23431 @subsection User-defined Command Hooks
23432 @cindex command hooks
23433 @cindex hooks, for commands
23434 @cindex hooks, pre-command
23437 You may define @dfn{hooks}, which are a special kind of user-defined
23438 command. Whenever you run the command @samp{foo}, if the user-defined
23439 command @samp{hook-foo} exists, it is executed (with no arguments)
23440 before that command.
23442 @cindex hooks, post-command
23444 A hook may also be defined which is run after the command you executed.
23445 Whenever you run the command @samp{foo}, if the user-defined command
23446 @samp{hookpost-foo} exists, it is executed (with no arguments) after
23447 that command. Post-execution hooks may exist simultaneously with
23448 pre-execution hooks, for the same command.
23450 It is valid for a hook to call the command which it hooks. If this
23451 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
23453 @c It would be nice if hookpost could be passed a parameter indicating
23454 @c if the command it hooks executed properly or not. FIXME!
23456 @kindex stop@r{, a pseudo-command}
23457 In addition, a pseudo-command, @samp{stop} exists. Defining
23458 (@samp{hook-stop}) makes the associated commands execute every time
23459 execution stops in your program: before breakpoint commands are run,
23460 displays are printed, or the stack frame is printed.
23462 For example, to ignore @code{SIGALRM} signals while
23463 single-stepping, but treat them normally during normal execution,
23468 handle SIGALRM nopass
23472 handle SIGALRM pass
23475 define hook-continue
23476 handle SIGALRM pass
23480 As a further example, to hook at the beginning and end of the @code{echo}
23481 command, and to add extra text to the beginning and end of the message,
23489 define hookpost-echo
23493 (@value{GDBP}) echo Hello World
23494 <<<---Hello World--->>>
23499 You can define a hook for any single-word command in @value{GDBN}, but
23500 not for command aliases; you should define a hook for the basic command
23501 name, e.g.@: @code{backtrace} rather than @code{bt}.
23502 @c FIXME! So how does Joe User discover whether a command is an alias
23504 You can hook a multi-word command by adding @code{hook-} or
23505 @code{hookpost-} to the last word of the command, e.g.@:
23506 @samp{define target hook-remote} to add a hook to @samp{target remote}.
23508 If an error occurs during the execution of your hook, execution of
23509 @value{GDBN} commands stops and @value{GDBN} issues a prompt
23510 (before the command that you actually typed had a chance to run).
23512 If you try to define a hook which does not match any known command, you
23513 get a warning from the @code{define} command.
23515 @node Command Files
23516 @subsection Command Files
23518 @cindex command files
23519 @cindex scripting commands
23520 A command file for @value{GDBN} is a text file made of lines that are
23521 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
23522 also be included. An empty line in a command file does nothing; it
23523 does not mean to repeat the last command, as it would from the
23526 You can request the execution of a command file with the @code{source}
23527 command. Note that the @code{source} command is also used to evaluate
23528 scripts that are not Command Files. The exact behavior can be configured
23529 using the @code{script-extension} setting.
23530 @xref{Extending GDB,, Extending GDB}.
23534 @cindex execute commands from a file
23535 @item source [-s] [-v] @var{filename}
23536 Execute the command file @var{filename}.
23539 The lines in a command file are generally executed sequentially,
23540 unless the order of execution is changed by one of the
23541 @emph{flow-control commands} described below. The commands are not
23542 printed as they are executed. An error in any command terminates
23543 execution of the command file and control is returned to the console.
23545 @value{GDBN} first searches for @var{filename} in the current directory.
23546 If the file is not found there, and @var{filename} does not specify a
23547 directory, then @value{GDBN} also looks for the file on the source search path
23548 (specified with the @samp{directory} command);
23549 except that @file{$cdir} is not searched because the compilation directory
23550 is not relevant to scripts.
23552 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
23553 on the search path even if @var{filename} specifies a directory.
23554 The search is done by appending @var{filename} to each element of the
23555 search path. So, for example, if @var{filename} is @file{mylib/myscript}
23556 and the search path contains @file{/home/user} then @value{GDBN} will
23557 look for the script @file{/home/user/mylib/myscript}.
23558 The search is also done if @var{filename} is an absolute path.
23559 For example, if @var{filename} is @file{/tmp/myscript} and
23560 the search path contains @file{/home/user} then @value{GDBN} will
23561 look for the script @file{/home/user/tmp/myscript}.
23562 For DOS-like systems, if @var{filename} contains a drive specification,
23563 it is stripped before concatenation. For example, if @var{filename} is
23564 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
23565 will look for the script @file{c:/tmp/myscript}.
23567 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
23568 each command as it is executed. The option must be given before
23569 @var{filename}, and is interpreted as part of the filename anywhere else.
23571 Commands that would ask for confirmation if used interactively proceed
23572 without asking when used in a command file. Many @value{GDBN} commands that
23573 normally print messages to say what they are doing omit the messages
23574 when called from command files.
23576 @value{GDBN} also accepts command input from standard input. In this
23577 mode, normal output goes to standard output and error output goes to
23578 standard error. Errors in a command file supplied on standard input do
23579 not terminate execution of the command file---execution continues with
23583 gdb < cmds > log 2>&1
23586 (The syntax above will vary depending on the shell used.) This example
23587 will execute commands from the file @file{cmds}. All output and errors
23588 would be directed to @file{log}.
23590 Since commands stored on command files tend to be more general than
23591 commands typed interactively, they frequently need to deal with
23592 complicated situations, such as different or unexpected values of
23593 variables and symbols, changes in how the program being debugged is
23594 built, etc. @value{GDBN} provides a set of flow-control commands to
23595 deal with these complexities. Using these commands, you can write
23596 complex scripts that loop over data structures, execute commands
23597 conditionally, etc.
23604 This command allows to include in your script conditionally executed
23605 commands. The @code{if} command takes a single argument, which is an
23606 expression to evaluate. It is followed by a series of commands that
23607 are executed only if the expression is true (its value is nonzero).
23608 There can then optionally be an @code{else} line, followed by a series
23609 of commands that are only executed if the expression was false. The
23610 end of the list is marked by a line containing @code{end}.
23614 This command allows to write loops. Its syntax is similar to
23615 @code{if}: the command takes a single argument, which is an expression
23616 to evaluate, and must be followed by the commands to execute, one per
23617 line, terminated by an @code{end}. These commands are called the
23618 @dfn{body} of the loop. The commands in the body of @code{while} are
23619 executed repeatedly as long as the expression evaluates to true.
23623 This command exits the @code{while} loop in whose body it is included.
23624 Execution of the script continues after that @code{while}s @code{end}
23627 @kindex loop_continue
23628 @item loop_continue
23629 This command skips the execution of the rest of the body of commands
23630 in the @code{while} loop in whose body it is included. Execution
23631 branches to the beginning of the @code{while} loop, where it evaluates
23632 the controlling expression.
23634 @kindex end@r{ (if/else/while commands)}
23636 Terminate the block of commands that are the body of @code{if},
23637 @code{else}, or @code{while} flow-control commands.
23642 @subsection Commands for Controlled Output
23644 During the execution of a command file or a user-defined command, normal
23645 @value{GDBN} output is suppressed; the only output that appears is what is
23646 explicitly printed by the commands in the definition. This section
23647 describes three commands useful for generating exactly the output you
23652 @item echo @var{text}
23653 @c I do not consider backslash-space a standard C escape sequence
23654 @c because it is not in ANSI.
23655 Print @var{text}. Nonprinting characters can be included in
23656 @var{text} using C escape sequences, such as @samp{\n} to print a
23657 newline. @strong{No newline is printed unless you specify one.}
23658 In addition to the standard C escape sequences, a backslash followed
23659 by a space stands for a space. This is useful for displaying a
23660 string with spaces at the beginning or the end, since leading and
23661 trailing spaces are otherwise trimmed from all arguments.
23662 To print @samp{@w{ }and foo =@w{ }}, use the command
23663 @samp{echo \@w{ }and foo = \@w{ }}.
23665 A backslash at the end of @var{text} can be used, as in C, to continue
23666 the command onto subsequent lines. For example,
23669 echo This is some text\n\
23670 which is continued\n\
23671 onto several lines.\n
23674 produces the same output as
23677 echo This is some text\n
23678 echo which is continued\n
23679 echo onto several lines.\n
23683 @item output @var{expression}
23684 Print the value of @var{expression} and nothing but that value: no
23685 newlines, no @samp{$@var{nn} = }. The value is not entered in the
23686 value history either. @xref{Expressions, ,Expressions}, for more information
23689 @item output/@var{fmt} @var{expression}
23690 Print the value of @var{expression} in format @var{fmt}. You can use
23691 the same formats as for @code{print}. @xref{Output Formats,,Output
23692 Formats}, for more information.
23695 @item printf @var{template}, @var{expressions}@dots{}
23696 Print the values of one or more @var{expressions} under the control of
23697 the string @var{template}. To print several values, make
23698 @var{expressions} be a comma-separated list of individual expressions,
23699 which may be either numbers or pointers. Their values are printed as
23700 specified by @var{template}, exactly as a C program would do by
23701 executing the code below:
23704 printf (@var{template}, @var{expressions}@dots{});
23707 As in @code{C} @code{printf}, ordinary characters in @var{template}
23708 are printed verbatim, while @dfn{conversion specification} introduced
23709 by the @samp{%} character cause subsequent @var{expressions} to be
23710 evaluated, their values converted and formatted according to type and
23711 style information encoded in the conversion specifications, and then
23714 For example, you can print two values in hex like this:
23717 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
23720 @code{printf} supports all the standard @code{C} conversion
23721 specifications, including the flags and modifiers between the @samp{%}
23722 character and the conversion letter, with the following exceptions:
23726 The argument-ordering modifiers, such as @samp{2$}, are not supported.
23729 The modifier @samp{*} is not supported for specifying precision or
23733 The @samp{'} flag (for separation of digits into groups according to
23734 @code{LC_NUMERIC'}) is not supported.
23737 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
23741 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
23744 The conversion letters @samp{a} and @samp{A} are not supported.
23748 Note that the @samp{ll} type modifier is supported only if the
23749 underlying @code{C} implementation used to build @value{GDBN} supports
23750 the @code{long long int} type, and the @samp{L} type modifier is
23751 supported only if @code{long double} type is available.
23753 As in @code{C}, @code{printf} supports simple backslash-escape
23754 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
23755 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
23756 single character. Octal and hexadecimal escape sequences are not
23759 Additionally, @code{printf} supports conversion specifications for DFP
23760 (@dfn{Decimal Floating Point}) types using the following length modifiers
23761 together with a floating point specifier.
23766 @samp{H} for printing @code{Decimal32} types.
23769 @samp{D} for printing @code{Decimal64} types.
23772 @samp{DD} for printing @code{Decimal128} types.
23775 If the underlying @code{C} implementation used to build @value{GDBN} has
23776 support for the three length modifiers for DFP types, other modifiers
23777 such as width and precision will also be available for @value{GDBN} to use.
23779 In case there is no such @code{C} support, no additional modifiers will be
23780 available and the value will be printed in the standard way.
23782 Here's an example of printing DFP types using the above conversion letters:
23784 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
23788 @item eval @var{template}, @var{expressions}@dots{}
23789 Convert the values of one or more @var{expressions} under the control of
23790 the string @var{template} to a command line, and call it.
23794 @node Auto-loading sequences
23795 @subsection Controlling auto-loading native @value{GDBN} scripts
23796 @cindex native script auto-loading
23798 When a new object file is read (for example, due to the @code{file}
23799 command, or because the inferior has loaded a shared library),
23800 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
23801 @xref{Auto-loading extensions}.
23803 Auto-loading can be enabled or disabled,
23804 and the list of auto-loaded scripts can be printed.
23807 @anchor{set auto-load gdb-scripts}
23808 @kindex set auto-load gdb-scripts
23809 @item set auto-load gdb-scripts [on|off]
23810 Enable or disable the auto-loading of canned sequences of commands scripts.
23812 @anchor{show auto-load gdb-scripts}
23813 @kindex show auto-load gdb-scripts
23814 @item show auto-load gdb-scripts
23815 Show whether auto-loading of canned sequences of commands scripts is enabled or
23818 @anchor{info auto-load gdb-scripts}
23819 @kindex info auto-load gdb-scripts
23820 @cindex print list of auto-loaded canned sequences of commands scripts
23821 @item info auto-load gdb-scripts [@var{regexp}]
23822 Print the list of all canned sequences of commands scripts that @value{GDBN}
23826 If @var{regexp} is supplied only canned sequences of commands scripts with
23827 matching names are printed.
23829 @c Python docs live in a separate file.
23830 @include python.texi
23832 @c Guile docs live in a separate file.
23833 @include guile.texi
23835 @node Auto-loading extensions
23836 @section Auto-loading extensions
23837 @cindex auto-loading extensions
23839 @value{GDBN} provides two mechanisms for automatically loading extensions
23840 when a new object file is read (for example, due to the @code{file}
23841 command, or because the inferior has loaded a shared library):
23842 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
23843 section of modern file formats like ELF.
23846 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
23847 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
23848 * Which flavor to choose?::
23851 The auto-loading feature is useful for supplying application-specific
23852 debugging commands and features.
23854 Auto-loading can be enabled or disabled,
23855 and the list of auto-loaded scripts can be printed.
23856 See the @samp{auto-loading} section of each extension language
23857 for more information.
23858 For @value{GDBN} command files see @ref{Auto-loading sequences}.
23859 For Python files see @ref{Python Auto-loading}.
23861 Note that loading of this script file also requires accordingly configured
23862 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
23864 @node objfile-gdbdotext file
23865 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
23866 @cindex @file{@var{objfile}-gdb.gdb}
23867 @cindex @file{@var{objfile}-gdb.py}
23868 @cindex @file{@var{objfile}-gdb.scm}
23870 When a new object file is read, @value{GDBN} looks for a file named
23871 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
23872 where @var{objfile} is the object file's name and
23873 where @var{ext} is the file extension for the extension language:
23876 @item @file{@var{objfile}-gdb.gdb}
23877 GDB's own command language
23878 @item @file{@var{objfile}-gdb.py}
23880 @item @file{@var{objfile}-gdb.scm}
23884 @var{script-name} is formed by ensuring that the file name of @var{objfile}
23885 is absolute, following all symlinks, and resolving @code{.} and @code{..}
23886 components, and appending the @file{-gdb.@var{ext}} suffix.
23887 If this file exists and is readable, @value{GDBN} will evaluate it as a
23888 script in the specified extension language.
23890 If this file does not exist, then @value{GDBN} will look for
23891 @var{script-name} file in all of the directories as specified below.
23893 Note that loading of these files requires an accordingly configured
23894 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
23896 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
23897 scripts normally according to its @file{.exe} filename. But if no scripts are
23898 found @value{GDBN} also tries script filenames matching the object file without
23899 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
23900 is attempted on any platform. This makes the script filenames compatible
23901 between Unix and MS-Windows hosts.
23904 @anchor{set auto-load scripts-directory}
23905 @kindex set auto-load scripts-directory
23906 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
23907 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
23908 may be delimited by the host platform path separator in use
23909 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
23911 Each entry here needs to be covered also by the security setting
23912 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
23914 @anchor{with-auto-load-dir}
23915 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
23916 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
23917 configuration option @option{--with-auto-load-dir}.
23919 Any reference to @file{$debugdir} will get replaced by
23920 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
23921 reference to @file{$datadir} will get replaced by @var{data-directory} which is
23922 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
23923 @file{$datadir} must be placed as a directory component --- either alone or
23924 delimited by @file{/} or @file{\} directory separators, depending on the host
23927 The list of directories uses path separator (@samp{:} on GNU and Unix
23928 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
23929 to the @env{PATH} environment variable.
23931 @anchor{show auto-load scripts-directory}
23932 @kindex show auto-load scripts-directory
23933 @item show auto-load scripts-directory
23934 Show @value{GDBN} auto-loaded scripts location.
23936 @anchor{add-auto-load-scripts-directory}
23937 @kindex add-auto-load-scripts-directory
23938 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
23939 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
23940 Multiple entries may be delimited by the host platform path separator in use.
23943 @value{GDBN} does not track which files it has already auto-loaded this way.
23944 @value{GDBN} will load the associated script every time the corresponding
23945 @var{objfile} is opened.
23946 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
23947 is evaluated more than once.
23949 @node dotdebug_gdb_scripts section
23950 @subsection The @code{.debug_gdb_scripts} section
23951 @cindex @code{.debug_gdb_scripts} section
23953 For systems using file formats like ELF and COFF,
23954 when @value{GDBN} loads a new object file
23955 it will look for a special section named @code{.debug_gdb_scripts}.
23956 If this section exists, its contents is a list of NUL-terminated names
23957 of scripts to load. Each entry begins with a non-NULL prefix byte that
23958 specifies the kind of entry, typically the extension language.
23960 @value{GDBN} will look for each specified script file first in the
23961 current directory and then along the source search path
23962 (@pxref{Source Path, ,Specifying Source Directories}),
23963 except that @file{$cdir} is not searched, since the compilation
23964 directory is not relevant to scripts.
23966 Entries can be placed in section @code{.debug_gdb_scripts} with,
23967 for example, this GCC macro for Python scripts.
23970 /* Note: The "MS" section flags are to remove duplicates. */
23971 #define DEFINE_GDB_PY_SCRIPT(script_name) \
23973 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
23974 .byte 1 /* Python */\n\
23975 .asciz \"" script_name "\"\n\
23981 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
23982 Then one can reference the macro in a header or source file like this:
23985 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
23988 The script name may include directories if desired.
23990 Note that loading of this script file also requires accordingly configured
23991 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
23993 If the macro invocation is put in a header, any application or library
23994 using this header will get a reference to the specified script,
23995 and with the use of @code{"MS"} attributes on the section, the linker
23996 will remove duplicates.
23998 @node Which flavor to choose?
23999 @subsection Which flavor to choose?
24001 Given the multiple ways of auto-loading extensions, it might not always
24002 be clear which one to choose. This section provides some guidance.
24005 Benefits of the @file{-gdb.@var{ext}} way:
24009 Can be used with file formats that don't support multiple sections.
24012 Ease of finding scripts for public libraries.
24014 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
24015 in the source search path.
24016 For publicly installed libraries, e.g., @file{libstdc++}, there typically
24017 isn't a source directory in which to find the script.
24020 Doesn't require source code additions.
24024 Benefits of the @code{.debug_gdb_scripts} way:
24028 Works with static linking.
24030 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
24031 trigger their loading. When an application is statically linked the only
24032 objfile available is the executable, and it is cumbersome to attach all the
24033 scripts from all the input libraries to the executable's
24034 @file{-gdb.@var{ext}} script.
24037 Works with classes that are entirely inlined.
24039 Some classes can be entirely inlined, and thus there may not be an associated
24040 shared library to attach a @file{-gdb.@var{ext}} script to.
24043 Scripts needn't be copied out of the source tree.
24045 In some circumstances, apps can be built out of large collections of internal
24046 libraries, and the build infrastructure necessary to install the
24047 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
24048 cumbersome. It may be easier to specify the scripts in the
24049 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
24050 top of the source tree to the source search path.
24053 @node Multiple Extension Languages
24054 @section Multiple Extension Languages
24056 The Guile and Python extension languages do not share any state,
24057 and generally do not interfere with each other.
24058 There are some things to be aware of, however.
24060 @subsection Python comes first
24062 Python was @value{GDBN}'s first extension language, and to avoid breaking
24063 existing behaviour Python comes first. This is generally solved by the
24064 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
24065 extension languages, and when it makes a call to an extension language,
24066 (say to pretty-print a value), it tries each in turn until an extension
24067 language indicates it has performed the request (e.g., has returned the
24068 pretty-printed form of a value).
24069 This extends to errors while performing such requests: If an error happens
24070 while, for example, trying to pretty-print an object then the error is
24071 reported and any following extension languages are not tried.
24074 @section Creating new spellings of existing commands
24075 @cindex aliases for commands
24077 It is often useful to define alternate spellings of existing commands.
24078 For example, if a new @value{GDBN} command defined in Python has
24079 a long name to type, it is handy to have an abbreviated version of it
24080 that involves less typing.
24082 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
24083 of the @samp{step} command even though it is otherwise an ambiguous
24084 abbreviation of other commands like @samp{set} and @samp{show}.
24086 Aliases are also used to provide shortened or more common versions
24087 of multi-word commands. For example, @value{GDBN} provides the
24088 @samp{tty} alias of the @samp{set inferior-tty} command.
24090 You can define a new alias with the @samp{alias} command.
24095 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
24099 @var{ALIAS} specifies the name of the new alias.
24100 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
24103 @var{COMMAND} specifies the name of an existing command
24104 that is being aliased.
24106 The @samp{-a} option specifies that the new alias is an abbreviation
24107 of the command. Abbreviations are not shown in command
24108 lists displayed by the @samp{help} command.
24110 The @samp{--} option specifies the end of options,
24111 and is useful when @var{ALIAS} begins with a dash.
24113 Here is a simple example showing how to make an abbreviation
24114 of a command so that there is less to type.
24115 Suppose you were tired of typing @samp{disas}, the current
24116 shortest unambiguous abbreviation of the @samp{disassemble} command
24117 and you wanted an even shorter version named @samp{di}.
24118 The following will accomplish this.
24121 (gdb) alias -a di = disas
24124 Note that aliases are different from user-defined commands.
24125 With a user-defined command, you also need to write documentation
24126 for it with the @samp{document} command.
24127 An alias automatically picks up the documentation of the existing command.
24129 Here is an example where we make @samp{elms} an abbreviation of
24130 @samp{elements} in the @samp{set print elements} command.
24131 This is to show that you can make an abbreviation of any part
24135 (gdb) alias -a set print elms = set print elements
24136 (gdb) alias -a show print elms = show print elements
24137 (gdb) set p elms 20
24139 Limit on string chars or array elements to print is 200.
24142 Note that if you are defining an alias of a @samp{set} command,
24143 and you want to have an alias for the corresponding @samp{show}
24144 command, then you need to define the latter separately.
24146 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
24147 @var{ALIAS}, just as they are normally.
24150 (gdb) alias -a set pr elms = set p ele
24153 Finally, here is an example showing the creation of a one word
24154 alias for a more complex command.
24155 This creates alias @samp{spe} of the command @samp{set print elements}.
24158 (gdb) alias spe = set print elements
24163 @chapter Command Interpreters
24164 @cindex command interpreters
24166 @value{GDBN} supports multiple command interpreters, and some command
24167 infrastructure to allow users or user interface writers to switch
24168 between interpreters or run commands in other interpreters.
24170 @value{GDBN} currently supports two command interpreters, the console
24171 interpreter (sometimes called the command-line interpreter or @sc{cli})
24172 and the machine interface interpreter (or @sc{gdb/mi}). This manual
24173 describes both of these interfaces in great detail.
24175 By default, @value{GDBN} will start with the console interpreter.
24176 However, the user may choose to start @value{GDBN} with another
24177 interpreter by specifying the @option{-i} or @option{--interpreter}
24178 startup options. Defined interpreters include:
24182 @cindex console interpreter
24183 The traditional console or command-line interpreter. This is the most often
24184 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
24185 @value{GDBN} will use this interpreter.
24188 @cindex mi interpreter
24189 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
24190 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
24191 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
24195 @cindex mi2 interpreter
24196 The current @sc{gdb/mi} interface.
24199 @cindex mi1 interpreter
24200 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
24204 @cindex invoke another interpreter
24205 The interpreter being used by @value{GDBN} may not be dynamically
24206 switched at runtime. Although possible, this could lead to a very
24207 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
24208 enters the command "interpreter-set console" in a console view,
24209 @value{GDBN} would switch to using the console interpreter, rendering
24210 the IDE inoperable!
24212 @kindex interpreter-exec
24213 Although you may only choose a single interpreter at startup, you may execute
24214 commands in any interpreter from the current interpreter using the appropriate
24215 command. If you are running the console interpreter, simply use the
24216 @code{interpreter-exec} command:
24219 interpreter-exec mi "-data-list-register-names"
24222 @sc{gdb/mi} has a similar command, although it is only available in versions of
24223 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
24226 @chapter @value{GDBN} Text User Interface
24228 @cindex Text User Interface
24231 * TUI Overview:: TUI overview
24232 * TUI Keys:: TUI key bindings
24233 * TUI Single Key Mode:: TUI single key mode
24234 * TUI Commands:: TUI-specific commands
24235 * TUI Configuration:: TUI configuration variables
24238 The @value{GDBN} Text User Interface (TUI) is a terminal
24239 interface which uses the @code{curses} library to show the source
24240 file, the assembly output, the program registers and @value{GDBN}
24241 commands in separate text windows. The TUI mode is supported only
24242 on platforms where a suitable version of the @code{curses} library
24245 The TUI mode is enabled by default when you invoke @value{GDBN} as
24246 @samp{@value{GDBP} -tui}.
24247 You can also switch in and out of TUI mode while @value{GDBN} runs by
24248 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
24249 @xref{TUI Keys, ,TUI Key Bindings}.
24252 @section TUI Overview
24254 In TUI mode, @value{GDBN} can display several text windows:
24258 This window is the @value{GDBN} command window with the @value{GDBN}
24259 prompt and the @value{GDBN} output. The @value{GDBN} input is still
24260 managed using readline.
24263 The source window shows the source file of the program. The current
24264 line and active breakpoints are displayed in this window.
24267 The assembly window shows the disassembly output of the program.
24270 This window shows the processor registers. Registers are highlighted
24271 when their values change.
24274 The source and assembly windows show the current program position
24275 by highlighting the current line and marking it with a @samp{>} marker.
24276 Breakpoints are indicated with two markers. The first marker
24277 indicates the breakpoint type:
24281 Breakpoint which was hit at least once.
24284 Breakpoint which was never hit.
24287 Hardware breakpoint which was hit at least once.
24290 Hardware breakpoint which was never hit.
24293 The second marker indicates whether the breakpoint is enabled or not:
24297 Breakpoint is enabled.
24300 Breakpoint is disabled.
24303 The source, assembly and register windows are updated when the current
24304 thread changes, when the frame changes, or when the program counter
24307 These windows are not all visible at the same time. The command
24308 window is always visible. The others can be arranged in several
24319 source and assembly,
24322 source and registers, or
24325 assembly and registers.
24328 A status line above the command window shows the following information:
24332 Indicates the current @value{GDBN} target.
24333 (@pxref{Targets, ,Specifying a Debugging Target}).
24336 Gives the current process or thread number.
24337 When no process is being debugged, this field is set to @code{No process}.
24340 Gives the current function name for the selected frame.
24341 The name is demangled if demangling is turned on (@pxref{Print Settings}).
24342 When there is no symbol corresponding to the current program counter,
24343 the string @code{??} is displayed.
24346 Indicates the current line number for the selected frame.
24347 When the current line number is not known, the string @code{??} is displayed.
24350 Indicates the current program counter address.
24354 @section TUI Key Bindings
24355 @cindex TUI key bindings
24357 The TUI installs several key bindings in the readline keymaps
24358 @ifset SYSTEM_READLINE
24359 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
24361 @ifclear SYSTEM_READLINE
24362 (@pxref{Command Line Editing}).
24364 The following key bindings are installed for both TUI mode and the
24365 @value{GDBN} standard mode.
24374 Enter or leave the TUI mode. When leaving the TUI mode,
24375 the curses window management stops and @value{GDBN} operates using
24376 its standard mode, writing on the terminal directly. When reentering
24377 the TUI mode, control is given back to the curses windows.
24378 The screen is then refreshed.
24382 Use a TUI layout with only one window. The layout will
24383 either be @samp{source} or @samp{assembly}. When the TUI mode
24384 is not active, it will switch to the TUI mode.
24386 Think of this key binding as the Emacs @kbd{C-x 1} binding.
24390 Use a TUI layout with at least two windows. When the current
24391 layout already has two windows, the next layout with two windows is used.
24392 When a new layout is chosen, one window will always be common to the
24393 previous layout and the new one.
24395 Think of it as the Emacs @kbd{C-x 2} binding.
24399 Change the active window. The TUI associates several key bindings
24400 (like scrolling and arrow keys) with the active window. This command
24401 gives the focus to the next TUI window.
24403 Think of it as the Emacs @kbd{C-x o} binding.
24407 Switch in and out of the TUI SingleKey mode that binds single
24408 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
24411 The following key bindings only work in the TUI mode:
24416 Scroll the active window one page up.
24420 Scroll the active window one page down.
24424 Scroll the active window one line up.
24428 Scroll the active window one line down.
24432 Scroll the active window one column left.
24436 Scroll the active window one column right.
24440 Refresh the screen.
24443 Because the arrow keys scroll the active window in the TUI mode, they
24444 are not available for their normal use by readline unless the command
24445 window has the focus. When another window is active, you must use
24446 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
24447 and @kbd{C-f} to control the command window.
24449 @node TUI Single Key Mode
24450 @section TUI Single Key Mode
24451 @cindex TUI single key mode
24453 The TUI also provides a @dfn{SingleKey} mode, which binds several
24454 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
24455 switch into this mode, where the following key bindings are used:
24458 @kindex c @r{(SingleKey TUI key)}
24462 @kindex d @r{(SingleKey TUI key)}
24466 @kindex f @r{(SingleKey TUI key)}
24470 @kindex n @r{(SingleKey TUI key)}
24474 @kindex q @r{(SingleKey TUI key)}
24476 exit the SingleKey mode.
24478 @kindex r @r{(SingleKey TUI key)}
24482 @kindex s @r{(SingleKey TUI key)}
24486 @kindex u @r{(SingleKey TUI key)}
24490 @kindex v @r{(SingleKey TUI key)}
24494 @kindex w @r{(SingleKey TUI key)}
24499 Other keys temporarily switch to the @value{GDBN} command prompt.
24500 The key that was pressed is inserted in the editing buffer so that
24501 it is possible to type most @value{GDBN} commands without interaction
24502 with the TUI SingleKey mode. Once the command is entered the TUI
24503 SingleKey mode is restored. The only way to permanently leave
24504 this mode is by typing @kbd{q} or @kbd{C-x s}.
24508 @section TUI-specific Commands
24509 @cindex TUI commands
24511 The TUI has specific commands to control the text windows.
24512 These commands are always available, even when @value{GDBN} is not in
24513 the TUI mode. When @value{GDBN} is in the standard mode, most
24514 of these commands will automatically switch to the TUI mode.
24516 Note that if @value{GDBN}'s @code{stdout} is not connected to a
24517 terminal, or @value{GDBN} has been started with the machine interface
24518 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
24519 these commands will fail with an error, because it would not be
24520 possible or desirable to enable curses window management.
24525 List and give the size of all displayed windows.
24529 Display the next layout.
24532 Display the previous layout.
24535 Display the source window only.
24538 Display the assembly window only.
24541 Display the source and assembly window.
24544 Display the register window together with the source or assembly window.
24548 Make the next window active for scrolling.
24551 Make the previous window active for scrolling.
24554 Make the source window active for scrolling.
24557 Make the assembly window active for scrolling.
24560 Make the register window active for scrolling.
24563 Make the command window active for scrolling.
24567 Refresh the screen. This is similar to typing @kbd{C-L}.
24569 @item tui reg float
24571 Show the floating point registers in the register window.
24573 @item tui reg general
24574 Show the general registers in the register window.
24577 Show the next register group. The list of register groups as well as
24578 their order is target specific. The predefined register groups are the
24579 following: @code{general}, @code{float}, @code{system}, @code{vector},
24580 @code{all}, @code{save}, @code{restore}.
24582 @item tui reg system
24583 Show the system registers in the register window.
24587 Update the source window and the current execution point.
24589 @item winheight @var{name} +@var{count}
24590 @itemx winheight @var{name} -@var{count}
24592 Change the height of the window @var{name} by @var{count}
24593 lines. Positive counts increase the height, while negative counts
24596 @item tabset @var{nchars}
24598 Set the width of tab stops to be @var{nchars} characters.
24601 @node TUI Configuration
24602 @section TUI Configuration Variables
24603 @cindex TUI configuration variables
24605 Several configuration variables control the appearance of TUI windows.
24608 @item set tui border-kind @var{kind}
24609 @kindex set tui border-kind
24610 Select the border appearance for the source, assembly and register windows.
24611 The possible values are the following:
24614 Use a space character to draw the border.
24617 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
24620 Use the Alternate Character Set to draw the border. The border is
24621 drawn using character line graphics if the terminal supports them.
24624 @item set tui border-mode @var{mode}
24625 @kindex set tui border-mode
24626 @itemx set tui active-border-mode @var{mode}
24627 @kindex set tui active-border-mode
24628 Select the display attributes for the borders of the inactive windows
24629 or the active window. The @var{mode} can be one of the following:
24632 Use normal attributes to display the border.
24638 Use reverse video mode.
24641 Use half bright mode.
24643 @item half-standout
24644 Use half bright and standout mode.
24647 Use extra bright or bold mode.
24649 @item bold-standout
24650 Use extra bright or bold and standout mode.
24655 @chapter Using @value{GDBN} under @sc{gnu} Emacs
24658 @cindex @sc{gnu} Emacs
24659 A special interface allows you to use @sc{gnu} Emacs to view (and
24660 edit) the source files for the program you are debugging with
24663 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
24664 executable file you want to debug as an argument. This command starts
24665 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
24666 created Emacs buffer.
24667 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
24669 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
24674 All ``terminal'' input and output goes through an Emacs buffer, called
24677 This applies both to @value{GDBN} commands and their output, and to the input
24678 and output done by the program you are debugging.
24680 This is useful because it means that you can copy the text of previous
24681 commands and input them again; you can even use parts of the output
24684 All the facilities of Emacs' Shell mode are available for interacting
24685 with your program. In particular, you can send signals the usual
24686 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
24690 @value{GDBN} displays source code through Emacs.
24692 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
24693 source file for that frame and puts an arrow (@samp{=>}) at the
24694 left margin of the current line. Emacs uses a separate buffer for
24695 source display, and splits the screen to show both your @value{GDBN} session
24698 Explicit @value{GDBN} @code{list} or search commands still produce output as
24699 usual, but you probably have no reason to use them from Emacs.
24702 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
24703 a graphical mode, enabled by default, which provides further buffers
24704 that can control the execution and describe the state of your program.
24705 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
24707 If you specify an absolute file name when prompted for the @kbd{M-x
24708 gdb} argument, then Emacs sets your current working directory to where
24709 your program resides. If you only specify the file name, then Emacs
24710 sets your current working directory to the directory associated
24711 with the previous buffer. In this case, @value{GDBN} may find your
24712 program by searching your environment's @code{PATH} variable, but on
24713 some operating systems it might not find the source. So, although the
24714 @value{GDBN} input and output session proceeds normally, the auxiliary
24715 buffer does not display the current source and line of execution.
24717 The initial working directory of @value{GDBN} is printed on the top
24718 line of the GUD buffer and this serves as a default for the commands
24719 that specify files for @value{GDBN} to operate on. @xref{Files,
24720 ,Commands to Specify Files}.
24722 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
24723 need to call @value{GDBN} by a different name (for example, if you
24724 keep several configurations around, with different names) you can
24725 customize the Emacs variable @code{gud-gdb-command-name} to run the
24728 In the GUD buffer, you can use these special Emacs commands in
24729 addition to the standard Shell mode commands:
24733 Describe the features of Emacs' GUD Mode.
24736 Execute to another source line, like the @value{GDBN} @code{step} command; also
24737 update the display window to show the current file and location.
24740 Execute to next source line in this function, skipping all function
24741 calls, like the @value{GDBN} @code{next} command. Then update the display window
24742 to show the current file and location.
24745 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
24746 display window accordingly.
24749 Execute until exit from the selected stack frame, like the @value{GDBN}
24750 @code{finish} command.
24753 Continue execution of your program, like the @value{GDBN} @code{continue}
24757 Go up the number of frames indicated by the numeric argument
24758 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
24759 like the @value{GDBN} @code{up} command.
24762 Go down the number of frames indicated by the numeric argument, like the
24763 @value{GDBN} @code{down} command.
24766 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
24767 tells @value{GDBN} to set a breakpoint on the source line point is on.
24769 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
24770 separate frame which shows a backtrace when the GUD buffer is current.
24771 Move point to any frame in the stack and type @key{RET} to make it
24772 become the current frame and display the associated source in the
24773 source buffer. Alternatively, click @kbd{Mouse-2} to make the
24774 selected frame become the current one. In graphical mode, the
24775 speedbar displays watch expressions.
24777 If you accidentally delete the source-display buffer, an easy way to get
24778 it back is to type the command @code{f} in the @value{GDBN} buffer, to
24779 request a frame display; when you run under Emacs, this recreates
24780 the source buffer if necessary to show you the context of the current
24783 The source files displayed in Emacs are in ordinary Emacs buffers
24784 which are visiting the source files in the usual way. You can edit
24785 the files with these buffers if you wish; but keep in mind that @value{GDBN}
24786 communicates with Emacs in terms of line numbers. If you add or
24787 delete lines from the text, the line numbers that @value{GDBN} knows cease
24788 to correspond properly with the code.
24790 A more detailed description of Emacs' interaction with @value{GDBN} is
24791 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
24795 @chapter The @sc{gdb/mi} Interface
24797 @unnumberedsec Function and Purpose
24799 @cindex @sc{gdb/mi}, its purpose
24800 @sc{gdb/mi} is a line based machine oriented text interface to
24801 @value{GDBN} and is activated by specifying using the
24802 @option{--interpreter} command line option (@pxref{Mode Options}). It
24803 is specifically intended to support the development of systems which
24804 use the debugger as just one small component of a larger system.
24806 This chapter is a specification of the @sc{gdb/mi} interface. It is written
24807 in the form of a reference manual.
24809 Note that @sc{gdb/mi} is still under construction, so some of the
24810 features described below are incomplete and subject to change
24811 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
24813 @unnumberedsec Notation and Terminology
24815 @cindex notational conventions, for @sc{gdb/mi}
24816 This chapter uses the following notation:
24820 @code{|} separates two alternatives.
24823 @code{[ @var{something} ]} indicates that @var{something} is optional:
24824 it may or may not be given.
24827 @code{( @var{group} )*} means that @var{group} inside the parentheses
24828 may repeat zero or more times.
24831 @code{( @var{group} )+} means that @var{group} inside the parentheses
24832 may repeat one or more times.
24835 @code{"@var{string}"} means a literal @var{string}.
24839 @heading Dependencies
24843 * GDB/MI General Design::
24844 * GDB/MI Command Syntax::
24845 * GDB/MI Compatibility with CLI::
24846 * GDB/MI Development and Front Ends::
24847 * GDB/MI Output Records::
24848 * GDB/MI Simple Examples::
24849 * GDB/MI Command Description Format::
24850 * GDB/MI Breakpoint Commands::
24851 * GDB/MI Catchpoint Commands::
24852 * GDB/MI Program Context::
24853 * GDB/MI Thread Commands::
24854 * GDB/MI Ada Tasking Commands::
24855 * GDB/MI Program Execution::
24856 * GDB/MI Stack Manipulation::
24857 * GDB/MI Variable Objects::
24858 * GDB/MI Data Manipulation::
24859 * GDB/MI Tracepoint Commands::
24860 * GDB/MI Symbol Query::
24861 * GDB/MI File Commands::
24863 * GDB/MI Kod Commands::
24864 * GDB/MI Memory Overlay Commands::
24865 * GDB/MI Signal Handling Commands::
24867 * GDB/MI Target Manipulation::
24868 * GDB/MI File Transfer Commands::
24869 * GDB/MI Ada Exceptions Commands::
24870 * GDB/MI Support Commands::
24871 * GDB/MI Miscellaneous Commands::
24874 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24875 @node GDB/MI General Design
24876 @section @sc{gdb/mi} General Design
24877 @cindex GDB/MI General Design
24879 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
24880 parts---commands sent to @value{GDBN}, responses to those commands
24881 and notifications. Each command results in exactly one response,
24882 indicating either successful completion of the command, or an error.
24883 For the commands that do not resume the target, the response contains the
24884 requested information. For the commands that resume the target, the
24885 response only indicates whether the target was successfully resumed.
24886 Notifications is the mechanism for reporting changes in the state of the
24887 target, or in @value{GDBN} state, that cannot conveniently be associated with
24888 a command and reported as part of that command response.
24890 The important examples of notifications are:
24894 Exec notifications. These are used to report changes in
24895 target state---when a target is resumed, or stopped. It would not
24896 be feasible to include this information in response of resuming
24897 commands, because one resume commands can result in multiple events in
24898 different threads. Also, quite some time may pass before any event
24899 happens in the target, while a frontend needs to know whether the resuming
24900 command itself was successfully executed.
24903 Console output, and status notifications. Console output
24904 notifications are used to report output of CLI commands, as well as
24905 diagnostics for other commands. Status notifications are used to
24906 report the progress of a long-running operation. Naturally, including
24907 this information in command response would mean no output is produced
24908 until the command is finished, which is undesirable.
24911 General notifications. Commands may have various side effects on
24912 the @value{GDBN} or target state beyond their official purpose. For example,
24913 a command may change the selected thread. Although such changes can
24914 be included in command response, using notification allows for more
24915 orthogonal frontend design.
24919 There's no guarantee that whenever an MI command reports an error,
24920 @value{GDBN} or the target are in any specific state, and especially,
24921 the state is not reverted to the state before the MI command was
24922 processed. Therefore, whenever an MI command results in an error,
24923 we recommend that the frontend refreshes all the information shown in
24924 the user interface.
24928 * Context management::
24929 * Asynchronous and non-stop modes::
24933 @node Context management
24934 @subsection Context management
24936 @subsubsection Threads and Frames
24938 In most cases when @value{GDBN} accesses the target, this access is
24939 done in context of a specific thread and frame (@pxref{Frames}).
24940 Often, even when accessing global data, the target requires that a thread
24941 be specified. The CLI interface maintains the selected thread and frame,
24942 and supplies them to target on each command. This is convenient,
24943 because a command line user would not want to specify that information
24944 explicitly on each command, and because user interacts with
24945 @value{GDBN} via a single terminal, so no confusion is possible as
24946 to what thread and frame are the current ones.
24948 In the case of MI, the concept of selected thread and frame is less
24949 useful. First, a frontend can easily remember this information
24950 itself. Second, a graphical frontend can have more than one window,
24951 each one used for debugging a different thread, and the frontend might
24952 want to access additional threads for internal purposes. This
24953 increases the risk that by relying on implicitly selected thread, the
24954 frontend may be operating on a wrong one. Therefore, each MI command
24955 should explicitly specify which thread and frame to operate on. To
24956 make it possible, each MI command accepts the @samp{--thread} and
24957 @samp{--frame} options, the value to each is @value{GDBN} identifier
24958 for thread and frame to operate on.
24960 Usually, each top-level window in a frontend allows the user to select
24961 a thread and a frame, and remembers the user selection for further
24962 operations. However, in some cases @value{GDBN} may suggest that the
24963 current thread be changed. For example, when stopping on a breakpoint
24964 it is reasonable to switch to the thread where breakpoint is hit. For
24965 another example, if the user issues the CLI @samp{thread} command via
24966 the frontend, it is desirable to change the frontend's selected thread to the
24967 one specified by user. @value{GDBN} communicates the suggestion to
24968 change current thread using the @samp{=thread-selected} notification.
24969 No such notification is available for the selected frame at the moment.
24971 Note that historically, MI shares the selected thread with CLI, so
24972 frontends used the @code{-thread-select} to execute commands in the
24973 right context. However, getting this to work right is cumbersome. The
24974 simplest way is for frontend to emit @code{-thread-select} command
24975 before every command. This doubles the number of commands that need
24976 to be sent. The alternative approach is to suppress @code{-thread-select}
24977 if the selected thread in @value{GDBN} is supposed to be identical to the
24978 thread the frontend wants to operate on. However, getting this
24979 optimization right can be tricky. In particular, if the frontend
24980 sends several commands to @value{GDBN}, and one of the commands changes the
24981 selected thread, then the behaviour of subsequent commands will
24982 change. So, a frontend should either wait for response from such
24983 problematic commands, or explicitly add @code{-thread-select} for
24984 all subsequent commands. No frontend is known to do this exactly
24985 right, so it is suggested to just always pass the @samp{--thread} and
24986 @samp{--frame} options.
24988 @subsubsection Language
24990 The execution of several commands depends on which language is selected.
24991 By default, the current language (@pxref{show language}) is used.
24992 But for commands known to be language-sensitive, it is recommended
24993 to use the @samp{--language} option. This option takes one argument,
24994 which is the name of the language to use while executing the command.
24998 -data-evaluate-expression --language c "sizeof (void*)"
25003 The valid language names are the same names accepted by the
25004 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
25005 @samp{local} or @samp{unknown}.
25007 @node Asynchronous and non-stop modes
25008 @subsection Asynchronous command execution and non-stop mode
25010 On some targets, @value{GDBN} is capable of processing MI commands
25011 even while the target is running. This is called @dfn{asynchronous
25012 command execution} (@pxref{Background Execution}). The frontend may
25013 specify a preferrence for asynchronous execution using the
25014 @code{-gdb-set mi-async 1} command, which should be emitted before
25015 either running the executable or attaching to the target. After the
25016 frontend has started the executable or attached to the target, it can
25017 find if asynchronous execution is enabled using the
25018 @code{-list-target-features} command.
25021 @item -gdb-set mi-async on
25022 @item -gdb-set mi-async off
25023 Set whether MI is in asynchronous mode.
25025 When @code{off}, which is the default, MI execution commands (e.g.,
25026 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
25027 for the program to stop before processing further commands.
25029 When @code{on}, MI execution commands are background execution
25030 commands (e.g., @code{-exec-continue} becomes the equivalent of the
25031 @code{c&} CLI command), and so @value{GDBN} is capable of processing
25032 MI commands even while the target is running.
25034 @item -gdb-show mi-async
25035 Show whether MI asynchronous mode is enabled.
25038 Note: In @value{GDBN} version 7.7 and earlier, this option was called
25039 @code{target-async} instead of @code{mi-async}, and it had the effect
25040 of both putting MI in asynchronous mode and making CLI background
25041 commands possible. CLI background commands are now always possible
25042 ``out of the box'' if the target supports them. The old spelling is
25043 kept as a deprecated alias for backwards compatibility.
25045 Even if @value{GDBN} can accept a command while target is running,
25046 many commands that access the target do not work when the target is
25047 running. Therefore, asynchronous command execution is most useful
25048 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
25049 it is possible to examine the state of one thread, while other threads
25052 When a given thread is running, MI commands that try to access the
25053 target in the context of that thread may not work, or may work only on
25054 some targets. In particular, commands that try to operate on thread's
25055 stack will not work, on any target. Commands that read memory, or
25056 modify breakpoints, may work or not work, depending on the target. Note
25057 that even commands that operate on global state, such as @code{print},
25058 @code{set}, and breakpoint commands, still access the target in the
25059 context of a specific thread, so frontend should try to find a
25060 stopped thread and perform the operation on that thread (using the
25061 @samp{--thread} option).
25063 Which commands will work in the context of a running thread is
25064 highly target dependent. However, the two commands
25065 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
25066 to find the state of a thread, will always work.
25068 @node Thread groups
25069 @subsection Thread groups
25070 @value{GDBN} may be used to debug several processes at the same time.
25071 On some platfroms, @value{GDBN} may support debugging of several
25072 hardware systems, each one having several cores with several different
25073 processes running on each core. This section describes the MI
25074 mechanism to support such debugging scenarios.
25076 The key observation is that regardless of the structure of the
25077 target, MI can have a global list of threads, because most commands that
25078 accept the @samp{--thread} option do not need to know what process that
25079 thread belongs to. Therefore, it is not necessary to introduce
25080 neither additional @samp{--process} option, nor an notion of the
25081 current process in the MI interface. The only strictly new feature
25082 that is required is the ability to find how the threads are grouped
25085 To allow the user to discover such grouping, and to support arbitrary
25086 hierarchy of machines/cores/processes, MI introduces the concept of a
25087 @dfn{thread group}. Thread group is a collection of threads and other
25088 thread groups. A thread group always has a string identifier, a type,
25089 and may have additional attributes specific to the type. A new
25090 command, @code{-list-thread-groups}, returns the list of top-level
25091 thread groups, which correspond to processes that @value{GDBN} is
25092 debugging at the moment. By passing an identifier of a thread group
25093 to the @code{-list-thread-groups} command, it is possible to obtain
25094 the members of specific thread group.
25096 To allow the user to easily discover processes, and other objects, he
25097 wishes to debug, a concept of @dfn{available thread group} is
25098 introduced. Available thread group is an thread group that
25099 @value{GDBN} is not debugging, but that can be attached to, using the
25100 @code{-target-attach} command. The list of available top-level thread
25101 groups can be obtained using @samp{-list-thread-groups --available}.
25102 In general, the content of a thread group may be only retrieved only
25103 after attaching to that thread group.
25105 Thread groups are related to inferiors (@pxref{Inferiors and
25106 Programs}). Each inferior corresponds to a thread group of a special
25107 type @samp{process}, and some additional operations are permitted on
25108 such thread groups.
25110 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25111 @node GDB/MI Command Syntax
25112 @section @sc{gdb/mi} Command Syntax
25115 * GDB/MI Input Syntax::
25116 * GDB/MI Output Syntax::
25119 @node GDB/MI Input Syntax
25120 @subsection @sc{gdb/mi} Input Syntax
25122 @cindex input syntax for @sc{gdb/mi}
25123 @cindex @sc{gdb/mi}, input syntax
25125 @item @var{command} @expansion{}
25126 @code{@var{cli-command} | @var{mi-command}}
25128 @item @var{cli-command} @expansion{}
25129 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
25130 @var{cli-command} is any existing @value{GDBN} CLI command.
25132 @item @var{mi-command} @expansion{}
25133 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
25134 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
25136 @item @var{token} @expansion{}
25137 "any sequence of digits"
25139 @item @var{option} @expansion{}
25140 @code{"-" @var{parameter} [ " " @var{parameter} ]}
25142 @item @var{parameter} @expansion{}
25143 @code{@var{non-blank-sequence} | @var{c-string}}
25145 @item @var{operation} @expansion{}
25146 @emph{any of the operations described in this chapter}
25148 @item @var{non-blank-sequence} @expansion{}
25149 @emph{anything, provided it doesn't contain special characters such as
25150 "-", @var{nl}, """ and of course " "}
25152 @item @var{c-string} @expansion{}
25153 @code{""" @var{seven-bit-iso-c-string-content} """}
25155 @item @var{nl} @expansion{}
25164 The CLI commands are still handled by the @sc{mi} interpreter; their
25165 output is described below.
25168 The @code{@var{token}}, when present, is passed back when the command
25172 Some @sc{mi} commands accept optional arguments as part of the parameter
25173 list. Each option is identified by a leading @samp{-} (dash) and may be
25174 followed by an optional argument parameter. Options occur first in the
25175 parameter list and can be delimited from normal parameters using
25176 @samp{--} (this is useful when some parameters begin with a dash).
25183 We want easy access to the existing CLI syntax (for debugging).
25186 We want it to be easy to spot a @sc{mi} operation.
25189 @node GDB/MI Output Syntax
25190 @subsection @sc{gdb/mi} Output Syntax
25192 @cindex output syntax of @sc{gdb/mi}
25193 @cindex @sc{gdb/mi}, output syntax
25194 The output from @sc{gdb/mi} consists of zero or more out-of-band records
25195 followed, optionally, by a single result record. This result record
25196 is for the most recent command. The sequence of output records is
25197 terminated by @samp{(gdb)}.
25199 If an input command was prefixed with a @code{@var{token}} then the
25200 corresponding output for that command will also be prefixed by that same
25204 @item @var{output} @expansion{}
25205 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
25207 @item @var{result-record} @expansion{}
25208 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
25210 @item @var{out-of-band-record} @expansion{}
25211 @code{@var{async-record} | @var{stream-record}}
25213 @item @var{async-record} @expansion{}
25214 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
25216 @item @var{exec-async-output} @expansion{}
25217 @code{[ @var{token} ] "*" @var{async-output nl}}
25219 @item @var{status-async-output} @expansion{}
25220 @code{[ @var{token} ] "+" @var{async-output nl}}
25222 @item @var{notify-async-output} @expansion{}
25223 @code{[ @var{token} ] "=" @var{async-output nl}}
25225 @item @var{async-output} @expansion{}
25226 @code{@var{async-class} ( "," @var{result} )*}
25228 @item @var{result-class} @expansion{}
25229 @code{"done" | "running" | "connected" | "error" | "exit"}
25231 @item @var{async-class} @expansion{}
25232 @code{"stopped" | @var{others}} (where @var{others} will be added
25233 depending on the needs---this is still in development).
25235 @item @var{result} @expansion{}
25236 @code{ @var{variable} "=" @var{value}}
25238 @item @var{variable} @expansion{}
25239 @code{ @var{string} }
25241 @item @var{value} @expansion{}
25242 @code{ @var{const} | @var{tuple} | @var{list} }
25244 @item @var{const} @expansion{}
25245 @code{@var{c-string}}
25247 @item @var{tuple} @expansion{}
25248 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
25250 @item @var{list} @expansion{}
25251 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
25252 @var{result} ( "," @var{result} )* "]" }
25254 @item @var{stream-record} @expansion{}
25255 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
25257 @item @var{console-stream-output} @expansion{}
25258 @code{"~" @var{c-string nl}}
25260 @item @var{target-stream-output} @expansion{}
25261 @code{"@@" @var{c-string nl}}
25263 @item @var{log-stream-output} @expansion{}
25264 @code{"&" @var{c-string nl}}
25266 @item @var{nl} @expansion{}
25269 @item @var{token} @expansion{}
25270 @emph{any sequence of digits}.
25278 All output sequences end in a single line containing a period.
25281 The @code{@var{token}} is from the corresponding request. Note that
25282 for all async output, while the token is allowed by the grammar and
25283 may be output by future versions of @value{GDBN} for select async
25284 output messages, it is generally omitted. Frontends should treat
25285 all async output as reporting general changes in the state of the
25286 target and there should be no need to associate async output to any
25290 @cindex status output in @sc{gdb/mi}
25291 @var{status-async-output} contains on-going status information about the
25292 progress of a slow operation. It can be discarded. All status output is
25293 prefixed by @samp{+}.
25296 @cindex async output in @sc{gdb/mi}
25297 @var{exec-async-output} contains asynchronous state change on the target
25298 (stopped, started, disappeared). All async output is prefixed by
25302 @cindex notify output in @sc{gdb/mi}
25303 @var{notify-async-output} contains supplementary information that the
25304 client should handle (e.g., a new breakpoint information). All notify
25305 output is prefixed by @samp{=}.
25308 @cindex console output in @sc{gdb/mi}
25309 @var{console-stream-output} is output that should be displayed as is in the
25310 console. It is the textual response to a CLI command. All the console
25311 output is prefixed by @samp{~}.
25314 @cindex target output in @sc{gdb/mi}
25315 @var{target-stream-output} is the output produced by the target program.
25316 All the target output is prefixed by @samp{@@}.
25319 @cindex log output in @sc{gdb/mi}
25320 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
25321 instance messages that should be displayed as part of an error log. All
25322 the log output is prefixed by @samp{&}.
25325 @cindex list output in @sc{gdb/mi}
25326 New @sc{gdb/mi} commands should only output @var{lists} containing
25332 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
25333 details about the various output records.
25335 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25336 @node GDB/MI Compatibility with CLI
25337 @section @sc{gdb/mi} Compatibility with CLI
25339 @cindex compatibility, @sc{gdb/mi} and CLI
25340 @cindex @sc{gdb/mi}, compatibility with CLI
25342 For the developers convenience CLI commands can be entered directly,
25343 but there may be some unexpected behaviour. For example, commands
25344 that query the user will behave as if the user replied yes, breakpoint
25345 command lists are not executed and some CLI commands, such as
25346 @code{if}, @code{when} and @code{define}, prompt for further input with
25347 @samp{>}, which is not valid MI output.
25349 This feature may be removed at some stage in the future and it is
25350 recommended that front ends use the @code{-interpreter-exec} command
25351 (@pxref{-interpreter-exec}).
25353 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25354 @node GDB/MI Development and Front Ends
25355 @section @sc{gdb/mi} Development and Front Ends
25356 @cindex @sc{gdb/mi} development
25358 The application which takes the MI output and presents the state of the
25359 program being debugged to the user is called a @dfn{front end}.
25361 Although @sc{gdb/mi} is still incomplete, it is currently being used
25362 by a variety of front ends to @value{GDBN}. This makes it difficult
25363 to introduce new functionality without breaking existing usage. This
25364 section tries to minimize the problems by describing how the protocol
25367 Some changes in MI need not break a carefully designed front end, and
25368 for these the MI version will remain unchanged. The following is a
25369 list of changes that may occur within one level, so front ends should
25370 parse MI output in a way that can handle them:
25374 New MI commands may be added.
25377 New fields may be added to the output of any MI command.
25380 The range of values for fields with specified values, e.g.,
25381 @code{in_scope} (@pxref{-var-update}) may be extended.
25383 @c The format of field's content e.g type prefix, may change so parse it
25384 @c at your own risk. Yes, in general?
25386 @c The order of fields may change? Shouldn't really matter but it might
25387 @c resolve inconsistencies.
25390 If the changes are likely to break front ends, the MI version level
25391 will be increased by one. This will allow the front end to parse the
25392 output according to the MI version. Apart from mi0, new versions of
25393 @value{GDBN} will not support old versions of MI and it will be the
25394 responsibility of the front end to work with the new one.
25396 @c Starting with mi3, add a new command -mi-version that prints the MI
25399 The best way to avoid unexpected changes in MI that might break your front
25400 end is to make your project known to @value{GDBN} developers and
25401 follow development on @email{gdb@@sourceware.org} and
25402 @email{gdb-patches@@sourceware.org}.
25403 @cindex mailing lists
25405 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25406 @node GDB/MI Output Records
25407 @section @sc{gdb/mi} Output Records
25410 * GDB/MI Result Records::
25411 * GDB/MI Stream Records::
25412 * GDB/MI Async Records::
25413 * GDB/MI Breakpoint Information::
25414 * GDB/MI Frame Information::
25415 * GDB/MI Thread Information::
25416 * GDB/MI Ada Exception Information::
25419 @node GDB/MI Result Records
25420 @subsection @sc{gdb/mi} Result Records
25422 @cindex result records in @sc{gdb/mi}
25423 @cindex @sc{gdb/mi}, result records
25424 In addition to a number of out-of-band notifications, the response to a
25425 @sc{gdb/mi} command includes one of the following result indications:
25429 @item "^done" [ "," @var{results} ]
25430 The synchronous operation was successful, @code{@var{results}} are the return
25435 This result record is equivalent to @samp{^done}. Historically, it
25436 was output instead of @samp{^done} if the command has resumed the
25437 target. This behaviour is maintained for backward compatibility, but
25438 all frontends should treat @samp{^done} and @samp{^running}
25439 identically and rely on the @samp{*running} output record to determine
25440 which threads are resumed.
25444 @value{GDBN} has connected to a remote target.
25446 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
25448 The operation failed. The @code{msg=@var{c-string}} variable contains
25449 the corresponding error message.
25451 If present, the @code{code=@var{c-string}} variable provides an error
25452 code on which consumers can rely on to detect the corresponding
25453 error condition. At present, only one error code is defined:
25456 @item "undefined-command"
25457 Indicates that the command causing the error does not exist.
25462 @value{GDBN} has terminated.
25466 @node GDB/MI Stream Records
25467 @subsection @sc{gdb/mi} Stream Records
25469 @cindex @sc{gdb/mi}, stream records
25470 @cindex stream records in @sc{gdb/mi}
25471 @value{GDBN} internally maintains a number of output streams: the console, the
25472 target, and the log. The output intended for each of these streams is
25473 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
25475 Each stream record begins with a unique @dfn{prefix character} which
25476 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
25477 Syntax}). In addition to the prefix, each stream record contains a
25478 @code{@var{string-output}}. This is either raw text (with an implicit new
25479 line) or a quoted C string (which does not contain an implicit newline).
25482 @item "~" @var{string-output}
25483 The console output stream contains text that should be displayed in the
25484 CLI console window. It contains the textual responses to CLI commands.
25486 @item "@@" @var{string-output}
25487 The target output stream contains any textual output from the running
25488 target. This is only present when GDB's event loop is truly
25489 asynchronous, which is currently only the case for remote targets.
25491 @item "&" @var{string-output}
25492 The log stream contains debugging messages being produced by @value{GDBN}'s
25496 @node GDB/MI Async Records
25497 @subsection @sc{gdb/mi} Async Records
25499 @cindex async records in @sc{gdb/mi}
25500 @cindex @sc{gdb/mi}, async records
25501 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
25502 additional changes that have occurred. Those changes can either be a
25503 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
25504 target activity (e.g., target stopped).
25506 The following is the list of possible async records:
25510 @item *running,thread-id="@var{thread}"
25511 The target is now running. The @var{thread} field tells which
25512 specific thread is now running, and can be @samp{all} if all threads
25513 are running. The frontend should assume that no interaction with a
25514 running thread is possible after this notification is produced.
25515 The frontend should not assume that this notification is output
25516 only once for any command. @value{GDBN} may emit this notification
25517 several times, either for different threads, because it cannot resume
25518 all threads together, or even for a single thread, if the thread must
25519 be stepped though some code before letting it run freely.
25521 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
25522 The target has stopped. The @var{reason} field can have one of the
25526 @item breakpoint-hit
25527 A breakpoint was reached.
25528 @item watchpoint-trigger
25529 A watchpoint was triggered.
25530 @item read-watchpoint-trigger
25531 A read watchpoint was triggered.
25532 @item access-watchpoint-trigger
25533 An access watchpoint was triggered.
25534 @item function-finished
25535 An -exec-finish or similar CLI command was accomplished.
25536 @item location-reached
25537 An -exec-until or similar CLI command was accomplished.
25538 @item watchpoint-scope
25539 A watchpoint has gone out of scope.
25540 @item end-stepping-range
25541 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
25542 similar CLI command was accomplished.
25543 @item exited-signalled
25544 The inferior exited because of a signal.
25546 The inferior exited.
25547 @item exited-normally
25548 The inferior exited normally.
25549 @item signal-received
25550 A signal was received by the inferior.
25552 The inferior has stopped due to a library being loaded or unloaded.
25553 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
25554 set or when a @code{catch load} or @code{catch unload} catchpoint is
25555 in use (@pxref{Set Catchpoints}).
25557 The inferior has forked. This is reported when @code{catch fork}
25558 (@pxref{Set Catchpoints}) has been used.
25560 The inferior has vforked. This is reported in when @code{catch vfork}
25561 (@pxref{Set Catchpoints}) has been used.
25562 @item syscall-entry
25563 The inferior entered a system call. This is reported when @code{catch
25564 syscall} (@pxref{Set Catchpoints}) has been used.
25565 @item syscall-entry
25566 The inferior returned from a system call. This is reported when
25567 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
25569 The inferior called @code{exec}. This is reported when @code{catch exec}
25570 (@pxref{Set Catchpoints}) has been used.
25573 The @var{id} field identifies the thread that directly caused the stop
25574 -- for example by hitting a breakpoint. Depending on whether all-stop
25575 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
25576 stop all threads, or only the thread that directly triggered the stop.
25577 If all threads are stopped, the @var{stopped} field will have the
25578 value of @code{"all"}. Otherwise, the value of the @var{stopped}
25579 field will be a list of thread identifiers. Presently, this list will
25580 always include a single thread, but frontend should be prepared to see
25581 several threads in the list. The @var{core} field reports the
25582 processor core on which the stop event has happened. This field may be absent
25583 if such information is not available.
25585 @item =thread-group-added,id="@var{id}"
25586 @itemx =thread-group-removed,id="@var{id}"
25587 A thread group was either added or removed. The @var{id} field
25588 contains the @value{GDBN} identifier of the thread group. When a thread
25589 group is added, it generally might not be associated with a running
25590 process. When a thread group is removed, its id becomes invalid and
25591 cannot be used in any way.
25593 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
25594 A thread group became associated with a running program,
25595 either because the program was just started or the thread group
25596 was attached to a program. The @var{id} field contains the
25597 @value{GDBN} identifier of the thread group. The @var{pid} field
25598 contains process identifier, specific to the operating system.
25600 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
25601 A thread group is no longer associated with a running program,
25602 either because the program has exited, or because it was detached
25603 from. The @var{id} field contains the @value{GDBN} identifier of the
25604 thread group. The @var{code} field is the exit code of the inferior; it exists
25605 only when the inferior exited with some code.
25607 @item =thread-created,id="@var{id}",group-id="@var{gid}"
25608 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
25609 A thread either was created, or has exited. The @var{id} field
25610 contains the @value{GDBN} identifier of the thread. The @var{gid}
25611 field identifies the thread group this thread belongs to.
25613 @item =thread-selected,id="@var{id}"
25614 Informs that the selected thread was changed as result of the last
25615 command. This notification is not emitted as result of @code{-thread-select}
25616 command but is emitted whenever an MI command that is not documented
25617 to change the selected thread actually changes it. In particular,
25618 invoking, directly or indirectly (via user-defined command), the CLI
25619 @code{thread} command, will generate this notification.
25621 We suggest that in response to this notification, front ends
25622 highlight the selected thread and cause subsequent commands to apply to
25625 @item =library-loaded,...
25626 Reports that a new library file was loaded by the program. This
25627 notification has 4 fields---@var{id}, @var{target-name},
25628 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
25629 opaque identifier of the library. For remote debugging case,
25630 @var{target-name} and @var{host-name} fields give the name of the
25631 library file on the target, and on the host respectively. For native
25632 debugging, both those fields have the same value. The
25633 @var{symbols-loaded} field is emitted only for backward compatibility
25634 and should not be relied on to convey any useful information. The
25635 @var{thread-group} field, if present, specifies the id of the thread
25636 group in whose context the library was loaded. If the field is
25637 absent, it means the library was loaded in the context of all present
25640 @item =library-unloaded,...
25641 Reports that a library was unloaded by the program. This notification
25642 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
25643 the same meaning as for the @code{=library-loaded} notification.
25644 The @var{thread-group} field, if present, specifies the id of the
25645 thread group in whose context the library was unloaded. If the field is
25646 absent, it means the library was unloaded in the context of all present
25649 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
25650 @itemx =traceframe-changed,end
25651 Reports that the trace frame was changed and its new number is
25652 @var{tfnum}. The number of the tracepoint associated with this trace
25653 frame is @var{tpnum}.
25655 @item =tsv-created,name=@var{name},initial=@var{initial}
25656 Reports that the new trace state variable @var{name} is created with
25657 initial value @var{initial}.
25659 @item =tsv-deleted,name=@var{name}
25660 @itemx =tsv-deleted
25661 Reports that the trace state variable @var{name} is deleted or all
25662 trace state variables are deleted.
25664 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
25665 Reports that the trace state variable @var{name} is modified with
25666 the initial value @var{initial}. The current value @var{current} of
25667 trace state variable is optional and is reported if the current
25668 value of trace state variable is known.
25670 @item =breakpoint-created,bkpt=@{...@}
25671 @itemx =breakpoint-modified,bkpt=@{...@}
25672 @itemx =breakpoint-deleted,id=@var{number}
25673 Reports that a breakpoint was created, modified, or deleted,
25674 respectively. Only user-visible breakpoints are reported to the MI
25677 The @var{bkpt} argument is of the same form as returned by the various
25678 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
25679 @var{number} is the ordinal number of the breakpoint.
25681 Note that if a breakpoint is emitted in the result record of a
25682 command, then it will not also be emitted in an async record.
25684 @item =record-started,thread-group="@var{id}"
25685 @itemx =record-stopped,thread-group="@var{id}"
25686 Execution log recording was either started or stopped on an
25687 inferior. The @var{id} is the @value{GDBN} identifier of the thread
25688 group corresponding to the affected inferior.
25690 @item =cmd-param-changed,param=@var{param},value=@var{value}
25691 Reports that a parameter of the command @code{set @var{param}} is
25692 changed to @var{value}. In the multi-word @code{set} command,
25693 the @var{param} is the whole parameter list to @code{set} command.
25694 For example, In command @code{set check type on}, @var{param}
25695 is @code{check type} and @var{value} is @code{on}.
25697 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
25698 Reports that bytes from @var{addr} to @var{data} + @var{len} were
25699 written in an inferior. The @var{id} is the identifier of the
25700 thread group corresponding to the affected inferior. The optional
25701 @code{type="code"} part is reported if the memory written to holds
25705 @node GDB/MI Breakpoint Information
25706 @subsection @sc{gdb/mi} Breakpoint Information
25708 When @value{GDBN} reports information about a breakpoint, a
25709 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
25714 The breakpoint number. For a breakpoint that represents one location
25715 of a multi-location breakpoint, this will be a dotted pair, like
25719 The type of the breakpoint. For ordinary breakpoints this will be
25720 @samp{breakpoint}, but many values are possible.
25723 If the type of the breakpoint is @samp{catchpoint}, then this
25724 indicates the exact type of catchpoint.
25727 This is the breakpoint disposition---either @samp{del}, meaning that
25728 the breakpoint will be deleted at the next stop, or @samp{keep},
25729 meaning that the breakpoint will not be deleted.
25732 This indicates whether the breakpoint is enabled, in which case the
25733 value is @samp{y}, or disabled, in which case the value is @samp{n}.
25734 Note that this is not the same as the field @code{enable}.
25737 The address of the breakpoint. This may be a hexidecimal number,
25738 giving the address; or the string @samp{<PENDING>}, for a pending
25739 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
25740 multiple locations. This field will not be present if no address can
25741 be determined. For example, a watchpoint does not have an address.
25744 If known, the function in which the breakpoint appears.
25745 If not known, this field is not present.
25748 The name of the source file which contains this function, if known.
25749 If not known, this field is not present.
25752 The full file name of the source file which contains this function, if
25753 known. If not known, this field is not present.
25756 The line number at which this breakpoint appears, if known.
25757 If not known, this field is not present.
25760 If the source file is not known, this field may be provided. If
25761 provided, this holds the address of the breakpoint, possibly followed
25765 If this breakpoint is pending, this field is present and holds the
25766 text used to set the breakpoint, as entered by the user.
25769 Where this breakpoint's condition is evaluated, either @samp{host} or
25773 If this is a thread-specific breakpoint, then this identifies the
25774 thread in which the breakpoint can trigger.
25777 If this breakpoint is restricted to a particular Ada task, then this
25778 field will hold the task identifier.
25781 If the breakpoint is conditional, this is the condition expression.
25784 The ignore count of the breakpoint.
25787 The enable count of the breakpoint.
25789 @item traceframe-usage
25792 @item static-tracepoint-marker-string-id
25793 For a static tracepoint, the name of the static tracepoint marker.
25796 For a masked watchpoint, this is the mask.
25799 A tracepoint's pass count.
25801 @item original-location
25802 The location of the breakpoint as originally specified by the user.
25803 This field is optional.
25806 The number of times the breakpoint has been hit.
25809 This field is only given for tracepoints. This is either @samp{y},
25810 meaning that the tracepoint is installed, or @samp{n}, meaning that it
25814 Some extra data, the exact contents of which are type-dependent.
25818 For example, here is what the output of @code{-break-insert}
25819 (@pxref{GDB/MI Breakpoint Commands}) might be:
25822 -> -break-insert main
25823 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
25824 enabled="y",addr="0x08048564",func="main",file="myprog.c",
25825 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
25830 @node GDB/MI Frame Information
25831 @subsection @sc{gdb/mi} Frame Information
25833 Response from many MI commands includes an information about stack
25834 frame. This information is a tuple that may have the following
25839 The level of the stack frame. The innermost frame has the level of
25840 zero. This field is always present.
25843 The name of the function corresponding to the frame. This field may
25844 be absent if @value{GDBN} is unable to determine the function name.
25847 The code address for the frame. This field is always present.
25850 The name of the source files that correspond to the frame's code
25851 address. This field may be absent.
25854 The source line corresponding to the frames' code address. This field
25858 The name of the binary file (either executable or shared library) the
25859 corresponds to the frame's code address. This field may be absent.
25863 @node GDB/MI Thread Information
25864 @subsection @sc{gdb/mi} Thread Information
25866 Whenever @value{GDBN} has to report an information about a thread, it
25867 uses a tuple with the following fields:
25871 The numeric id assigned to the thread by @value{GDBN}. This field is
25875 Target-specific string identifying the thread. This field is always present.
25878 Additional information about the thread provided by the target.
25879 It is supposed to be human-readable and not interpreted by the
25880 frontend. This field is optional.
25883 Either @samp{stopped} or @samp{running}, depending on whether the
25884 thread is presently running. This field is always present.
25887 The value of this field is an integer number of the processor core the
25888 thread was last seen on. This field is optional.
25891 @node GDB/MI Ada Exception Information
25892 @subsection @sc{gdb/mi} Ada Exception Information
25894 Whenever a @code{*stopped} record is emitted because the program
25895 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
25896 @value{GDBN} provides the name of the exception that was raised via
25897 the @code{exception-name} field.
25899 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25900 @node GDB/MI Simple Examples
25901 @section Simple Examples of @sc{gdb/mi} Interaction
25902 @cindex @sc{gdb/mi}, simple examples
25904 This subsection presents several simple examples of interaction using
25905 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
25906 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
25907 the output received from @sc{gdb/mi}.
25909 Note the line breaks shown in the examples are here only for
25910 readability, they don't appear in the real output.
25912 @subheading Setting a Breakpoint
25914 Setting a breakpoint generates synchronous output which contains detailed
25915 information of the breakpoint.
25918 -> -break-insert main
25919 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
25920 enabled="y",addr="0x08048564",func="main",file="myprog.c",
25921 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
25926 @subheading Program Execution
25928 Program execution generates asynchronous records and MI gives the
25929 reason that execution stopped.
25935 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
25936 frame=@{addr="0x08048564",func="main",
25937 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
25938 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
25943 <- *stopped,reason="exited-normally"
25947 @subheading Quitting @value{GDBN}
25949 Quitting @value{GDBN} just prints the result class @samp{^exit}.
25957 Please note that @samp{^exit} is printed immediately, but it might
25958 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
25959 performs necessary cleanups, including killing programs being debugged
25960 or disconnecting from debug hardware, so the frontend should wait till
25961 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
25962 fails to exit in reasonable time.
25964 @subheading A Bad Command
25966 Here's what happens if you pass a non-existent command:
25970 <- ^error,msg="Undefined MI command: rubbish"
25975 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25976 @node GDB/MI Command Description Format
25977 @section @sc{gdb/mi} Command Description Format
25979 The remaining sections describe blocks of commands. Each block of
25980 commands is laid out in a fashion similar to this section.
25982 @subheading Motivation
25984 The motivation for this collection of commands.
25986 @subheading Introduction
25988 A brief introduction to this collection of commands as a whole.
25990 @subheading Commands
25992 For each command in the block, the following is described:
25994 @subsubheading Synopsis
25997 -command @var{args}@dots{}
26000 @subsubheading Result
26002 @subsubheading @value{GDBN} Command
26004 The corresponding @value{GDBN} CLI command(s), if any.
26006 @subsubheading Example
26008 Example(s) formatted for readability. Some of the described commands have
26009 not been implemented yet and these are labeled N.A.@: (not available).
26012 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26013 @node GDB/MI Breakpoint Commands
26014 @section @sc{gdb/mi} Breakpoint Commands
26016 @cindex breakpoint commands for @sc{gdb/mi}
26017 @cindex @sc{gdb/mi}, breakpoint commands
26018 This section documents @sc{gdb/mi} commands for manipulating
26021 @subheading The @code{-break-after} Command
26022 @findex -break-after
26024 @subsubheading Synopsis
26027 -break-after @var{number} @var{count}
26030 The breakpoint number @var{number} is not in effect until it has been
26031 hit @var{count} times. To see how this is reflected in the output of
26032 the @samp{-break-list} command, see the description of the
26033 @samp{-break-list} command below.
26035 @subsubheading @value{GDBN} Command
26037 The corresponding @value{GDBN} command is @samp{ignore}.
26039 @subsubheading Example
26044 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
26045 enabled="y",addr="0x000100d0",func="main",file="hello.c",
26046 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
26054 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
26055 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26056 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26057 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26058 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26059 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26060 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26061 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26062 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
26063 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
26068 @subheading The @code{-break-catch} Command
26069 @findex -break-catch
26072 @subheading The @code{-break-commands} Command
26073 @findex -break-commands
26075 @subsubheading Synopsis
26078 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
26081 Specifies the CLI commands that should be executed when breakpoint
26082 @var{number} is hit. The parameters @var{command1} to @var{commandN}
26083 are the commands. If no command is specified, any previously-set
26084 commands are cleared. @xref{Break Commands}. Typical use of this
26085 functionality is tracing a program, that is, printing of values of
26086 some variables whenever breakpoint is hit and then continuing.
26088 @subsubheading @value{GDBN} Command
26090 The corresponding @value{GDBN} command is @samp{commands}.
26092 @subsubheading Example
26097 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
26098 enabled="y",addr="0x000100d0",func="main",file="hello.c",
26099 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
26102 -break-commands 1 "print v" "continue"
26107 @subheading The @code{-break-condition} Command
26108 @findex -break-condition
26110 @subsubheading Synopsis
26113 -break-condition @var{number} @var{expr}
26116 Breakpoint @var{number} will stop the program only if the condition in
26117 @var{expr} is true. The condition becomes part of the
26118 @samp{-break-list} output (see the description of the @samp{-break-list}
26121 @subsubheading @value{GDBN} Command
26123 The corresponding @value{GDBN} command is @samp{condition}.
26125 @subsubheading Example
26129 -break-condition 1 1
26133 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
26134 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26135 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26136 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26137 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26138 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26139 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26140 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26141 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
26142 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
26146 @subheading The @code{-break-delete} Command
26147 @findex -break-delete
26149 @subsubheading Synopsis
26152 -break-delete ( @var{breakpoint} )+
26155 Delete the breakpoint(s) whose number(s) are specified in the argument
26156 list. This is obviously reflected in the breakpoint list.
26158 @subsubheading @value{GDBN} Command
26160 The corresponding @value{GDBN} command is @samp{delete}.
26162 @subsubheading Example
26170 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
26171 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26172 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26173 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26174 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26175 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26176 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26181 @subheading The @code{-break-disable} Command
26182 @findex -break-disable
26184 @subsubheading Synopsis
26187 -break-disable ( @var{breakpoint} )+
26190 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
26191 break list is now set to @samp{n} for the named @var{breakpoint}(s).
26193 @subsubheading @value{GDBN} Command
26195 The corresponding @value{GDBN} command is @samp{disable}.
26197 @subsubheading Example
26205 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
26206 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26207 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26208 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26209 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26210 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26211 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26212 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
26213 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
26214 line="5",thread-groups=["i1"],times="0"@}]@}
26218 @subheading The @code{-break-enable} Command
26219 @findex -break-enable
26221 @subsubheading Synopsis
26224 -break-enable ( @var{breakpoint} )+
26227 Enable (previously disabled) @var{breakpoint}(s).
26229 @subsubheading @value{GDBN} Command
26231 The corresponding @value{GDBN} command is @samp{enable}.
26233 @subsubheading Example
26241 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
26242 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26243 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26244 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26245 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26246 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26247 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26248 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
26249 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
26250 line="5",thread-groups=["i1"],times="0"@}]@}
26254 @subheading The @code{-break-info} Command
26255 @findex -break-info
26257 @subsubheading Synopsis
26260 -break-info @var{breakpoint}
26264 Get information about a single breakpoint.
26266 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
26267 Information}, for details on the format of each breakpoint in the
26270 @subsubheading @value{GDBN} Command
26272 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
26274 @subsubheading Example
26277 @subheading The @code{-break-insert} Command
26278 @findex -break-insert
26280 @subsubheading Synopsis
26283 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
26284 [ -c @var{condition} ] [ -i @var{ignore-count} ]
26285 [ -p @var{thread-id} ] [ @var{location} ]
26289 If specified, @var{location}, can be one of:
26296 @item filename:linenum
26297 @item filename:function
26301 The possible optional parameters of this command are:
26305 Insert a temporary breakpoint.
26307 Insert a hardware breakpoint.
26309 If @var{location} cannot be parsed (for example if it
26310 refers to unknown files or functions), create a pending
26311 breakpoint. Without this flag, @value{GDBN} will report
26312 an error, and won't create a breakpoint, if @var{location}
26315 Create a disabled breakpoint.
26317 Create a tracepoint. @xref{Tracepoints}. When this parameter
26318 is used together with @samp{-h}, a fast tracepoint is created.
26319 @item -c @var{condition}
26320 Make the breakpoint conditional on @var{condition}.
26321 @item -i @var{ignore-count}
26322 Initialize the @var{ignore-count}.
26323 @item -p @var{thread-id}
26324 Restrict the breakpoint to the specified @var{thread-id}.
26327 @subsubheading Result
26329 @xref{GDB/MI Breakpoint Information}, for details on the format of the
26330 resulting breakpoint.
26332 Note: this format is open to change.
26333 @c An out-of-band breakpoint instead of part of the result?
26335 @subsubheading @value{GDBN} Command
26337 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
26338 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
26340 @subsubheading Example
26345 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
26346 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
26349 -break-insert -t foo
26350 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
26351 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
26355 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
26356 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26357 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26358 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26359 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26360 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26361 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26362 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26363 addr="0x0001072c", func="main",file="recursive2.c",
26364 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
26366 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
26367 addr="0x00010774",func="foo",file="recursive2.c",
26368 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
26371 @c -break-insert -r foo.*
26372 @c ~int foo(int, int);
26373 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
26374 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
26379 @subheading The @code{-dprintf-insert} Command
26380 @findex -dprintf-insert
26382 @subsubheading Synopsis
26385 -dprintf-insert [ -t ] [ -f ] [ -d ]
26386 [ -c @var{condition} ] [ -i @var{ignore-count} ]
26387 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
26392 If specified, @var{location}, can be one of:
26395 @item @var{function}
26398 @c @item @var{linenum}
26399 @item @var{filename}:@var{linenum}
26400 @item @var{filename}:function
26401 @item *@var{address}
26404 The possible optional parameters of this command are:
26408 Insert a temporary breakpoint.
26410 If @var{location} cannot be parsed (for example, if it
26411 refers to unknown files or functions), create a pending
26412 breakpoint. Without this flag, @value{GDBN} will report
26413 an error, and won't create a breakpoint, if @var{location}
26416 Create a disabled breakpoint.
26417 @item -c @var{condition}
26418 Make the breakpoint conditional on @var{condition}.
26419 @item -i @var{ignore-count}
26420 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
26421 to @var{ignore-count}.
26422 @item -p @var{thread-id}
26423 Restrict the breakpoint to the specified @var{thread-id}.
26426 @subsubheading Result
26428 @xref{GDB/MI Breakpoint Information}, for details on the format of the
26429 resulting breakpoint.
26431 @c An out-of-band breakpoint instead of part of the result?
26433 @subsubheading @value{GDBN} Command
26435 The corresponding @value{GDBN} command is @samp{dprintf}.
26437 @subsubheading Example
26441 4-dprintf-insert foo "At foo entry\n"
26442 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
26443 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
26444 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
26445 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
26446 original-location="foo"@}
26448 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
26449 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
26450 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
26451 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
26452 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
26453 original-location="mi-dprintf.c:26"@}
26457 @subheading The @code{-break-list} Command
26458 @findex -break-list
26460 @subsubheading Synopsis
26466 Displays the list of inserted breakpoints, showing the following fields:
26470 number of the breakpoint
26472 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
26474 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
26477 is the breakpoint enabled or no: @samp{y} or @samp{n}
26479 memory location at which the breakpoint is set
26481 logical location of the breakpoint, expressed by function name, file
26483 @item Thread-groups
26484 list of thread groups to which this breakpoint applies
26486 number of times the breakpoint has been hit
26489 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
26490 @code{body} field is an empty list.
26492 @subsubheading @value{GDBN} Command
26494 The corresponding @value{GDBN} command is @samp{info break}.
26496 @subsubheading Example
26501 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
26502 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26503 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26504 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26505 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26506 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26507 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26508 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26509 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
26511 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
26512 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
26513 line="13",thread-groups=["i1"],times="0"@}]@}
26517 Here's an example of the result when there are no breakpoints:
26522 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
26523 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26524 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26525 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26526 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26527 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26528 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26533 @subheading The @code{-break-passcount} Command
26534 @findex -break-passcount
26536 @subsubheading Synopsis
26539 -break-passcount @var{tracepoint-number} @var{passcount}
26542 Set the passcount for tracepoint @var{tracepoint-number} to
26543 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
26544 is not a tracepoint, error is emitted. This corresponds to CLI
26545 command @samp{passcount}.
26547 @subheading The @code{-break-watch} Command
26548 @findex -break-watch
26550 @subsubheading Synopsis
26553 -break-watch [ -a | -r ]
26556 Create a watchpoint. With the @samp{-a} option it will create an
26557 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
26558 read from or on a write to the memory location. With the @samp{-r}
26559 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
26560 trigger only when the memory location is accessed for reading. Without
26561 either of the options, the watchpoint created is a regular watchpoint,
26562 i.e., it will trigger when the memory location is accessed for writing.
26563 @xref{Set Watchpoints, , Setting Watchpoints}.
26565 Note that @samp{-break-list} will report a single list of watchpoints and
26566 breakpoints inserted.
26568 @subsubheading @value{GDBN} Command
26570 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
26573 @subsubheading Example
26575 Setting a watchpoint on a variable in the @code{main} function:
26580 ^done,wpt=@{number="2",exp="x"@}
26585 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
26586 value=@{old="-268439212",new="55"@},
26587 frame=@{func="main",args=[],file="recursive2.c",
26588 fullname="/home/foo/bar/recursive2.c",line="5"@}
26592 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
26593 the program execution twice: first for the variable changing value, then
26594 for the watchpoint going out of scope.
26599 ^done,wpt=@{number="5",exp="C"@}
26604 *stopped,reason="watchpoint-trigger",
26605 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
26606 frame=@{func="callee4",args=[],
26607 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26608 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
26613 *stopped,reason="watchpoint-scope",wpnum="5",
26614 frame=@{func="callee3",args=[@{name="strarg",
26615 value="0x11940 \"A string argument.\""@}],
26616 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26617 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
26621 Listing breakpoints and watchpoints, at different points in the program
26622 execution. Note that once the watchpoint goes out of scope, it is
26628 ^done,wpt=@{number="2",exp="C"@}
26631 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
26632 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26633 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26634 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26635 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26636 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26637 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26638 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26639 addr="0x00010734",func="callee4",
26640 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26641 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
26643 bkpt=@{number="2",type="watchpoint",disp="keep",
26644 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
26649 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
26650 value=@{old="-276895068",new="3"@},
26651 frame=@{func="callee4",args=[],
26652 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26653 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
26656 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
26657 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26658 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26659 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26660 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26661 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26662 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26663 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26664 addr="0x00010734",func="callee4",
26665 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26666 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
26668 bkpt=@{number="2",type="watchpoint",disp="keep",
26669 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
26673 ^done,reason="watchpoint-scope",wpnum="2",
26674 frame=@{func="callee3",args=[@{name="strarg",
26675 value="0x11940 \"A string argument.\""@}],
26676 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26677 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
26680 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
26681 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26682 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26683 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26684 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26685 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26686 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26687 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26688 addr="0x00010734",func="callee4",
26689 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26690 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
26691 thread-groups=["i1"],times="1"@}]@}
26696 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26697 @node GDB/MI Catchpoint Commands
26698 @section @sc{gdb/mi} Catchpoint Commands
26700 This section documents @sc{gdb/mi} commands for manipulating
26704 * Shared Library GDB/MI Catchpoint Commands::
26705 * Ada Exception GDB/MI Catchpoint Commands::
26708 @node Shared Library GDB/MI Catchpoint Commands
26709 @subsection Shared Library @sc{gdb/mi} Catchpoints
26711 @subheading The @code{-catch-load} Command
26712 @findex -catch-load
26714 @subsubheading Synopsis
26717 -catch-load [ -t ] [ -d ] @var{regexp}
26720 Add a catchpoint for library load events. If the @samp{-t} option is used,
26721 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
26722 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
26723 in a disabled state. The @samp{regexp} argument is a regular
26724 expression used to match the name of the loaded library.
26727 @subsubheading @value{GDBN} Command
26729 The corresponding @value{GDBN} command is @samp{catch load}.
26731 @subsubheading Example
26734 -catch-load -t foo.so
26735 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
26736 what="load of library matching foo.so",catch-type="load",times="0"@}
26741 @subheading The @code{-catch-unload} Command
26742 @findex -catch-unload
26744 @subsubheading Synopsis
26747 -catch-unload [ -t ] [ -d ] @var{regexp}
26750 Add a catchpoint for library unload events. If the @samp{-t} option is
26751 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
26752 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
26753 created in a disabled state. The @samp{regexp} argument is a regular
26754 expression used to match the name of the unloaded library.
26756 @subsubheading @value{GDBN} Command
26758 The corresponding @value{GDBN} command is @samp{catch unload}.
26760 @subsubheading Example
26763 -catch-unload -d bar.so
26764 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
26765 what="load of library matching bar.so",catch-type="unload",times="0"@}
26769 @node Ada Exception GDB/MI Catchpoint Commands
26770 @subsection Ada Exception @sc{gdb/mi} Catchpoints
26772 The following @sc{gdb/mi} commands can be used to create catchpoints
26773 that stop the execution when Ada exceptions are being raised.
26775 @subheading The @code{-catch-assert} Command
26776 @findex -catch-assert
26778 @subsubheading Synopsis
26781 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
26784 Add a catchpoint for failed Ada assertions.
26786 The possible optional parameters for this command are:
26789 @item -c @var{condition}
26790 Make the catchpoint conditional on @var{condition}.
26792 Create a disabled catchpoint.
26794 Create a temporary catchpoint.
26797 @subsubheading @value{GDBN} Command
26799 The corresponding @value{GDBN} command is @samp{catch assert}.
26801 @subsubheading Example
26805 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
26806 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
26807 thread-groups=["i1"],times="0",
26808 original-location="__gnat_debug_raise_assert_failure"@}
26812 @subheading The @code{-catch-exception} Command
26813 @findex -catch-exception
26815 @subsubheading Synopsis
26818 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
26822 Add a catchpoint stopping when Ada exceptions are raised.
26823 By default, the command stops the program when any Ada exception
26824 gets raised. But it is also possible, by using some of the
26825 optional parameters described below, to create more selective
26828 The possible optional parameters for this command are:
26831 @item -c @var{condition}
26832 Make the catchpoint conditional on @var{condition}.
26834 Create a disabled catchpoint.
26835 @item -e @var{exception-name}
26836 Only stop when @var{exception-name} is raised. This option cannot
26837 be used combined with @samp{-u}.
26839 Create a temporary catchpoint.
26841 Stop only when an unhandled exception gets raised. This option
26842 cannot be used combined with @samp{-e}.
26845 @subsubheading @value{GDBN} Command
26847 The corresponding @value{GDBN} commands are @samp{catch exception}
26848 and @samp{catch exception unhandled}.
26850 @subsubheading Example
26853 -catch-exception -e Program_Error
26854 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
26855 enabled="y",addr="0x0000000000404874",
26856 what="`Program_Error' Ada exception", thread-groups=["i1"],
26857 times="0",original-location="__gnat_debug_raise_exception"@}
26861 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26862 @node GDB/MI Program Context
26863 @section @sc{gdb/mi} Program Context
26865 @subheading The @code{-exec-arguments} Command
26866 @findex -exec-arguments
26869 @subsubheading Synopsis
26872 -exec-arguments @var{args}
26875 Set the inferior program arguments, to be used in the next
26878 @subsubheading @value{GDBN} Command
26880 The corresponding @value{GDBN} command is @samp{set args}.
26882 @subsubheading Example
26886 -exec-arguments -v word
26893 @subheading The @code{-exec-show-arguments} Command
26894 @findex -exec-show-arguments
26896 @subsubheading Synopsis
26899 -exec-show-arguments
26902 Print the arguments of the program.
26904 @subsubheading @value{GDBN} Command
26906 The corresponding @value{GDBN} command is @samp{show args}.
26908 @subsubheading Example
26913 @subheading The @code{-environment-cd} Command
26914 @findex -environment-cd
26916 @subsubheading Synopsis
26919 -environment-cd @var{pathdir}
26922 Set @value{GDBN}'s working directory.
26924 @subsubheading @value{GDBN} Command
26926 The corresponding @value{GDBN} command is @samp{cd}.
26928 @subsubheading Example
26932 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
26938 @subheading The @code{-environment-directory} Command
26939 @findex -environment-directory
26941 @subsubheading Synopsis
26944 -environment-directory [ -r ] [ @var{pathdir} ]+
26947 Add directories @var{pathdir} to beginning of search path for source files.
26948 If the @samp{-r} option is used, the search path is reset to the default
26949 search path. If directories @var{pathdir} are supplied in addition to the
26950 @samp{-r} option, the search path is first reset and then addition
26952 Multiple directories may be specified, separated by blanks. Specifying
26953 multiple directories in a single command
26954 results in the directories added to the beginning of the
26955 search path in the same order they were presented in the command.
26956 If blanks are needed as
26957 part of a directory name, double-quotes should be used around
26958 the name. In the command output, the path will show up separated
26959 by the system directory-separator character. The directory-separator
26960 character must not be used
26961 in any directory name.
26962 If no directories are specified, the current search path is displayed.
26964 @subsubheading @value{GDBN} Command
26966 The corresponding @value{GDBN} command is @samp{dir}.
26968 @subsubheading Example
26972 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
26973 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
26975 -environment-directory ""
26976 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
26978 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
26979 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
26981 -environment-directory -r
26982 ^done,source-path="$cdir:$cwd"
26987 @subheading The @code{-environment-path} Command
26988 @findex -environment-path
26990 @subsubheading Synopsis
26993 -environment-path [ -r ] [ @var{pathdir} ]+
26996 Add directories @var{pathdir} to beginning of search path for object files.
26997 If the @samp{-r} option is used, the search path is reset to the original
26998 search path that existed at gdb start-up. If directories @var{pathdir} are
26999 supplied in addition to the
27000 @samp{-r} option, the search path is first reset and then addition
27002 Multiple directories may be specified, separated by blanks. Specifying
27003 multiple directories in a single command
27004 results in the directories added to the beginning of the
27005 search path in the same order they were presented in the command.
27006 If blanks are needed as
27007 part of a directory name, double-quotes should be used around
27008 the name. In the command output, the path will show up separated
27009 by the system directory-separator character. The directory-separator
27010 character must not be used
27011 in any directory name.
27012 If no directories are specified, the current path is displayed.
27015 @subsubheading @value{GDBN} Command
27017 The corresponding @value{GDBN} command is @samp{path}.
27019 @subsubheading Example
27024 ^done,path="/usr/bin"
27026 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
27027 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
27029 -environment-path -r /usr/local/bin
27030 ^done,path="/usr/local/bin:/usr/bin"
27035 @subheading The @code{-environment-pwd} Command
27036 @findex -environment-pwd
27038 @subsubheading Synopsis
27044 Show the current working directory.
27046 @subsubheading @value{GDBN} Command
27048 The corresponding @value{GDBN} command is @samp{pwd}.
27050 @subsubheading Example
27055 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
27059 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27060 @node GDB/MI Thread Commands
27061 @section @sc{gdb/mi} Thread Commands
27064 @subheading The @code{-thread-info} Command
27065 @findex -thread-info
27067 @subsubheading Synopsis
27070 -thread-info [ @var{thread-id} ]
27073 Reports information about either a specific thread, if
27074 the @var{thread-id} parameter is present, or about all
27075 threads. When printing information about all threads,
27076 also reports the current thread.
27078 @subsubheading @value{GDBN} Command
27080 The @samp{info thread} command prints the same information
27083 @subsubheading Result
27085 The result is a list of threads. The following attributes are
27086 defined for a given thread:
27090 This field exists only for the current thread. It has the value @samp{*}.
27093 The identifier that @value{GDBN} uses to refer to the thread.
27096 The identifier that the target uses to refer to the thread.
27099 Extra information about the thread, in a target-specific format. This
27103 The name of the thread. If the user specified a name using the
27104 @code{thread name} command, then this name is given. Otherwise, if
27105 @value{GDBN} can extract the thread name from the target, then that
27106 name is given. If @value{GDBN} cannot find the thread name, then this
27110 The stack frame currently executing in the thread.
27113 The thread's state. The @samp{state} field may have the following
27118 The thread is stopped. Frame information is available for stopped
27122 The thread is running. There's no frame information for running
27128 If @value{GDBN} can find the CPU core on which this thread is running,
27129 then this field is the core identifier. This field is optional.
27133 @subsubheading Example
27138 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
27139 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
27140 args=[]@},state="running"@},
27141 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
27142 frame=@{level="0",addr="0x0804891f",func="foo",
27143 args=[@{name="i",value="10"@}],
27144 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},
27145 state="running"@}],
27146 current-thread-id="1"
27150 @subheading The @code{-thread-list-ids} Command
27151 @findex -thread-list-ids
27153 @subsubheading Synopsis
27159 Produces a list of the currently known @value{GDBN} thread ids. At the
27160 end of the list it also prints the total number of such threads.
27162 This command is retained for historical reasons, the
27163 @code{-thread-info} command should be used instead.
27165 @subsubheading @value{GDBN} Command
27167 Part of @samp{info threads} supplies the same information.
27169 @subsubheading Example
27174 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
27175 current-thread-id="1",number-of-threads="3"
27180 @subheading The @code{-thread-select} Command
27181 @findex -thread-select
27183 @subsubheading Synopsis
27186 -thread-select @var{threadnum}
27189 Make @var{threadnum} the current thread. It prints the number of the new
27190 current thread, and the topmost frame for that thread.
27192 This command is deprecated in favor of explicitly using the
27193 @samp{--thread} option to each command.
27195 @subsubheading @value{GDBN} Command
27197 The corresponding @value{GDBN} command is @samp{thread}.
27199 @subsubheading Example
27206 *stopped,reason="end-stepping-range",thread-id="2",line="187",
27207 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
27211 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
27212 number-of-threads="3"
27215 ^done,new-thread-id="3",
27216 frame=@{level="0",func="vprintf",
27217 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
27218 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
27222 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27223 @node GDB/MI Ada Tasking Commands
27224 @section @sc{gdb/mi} Ada Tasking Commands
27226 @subheading The @code{-ada-task-info} Command
27227 @findex -ada-task-info
27229 @subsubheading Synopsis
27232 -ada-task-info [ @var{task-id} ]
27235 Reports information about either a specific Ada task, if the
27236 @var{task-id} parameter is present, or about all Ada tasks.
27238 @subsubheading @value{GDBN} Command
27240 The @samp{info tasks} command prints the same information
27241 about all Ada tasks (@pxref{Ada Tasks}).
27243 @subsubheading Result
27245 The result is a table of Ada tasks. The following columns are
27246 defined for each Ada task:
27250 This field exists only for the current thread. It has the value @samp{*}.
27253 The identifier that @value{GDBN} uses to refer to the Ada task.
27256 The identifier that the target uses to refer to the Ada task.
27259 The identifier of the thread corresponding to the Ada task.
27261 This field should always exist, as Ada tasks are always implemented
27262 on top of a thread. But if @value{GDBN} cannot find this corresponding
27263 thread for any reason, the field is omitted.
27266 This field exists only when the task was created by another task.
27267 In this case, it provides the ID of the parent task.
27270 The base priority of the task.
27273 The current state of the task. For a detailed description of the
27274 possible states, see @ref{Ada Tasks}.
27277 The name of the task.
27281 @subsubheading Example
27285 ^done,tasks=@{nr_rows="3",nr_cols="8",
27286 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
27287 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
27288 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
27289 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
27290 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
27291 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
27292 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
27293 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
27294 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
27295 state="Child Termination Wait",name="main_task"@}]@}
27299 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27300 @node GDB/MI Program Execution
27301 @section @sc{gdb/mi} Program Execution
27303 These are the asynchronous commands which generate the out-of-band
27304 record @samp{*stopped}. Currently @value{GDBN} only really executes
27305 asynchronously with remote targets and this interaction is mimicked in
27308 @subheading The @code{-exec-continue} Command
27309 @findex -exec-continue
27311 @subsubheading Synopsis
27314 -exec-continue [--reverse] [--all|--thread-group N]
27317 Resumes the execution of the inferior program, which will continue
27318 to execute until it reaches a debugger stop event. If the
27319 @samp{--reverse} option is specified, execution resumes in reverse until
27320 it reaches a stop event. Stop events may include
27323 breakpoints or watchpoints
27325 signals or exceptions
27327 the end of the process (or its beginning under @samp{--reverse})
27329 the end or beginning of a replay log if one is being used.
27331 In all-stop mode (@pxref{All-Stop
27332 Mode}), may resume only one thread, or all threads, depending on the
27333 value of the @samp{scheduler-locking} variable. If @samp{--all} is
27334 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
27335 ignored in all-stop mode. If the @samp{--thread-group} options is
27336 specified, then all threads in that thread group are resumed.
27338 @subsubheading @value{GDBN} Command
27340 The corresponding @value{GDBN} corresponding is @samp{continue}.
27342 @subsubheading Example
27349 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
27350 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
27356 @subheading The @code{-exec-finish} Command
27357 @findex -exec-finish
27359 @subsubheading Synopsis
27362 -exec-finish [--reverse]
27365 Resumes the execution of the inferior program until the current
27366 function is exited. Displays the results returned by the function.
27367 If the @samp{--reverse} option is specified, resumes the reverse
27368 execution of the inferior program until the point where current
27369 function was called.
27371 @subsubheading @value{GDBN} Command
27373 The corresponding @value{GDBN} command is @samp{finish}.
27375 @subsubheading Example
27377 Function returning @code{void}.
27384 *stopped,reason="function-finished",frame=@{func="main",args=[],
27385 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
27389 Function returning other than @code{void}. The name of the internal
27390 @value{GDBN} variable storing the result is printed, together with the
27397 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
27398 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
27399 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27400 gdb-result-var="$1",return-value="0"
27405 @subheading The @code{-exec-interrupt} Command
27406 @findex -exec-interrupt
27408 @subsubheading Synopsis
27411 -exec-interrupt [--all|--thread-group N]
27414 Interrupts the background execution of the target. Note how the token
27415 associated with the stop message is the one for the execution command
27416 that has been interrupted. The token for the interrupt itself only
27417 appears in the @samp{^done} output. If the user is trying to
27418 interrupt a non-running program, an error message will be printed.
27420 Note that when asynchronous execution is enabled, this command is
27421 asynchronous just like other execution commands. That is, first the
27422 @samp{^done} response will be printed, and the target stop will be
27423 reported after that using the @samp{*stopped} notification.
27425 In non-stop mode, only the context thread is interrupted by default.
27426 All threads (in all inferiors) will be interrupted if the
27427 @samp{--all} option is specified. If the @samp{--thread-group}
27428 option is specified, all threads in that group will be interrupted.
27430 @subsubheading @value{GDBN} Command
27432 The corresponding @value{GDBN} command is @samp{interrupt}.
27434 @subsubheading Example
27445 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
27446 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
27447 fullname="/home/foo/bar/try.c",line="13"@}
27452 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
27456 @subheading The @code{-exec-jump} Command
27459 @subsubheading Synopsis
27462 -exec-jump @var{location}
27465 Resumes execution of the inferior program at the location specified by
27466 parameter. @xref{Specify Location}, for a description of the
27467 different forms of @var{location}.
27469 @subsubheading @value{GDBN} Command
27471 The corresponding @value{GDBN} command is @samp{jump}.
27473 @subsubheading Example
27476 -exec-jump foo.c:10
27477 *running,thread-id="all"
27482 @subheading The @code{-exec-next} Command
27485 @subsubheading Synopsis
27488 -exec-next [--reverse]
27491 Resumes execution of the inferior program, stopping when the beginning
27492 of the next source line is reached.
27494 If the @samp{--reverse} option is specified, resumes reverse execution
27495 of the inferior program, stopping at the beginning of the previous
27496 source line. If you issue this command on the first line of a
27497 function, it will take you back to the caller of that function, to the
27498 source line where the function was called.
27501 @subsubheading @value{GDBN} Command
27503 The corresponding @value{GDBN} command is @samp{next}.
27505 @subsubheading Example
27511 *stopped,reason="end-stepping-range",line="8",file="hello.c"
27516 @subheading The @code{-exec-next-instruction} Command
27517 @findex -exec-next-instruction
27519 @subsubheading Synopsis
27522 -exec-next-instruction [--reverse]
27525 Executes one machine instruction. If the instruction is a function
27526 call, continues until the function returns. If the program stops at an
27527 instruction in the middle of a source line, the address will be
27530 If the @samp{--reverse} option is specified, resumes reverse execution
27531 of the inferior program, stopping at the previous instruction. If the
27532 previously executed instruction was a return from another function,
27533 it will continue to execute in reverse until the call to that function
27534 (from the current stack frame) is reached.
27536 @subsubheading @value{GDBN} Command
27538 The corresponding @value{GDBN} command is @samp{nexti}.
27540 @subsubheading Example
27544 -exec-next-instruction
27548 *stopped,reason="end-stepping-range",
27549 addr="0x000100d4",line="5",file="hello.c"
27554 @subheading The @code{-exec-return} Command
27555 @findex -exec-return
27557 @subsubheading Synopsis
27563 Makes current function return immediately. Doesn't execute the inferior.
27564 Displays the new current frame.
27566 @subsubheading @value{GDBN} Command
27568 The corresponding @value{GDBN} command is @samp{return}.
27570 @subsubheading Example
27574 200-break-insert callee4
27575 200^done,bkpt=@{number="1",addr="0x00010734",
27576 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
27581 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
27582 frame=@{func="callee4",args=[],
27583 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27584 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
27590 111^done,frame=@{level="0",func="callee3",
27591 args=[@{name="strarg",
27592 value="0x11940 \"A string argument.\""@}],
27593 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27594 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
27599 @subheading The @code{-exec-run} Command
27602 @subsubheading Synopsis
27605 -exec-run [ --all | --thread-group N ] [ --start ]
27608 Starts execution of the inferior from the beginning. The inferior
27609 executes until either a breakpoint is encountered or the program
27610 exits. In the latter case the output will include an exit code, if
27611 the program has exited exceptionally.
27613 When neither the @samp{--all} nor the @samp{--thread-group} option
27614 is specified, the current inferior is started. If the
27615 @samp{--thread-group} option is specified, it should refer to a thread
27616 group of type @samp{process}, and that thread group will be started.
27617 If the @samp{--all} option is specified, then all inferiors will be started.
27619 Using the @samp{--start} option instructs the debugger to stop
27620 the execution at the start of the inferior's main subprogram,
27621 following the same behavior as the @code{start} command
27622 (@pxref{Starting}).
27624 @subsubheading @value{GDBN} Command
27626 The corresponding @value{GDBN} command is @samp{run}.
27628 @subsubheading Examples
27633 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
27638 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
27639 frame=@{func="main",args=[],file="recursive2.c",
27640 fullname="/home/foo/bar/recursive2.c",line="4"@}
27645 Program exited normally:
27653 *stopped,reason="exited-normally"
27658 Program exited exceptionally:
27666 *stopped,reason="exited",exit-code="01"
27670 Another way the program can terminate is if it receives a signal such as
27671 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
27675 *stopped,reason="exited-signalled",signal-name="SIGINT",
27676 signal-meaning="Interrupt"
27680 @c @subheading -exec-signal
27683 @subheading The @code{-exec-step} Command
27686 @subsubheading Synopsis
27689 -exec-step [--reverse]
27692 Resumes execution of the inferior program, stopping when the beginning
27693 of the next source line is reached, if the next source line is not a
27694 function call. If it is, stop at the first instruction of the called
27695 function. If the @samp{--reverse} option is specified, resumes reverse
27696 execution of the inferior program, stopping at the beginning of the
27697 previously executed source line.
27699 @subsubheading @value{GDBN} Command
27701 The corresponding @value{GDBN} command is @samp{step}.
27703 @subsubheading Example
27705 Stepping into a function:
27711 *stopped,reason="end-stepping-range",
27712 frame=@{func="foo",args=[@{name="a",value="10"@},
27713 @{name="b",value="0"@}],file="recursive2.c",
27714 fullname="/home/foo/bar/recursive2.c",line="11"@}
27724 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
27729 @subheading The @code{-exec-step-instruction} Command
27730 @findex -exec-step-instruction
27732 @subsubheading Synopsis
27735 -exec-step-instruction [--reverse]
27738 Resumes the inferior which executes one machine instruction. If the
27739 @samp{--reverse} option is specified, resumes reverse execution of the
27740 inferior program, stopping at the previously executed instruction.
27741 The output, once @value{GDBN} has stopped, will vary depending on
27742 whether we have stopped in the middle of a source line or not. In the
27743 former case, the address at which the program stopped will be printed
27746 @subsubheading @value{GDBN} Command
27748 The corresponding @value{GDBN} command is @samp{stepi}.
27750 @subsubheading Example
27754 -exec-step-instruction
27758 *stopped,reason="end-stepping-range",
27759 frame=@{func="foo",args=[],file="try.c",
27760 fullname="/home/foo/bar/try.c",line="10"@}
27762 -exec-step-instruction
27766 *stopped,reason="end-stepping-range",
27767 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
27768 fullname="/home/foo/bar/try.c",line="10"@}
27773 @subheading The @code{-exec-until} Command
27774 @findex -exec-until
27776 @subsubheading Synopsis
27779 -exec-until [ @var{location} ]
27782 Executes the inferior until the @var{location} specified in the
27783 argument is reached. If there is no argument, the inferior executes
27784 until a source line greater than the current one is reached. The
27785 reason for stopping in this case will be @samp{location-reached}.
27787 @subsubheading @value{GDBN} Command
27789 The corresponding @value{GDBN} command is @samp{until}.
27791 @subsubheading Example
27795 -exec-until recursive2.c:6
27799 *stopped,reason="location-reached",frame=@{func="main",args=[],
27800 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
27805 @subheading -file-clear
27806 Is this going away????
27809 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27810 @node GDB/MI Stack Manipulation
27811 @section @sc{gdb/mi} Stack Manipulation Commands
27813 @subheading The @code{-enable-frame-filters} Command
27814 @findex -enable-frame-filters
27817 -enable-frame-filters
27820 @value{GDBN} allows Python-based frame filters to affect the output of
27821 the MI commands relating to stack traces. As there is no way to
27822 implement this in a fully backward-compatible way, a front end must
27823 request that this functionality be enabled.
27825 Once enabled, this feature cannot be disabled.
27827 Note that if Python support has not been compiled into @value{GDBN},
27828 this command will still succeed (and do nothing).
27830 @subheading The @code{-stack-info-frame} Command
27831 @findex -stack-info-frame
27833 @subsubheading Synopsis
27839 Get info on the selected frame.
27841 @subsubheading @value{GDBN} Command
27843 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
27844 (without arguments).
27846 @subsubheading Example
27851 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
27852 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27853 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
27857 @subheading The @code{-stack-info-depth} Command
27858 @findex -stack-info-depth
27860 @subsubheading Synopsis
27863 -stack-info-depth [ @var{max-depth} ]
27866 Return the depth of the stack. If the integer argument @var{max-depth}
27867 is specified, do not count beyond @var{max-depth} frames.
27869 @subsubheading @value{GDBN} Command
27871 There's no equivalent @value{GDBN} command.
27873 @subsubheading Example
27875 For a stack with frame levels 0 through 11:
27882 -stack-info-depth 4
27885 -stack-info-depth 12
27888 -stack-info-depth 11
27891 -stack-info-depth 13
27896 @anchor{-stack-list-arguments}
27897 @subheading The @code{-stack-list-arguments} Command
27898 @findex -stack-list-arguments
27900 @subsubheading Synopsis
27903 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
27904 [ @var{low-frame} @var{high-frame} ]
27907 Display a list of the arguments for the frames between @var{low-frame}
27908 and @var{high-frame} (inclusive). If @var{low-frame} and
27909 @var{high-frame} are not provided, list the arguments for the whole
27910 call stack. If the two arguments are equal, show the single frame
27911 at the corresponding level. It is an error if @var{low-frame} is
27912 larger than the actual number of frames. On the other hand,
27913 @var{high-frame} may be larger than the actual number of frames, in
27914 which case only existing frames will be returned.
27916 If @var{print-values} is 0 or @code{--no-values}, print only the names of
27917 the variables; if it is 1 or @code{--all-values}, print also their
27918 values; and if it is 2 or @code{--simple-values}, print the name,
27919 type and value for simple data types, and the name and type for arrays,
27920 structures and unions. If the option @code{--no-frame-filters} is
27921 supplied, then Python frame filters will not be executed.
27923 If the @code{--skip-unavailable} option is specified, arguments that
27924 are not available are not listed. Partially available arguments
27925 are still displayed, however.
27927 Use of this command to obtain arguments in a single frame is
27928 deprecated in favor of the @samp{-stack-list-variables} command.
27930 @subsubheading @value{GDBN} Command
27932 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
27933 @samp{gdb_get_args} command which partially overlaps with the
27934 functionality of @samp{-stack-list-arguments}.
27936 @subsubheading Example
27943 frame=@{level="0",addr="0x00010734",func="callee4",
27944 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27945 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
27946 frame=@{level="1",addr="0x0001076c",func="callee3",
27947 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27948 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
27949 frame=@{level="2",addr="0x0001078c",func="callee2",
27950 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27951 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
27952 frame=@{level="3",addr="0x000107b4",func="callee1",
27953 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27954 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
27955 frame=@{level="4",addr="0x000107e0",func="main",
27956 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27957 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
27959 -stack-list-arguments 0
27962 frame=@{level="0",args=[]@},
27963 frame=@{level="1",args=[name="strarg"]@},
27964 frame=@{level="2",args=[name="intarg",name="strarg"]@},
27965 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
27966 frame=@{level="4",args=[]@}]
27968 -stack-list-arguments 1
27971 frame=@{level="0",args=[]@},
27973 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
27974 frame=@{level="2",args=[
27975 @{name="intarg",value="2"@},
27976 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
27977 @{frame=@{level="3",args=[
27978 @{name="intarg",value="2"@},
27979 @{name="strarg",value="0x11940 \"A string argument.\""@},
27980 @{name="fltarg",value="3.5"@}]@},
27981 frame=@{level="4",args=[]@}]
27983 -stack-list-arguments 0 2 2
27984 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
27986 -stack-list-arguments 1 2 2
27987 ^done,stack-args=[frame=@{level="2",
27988 args=[@{name="intarg",value="2"@},
27989 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
27993 @c @subheading -stack-list-exception-handlers
27996 @anchor{-stack-list-frames}
27997 @subheading The @code{-stack-list-frames} Command
27998 @findex -stack-list-frames
28000 @subsubheading Synopsis
28003 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
28006 List the frames currently on the stack. For each frame it displays the
28011 The frame number, 0 being the topmost frame, i.e., the innermost function.
28013 The @code{$pc} value for that frame.
28017 File name of the source file where the function lives.
28018 @item @var{fullname}
28019 The full file name of the source file where the function lives.
28021 Line number corresponding to the @code{$pc}.
28023 The shared library where this function is defined. This is only given
28024 if the frame's function is not known.
28027 If invoked without arguments, this command prints a backtrace for the
28028 whole stack. If given two integer arguments, it shows the frames whose
28029 levels are between the two arguments (inclusive). If the two arguments
28030 are equal, it shows the single frame at the corresponding level. It is
28031 an error if @var{low-frame} is larger than the actual number of
28032 frames. On the other hand, @var{high-frame} may be larger than the
28033 actual number of frames, in which case only existing frames will be
28034 returned. If the option @code{--no-frame-filters} is supplied, then
28035 Python frame filters will not be executed.
28037 @subsubheading @value{GDBN} Command
28039 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
28041 @subsubheading Example
28043 Full stack backtrace:
28049 [frame=@{level="0",addr="0x0001076c",func="foo",
28050 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
28051 frame=@{level="1",addr="0x000107a4",func="foo",
28052 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28053 frame=@{level="2",addr="0x000107a4",func="foo",
28054 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28055 frame=@{level="3",addr="0x000107a4",func="foo",
28056 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28057 frame=@{level="4",addr="0x000107a4",func="foo",
28058 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28059 frame=@{level="5",addr="0x000107a4",func="foo",
28060 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28061 frame=@{level="6",addr="0x000107a4",func="foo",
28062 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28063 frame=@{level="7",addr="0x000107a4",func="foo",
28064 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28065 frame=@{level="8",addr="0x000107a4",func="foo",
28066 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28067 frame=@{level="9",addr="0x000107a4",func="foo",
28068 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28069 frame=@{level="10",addr="0x000107a4",func="foo",
28070 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28071 frame=@{level="11",addr="0x00010738",func="main",
28072 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
28076 Show frames between @var{low_frame} and @var{high_frame}:
28080 -stack-list-frames 3 5
28082 [frame=@{level="3",addr="0x000107a4",func="foo",
28083 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28084 frame=@{level="4",addr="0x000107a4",func="foo",
28085 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28086 frame=@{level="5",addr="0x000107a4",func="foo",
28087 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
28091 Show a single frame:
28095 -stack-list-frames 3 3
28097 [frame=@{level="3",addr="0x000107a4",func="foo",
28098 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
28103 @subheading The @code{-stack-list-locals} Command
28104 @findex -stack-list-locals
28105 @anchor{-stack-list-locals}
28107 @subsubheading Synopsis
28110 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
28113 Display the local variable names for the selected frame. If
28114 @var{print-values} is 0 or @code{--no-values}, print only the names of
28115 the variables; if it is 1 or @code{--all-values}, print also their
28116 values; and if it is 2 or @code{--simple-values}, print the name,
28117 type and value for simple data types, and the name and type for arrays,
28118 structures and unions. In this last case, a frontend can immediately
28119 display the value of simple data types and create variable objects for
28120 other data types when the user wishes to explore their values in
28121 more detail. If the option @code{--no-frame-filters} is supplied, then
28122 Python frame filters will not be executed.
28124 If the @code{--skip-unavailable} option is specified, local variables
28125 that are not available are not listed. Partially available local
28126 variables are still displayed, however.
28128 This command is deprecated in favor of the
28129 @samp{-stack-list-variables} command.
28131 @subsubheading @value{GDBN} Command
28133 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
28135 @subsubheading Example
28139 -stack-list-locals 0
28140 ^done,locals=[name="A",name="B",name="C"]
28142 -stack-list-locals --all-values
28143 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
28144 @{name="C",value="@{1, 2, 3@}"@}]
28145 -stack-list-locals --simple-values
28146 ^done,locals=[@{name="A",type="int",value="1"@},
28147 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
28151 @anchor{-stack-list-variables}
28152 @subheading The @code{-stack-list-variables} Command
28153 @findex -stack-list-variables
28155 @subsubheading Synopsis
28158 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
28161 Display the names of local variables and function arguments for the selected frame. If
28162 @var{print-values} is 0 or @code{--no-values}, print only the names of
28163 the variables; if it is 1 or @code{--all-values}, print also their
28164 values; and if it is 2 or @code{--simple-values}, print the name,
28165 type and value for simple data types, and the name and type for arrays,
28166 structures and unions. If the option @code{--no-frame-filters} is
28167 supplied, then Python frame filters will not be executed.
28169 If the @code{--skip-unavailable} option is specified, local variables
28170 and arguments that are not available are not listed. Partially
28171 available arguments and local variables are still displayed, however.
28173 @subsubheading Example
28177 -stack-list-variables --thread 1 --frame 0 --all-values
28178 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
28183 @subheading The @code{-stack-select-frame} Command
28184 @findex -stack-select-frame
28186 @subsubheading Synopsis
28189 -stack-select-frame @var{framenum}
28192 Change the selected frame. Select a different frame @var{framenum} on
28195 This command in deprecated in favor of passing the @samp{--frame}
28196 option to every command.
28198 @subsubheading @value{GDBN} Command
28200 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
28201 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
28203 @subsubheading Example
28207 -stack-select-frame 2
28212 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28213 @node GDB/MI Variable Objects
28214 @section @sc{gdb/mi} Variable Objects
28218 @subheading Motivation for Variable Objects in @sc{gdb/mi}
28220 For the implementation of a variable debugger window (locals, watched
28221 expressions, etc.), we are proposing the adaptation of the existing code
28222 used by @code{Insight}.
28224 The two main reasons for that are:
28228 It has been proven in practice (it is already on its second generation).
28231 It will shorten development time (needless to say how important it is
28235 The original interface was designed to be used by Tcl code, so it was
28236 slightly changed so it could be used through @sc{gdb/mi}. This section
28237 describes the @sc{gdb/mi} operations that will be available and gives some
28238 hints about their use.
28240 @emph{Note}: In addition to the set of operations described here, we
28241 expect the @sc{gui} implementation of a variable window to require, at
28242 least, the following operations:
28245 @item @code{-gdb-show} @code{output-radix}
28246 @item @code{-stack-list-arguments}
28247 @item @code{-stack-list-locals}
28248 @item @code{-stack-select-frame}
28253 @subheading Introduction to Variable Objects
28255 @cindex variable objects in @sc{gdb/mi}
28257 Variable objects are "object-oriented" MI interface for examining and
28258 changing values of expressions. Unlike some other MI interfaces that
28259 work with expressions, variable objects are specifically designed for
28260 simple and efficient presentation in the frontend. A variable object
28261 is identified by string name. When a variable object is created, the
28262 frontend specifies the expression for that variable object. The
28263 expression can be a simple variable, or it can be an arbitrary complex
28264 expression, and can even involve CPU registers. After creating a
28265 variable object, the frontend can invoke other variable object
28266 operations---for example to obtain or change the value of a variable
28267 object, or to change display format.
28269 Variable objects have hierarchical tree structure. Any variable object
28270 that corresponds to a composite type, such as structure in C, has
28271 a number of child variable objects, for example corresponding to each
28272 element of a structure. A child variable object can itself have
28273 children, recursively. Recursion ends when we reach
28274 leaf variable objects, which always have built-in types. Child variable
28275 objects are created only by explicit request, so if a frontend
28276 is not interested in the children of a particular variable object, no
28277 child will be created.
28279 For a leaf variable object it is possible to obtain its value as a
28280 string, or set the value from a string. String value can be also
28281 obtained for a non-leaf variable object, but it's generally a string
28282 that only indicates the type of the object, and does not list its
28283 contents. Assignment to a non-leaf variable object is not allowed.
28285 A frontend does not need to read the values of all variable objects each time
28286 the program stops. Instead, MI provides an update command that lists all
28287 variable objects whose values has changed since the last update
28288 operation. This considerably reduces the amount of data that must
28289 be transferred to the frontend. As noted above, children variable
28290 objects are created on demand, and only leaf variable objects have a
28291 real value. As result, gdb will read target memory only for leaf
28292 variables that frontend has created.
28294 The automatic update is not always desirable. For example, a frontend
28295 might want to keep a value of some expression for future reference,
28296 and never update it. For another example, fetching memory is
28297 relatively slow for embedded targets, so a frontend might want
28298 to disable automatic update for the variables that are either not
28299 visible on the screen, or ``closed''. This is possible using so
28300 called ``frozen variable objects''. Such variable objects are never
28301 implicitly updated.
28303 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
28304 fixed variable object, the expression is parsed when the variable
28305 object is created, including associating identifiers to specific
28306 variables. The meaning of expression never changes. For a floating
28307 variable object the values of variables whose names appear in the
28308 expressions are re-evaluated every time in the context of the current
28309 frame. Consider this example:
28314 struct work_state state;
28321 If a fixed variable object for the @code{state} variable is created in
28322 this function, and we enter the recursive call, the variable
28323 object will report the value of @code{state} in the top-level
28324 @code{do_work} invocation. On the other hand, a floating variable
28325 object will report the value of @code{state} in the current frame.
28327 If an expression specified when creating a fixed variable object
28328 refers to a local variable, the variable object becomes bound to the
28329 thread and frame in which the variable object is created. When such
28330 variable object is updated, @value{GDBN} makes sure that the
28331 thread/frame combination the variable object is bound to still exists,
28332 and re-evaluates the variable object in context of that thread/frame.
28334 The following is the complete set of @sc{gdb/mi} operations defined to
28335 access this functionality:
28337 @multitable @columnfractions .4 .6
28338 @item @strong{Operation}
28339 @tab @strong{Description}
28341 @item @code{-enable-pretty-printing}
28342 @tab enable Python-based pretty-printing
28343 @item @code{-var-create}
28344 @tab create a variable object
28345 @item @code{-var-delete}
28346 @tab delete the variable object and/or its children
28347 @item @code{-var-set-format}
28348 @tab set the display format of this variable
28349 @item @code{-var-show-format}
28350 @tab show the display format of this variable
28351 @item @code{-var-info-num-children}
28352 @tab tells how many children this object has
28353 @item @code{-var-list-children}
28354 @tab return a list of the object's children
28355 @item @code{-var-info-type}
28356 @tab show the type of this variable object
28357 @item @code{-var-info-expression}
28358 @tab print parent-relative expression that this variable object represents
28359 @item @code{-var-info-path-expression}
28360 @tab print full expression that this variable object represents
28361 @item @code{-var-show-attributes}
28362 @tab is this variable editable? does it exist here?
28363 @item @code{-var-evaluate-expression}
28364 @tab get the value of this variable
28365 @item @code{-var-assign}
28366 @tab set the value of this variable
28367 @item @code{-var-update}
28368 @tab update the variable and its children
28369 @item @code{-var-set-frozen}
28370 @tab set frozeness attribute
28371 @item @code{-var-set-update-range}
28372 @tab set range of children to display on update
28375 In the next subsection we describe each operation in detail and suggest
28376 how it can be used.
28378 @subheading Description And Use of Operations on Variable Objects
28380 @subheading The @code{-enable-pretty-printing} Command
28381 @findex -enable-pretty-printing
28384 -enable-pretty-printing
28387 @value{GDBN} allows Python-based visualizers to affect the output of the
28388 MI variable object commands. However, because there was no way to
28389 implement this in a fully backward-compatible way, a front end must
28390 request that this functionality be enabled.
28392 Once enabled, this feature cannot be disabled.
28394 Note that if Python support has not been compiled into @value{GDBN},
28395 this command will still succeed (and do nothing).
28397 This feature is currently (as of @value{GDBN} 7.0) experimental, and
28398 may work differently in future versions of @value{GDBN}.
28400 @subheading The @code{-var-create} Command
28401 @findex -var-create
28403 @subsubheading Synopsis
28406 -var-create @{@var{name} | "-"@}
28407 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
28410 This operation creates a variable object, which allows the monitoring of
28411 a variable, the result of an expression, a memory cell or a CPU
28414 The @var{name} parameter is the string by which the object can be
28415 referenced. It must be unique. If @samp{-} is specified, the varobj
28416 system will generate a string ``varNNNNNN'' automatically. It will be
28417 unique provided that one does not specify @var{name} of that format.
28418 The command fails if a duplicate name is found.
28420 The frame under which the expression should be evaluated can be
28421 specified by @var{frame-addr}. A @samp{*} indicates that the current
28422 frame should be used. A @samp{@@} indicates that a floating variable
28423 object must be created.
28425 @var{expression} is any expression valid on the current language set (must not
28426 begin with a @samp{*}), or one of the following:
28430 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
28433 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
28436 @samp{$@var{regname}} --- a CPU register name
28439 @cindex dynamic varobj
28440 A varobj's contents may be provided by a Python-based pretty-printer. In this
28441 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
28442 have slightly different semantics in some cases. If the
28443 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
28444 will never create a dynamic varobj. This ensures backward
28445 compatibility for existing clients.
28447 @subsubheading Result
28449 This operation returns attributes of the newly-created varobj. These
28454 The name of the varobj.
28457 The number of children of the varobj. This number is not necessarily
28458 reliable for a dynamic varobj. Instead, you must examine the
28459 @samp{has_more} attribute.
28462 The varobj's scalar value. For a varobj whose type is some sort of
28463 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
28464 will not be interesting.
28467 The varobj's type. This is a string representation of the type, as
28468 would be printed by the @value{GDBN} CLI. If @samp{print object}
28469 (@pxref{Print Settings, set print object}) is set to @code{on}, the
28470 @emph{actual} (derived) type of the object is shown rather than the
28471 @emph{declared} one.
28474 If a variable object is bound to a specific thread, then this is the
28475 thread's identifier.
28478 For a dynamic varobj, this indicates whether there appear to be any
28479 children available. For a non-dynamic varobj, this will be 0.
28482 This attribute will be present and have the value @samp{1} if the
28483 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
28484 then this attribute will not be present.
28487 A dynamic varobj can supply a display hint to the front end. The
28488 value comes directly from the Python pretty-printer object's
28489 @code{display_hint} method. @xref{Pretty Printing API}.
28492 Typical output will look like this:
28495 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
28496 has_more="@var{has_more}"
28500 @subheading The @code{-var-delete} Command
28501 @findex -var-delete
28503 @subsubheading Synopsis
28506 -var-delete [ -c ] @var{name}
28509 Deletes a previously created variable object and all of its children.
28510 With the @samp{-c} option, just deletes the children.
28512 Returns an error if the object @var{name} is not found.
28515 @subheading The @code{-var-set-format} Command
28516 @findex -var-set-format
28518 @subsubheading Synopsis
28521 -var-set-format @var{name} @var{format-spec}
28524 Sets the output format for the value of the object @var{name} to be
28527 @anchor{-var-set-format}
28528 The syntax for the @var{format-spec} is as follows:
28531 @var{format-spec} @expansion{}
28532 @{binary | decimal | hexadecimal | octal | natural@}
28535 The natural format is the default format choosen automatically
28536 based on the variable type (like decimal for an @code{int}, hex
28537 for pointers, etc.).
28539 For a variable with children, the format is set only on the
28540 variable itself, and the children are not affected.
28542 @subheading The @code{-var-show-format} Command
28543 @findex -var-show-format
28545 @subsubheading Synopsis
28548 -var-show-format @var{name}
28551 Returns the format used to display the value of the object @var{name}.
28554 @var{format} @expansion{}
28559 @subheading The @code{-var-info-num-children} Command
28560 @findex -var-info-num-children
28562 @subsubheading Synopsis
28565 -var-info-num-children @var{name}
28568 Returns the number of children of a variable object @var{name}:
28574 Note that this number is not completely reliable for a dynamic varobj.
28575 It will return the current number of children, but more children may
28579 @subheading The @code{-var-list-children} Command
28580 @findex -var-list-children
28582 @subsubheading Synopsis
28585 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
28587 @anchor{-var-list-children}
28589 Return a list of the children of the specified variable object and
28590 create variable objects for them, if they do not already exist. With
28591 a single argument or if @var{print-values} has a value of 0 or
28592 @code{--no-values}, print only the names of the variables; if
28593 @var{print-values} is 1 or @code{--all-values}, also print their
28594 values; and if it is 2 or @code{--simple-values} print the name and
28595 value for simple data types and just the name for arrays, structures
28598 @var{from} and @var{to}, if specified, indicate the range of children
28599 to report. If @var{from} or @var{to} is less than zero, the range is
28600 reset and all children will be reported. Otherwise, children starting
28601 at @var{from} (zero-based) and up to and excluding @var{to} will be
28604 If a child range is requested, it will only affect the current call to
28605 @code{-var-list-children}, but not future calls to @code{-var-update}.
28606 For this, you must instead use @code{-var-set-update-range}. The
28607 intent of this approach is to enable a front end to implement any
28608 update approach it likes; for example, scrolling a view may cause the
28609 front end to request more children with @code{-var-list-children}, and
28610 then the front end could call @code{-var-set-update-range} with a
28611 different range to ensure that future updates are restricted to just
28614 For each child the following results are returned:
28619 Name of the variable object created for this child.
28622 The expression to be shown to the user by the front end to designate this child.
28623 For example this may be the name of a structure member.
28625 For a dynamic varobj, this value cannot be used to form an
28626 expression. There is no way to do this at all with a dynamic varobj.
28628 For C/C@t{++} structures there are several pseudo children returned to
28629 designate access qualifiers. For these pseudo children @var{exp} is
28630 @samp{public}, @samp{private}, or @samp{protected}. In this case the
28631 type and value are not present.
28633 A dynamic varobj will not report the access qualifying
28634 pseudo-children, regardless of the language. This information is not
28635 available at all with a dynamic varobj.
28638 Number of children this child has. For a dynamic varobj, this will be
28642 The type of the child. If @samp{print object}
28643 (@pxref{Print Settings, set print object}) is set to @code{on}, the
28644 @emph{actual} (derived) type of the object is shown rather than the
28645 @emph{declared} one.
28648 If values were requested, this is the value.
28651 If this variable object is associated with a thread, this is the thread id.
28652 Otherwise this result is not present.
28655 If the variable object is frozen, this variable will be present with a value of 1.
28658 A dynamic varobj can supply a display hint to the front end. The
28659 value comes directly from the Python pretty-printer object's
28660 @code{display_hint} method. @xref{Pretty Printing API}.
28663 This attribute will be present and have the value @samp{1} if the
28664 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
28665 then this attribute will not be present.
28669 The result may have its own attributes:
28673 A dynamic varobj can supply a display hint to the front end. The
28674 value comes directly from the Python pretty-printer object's
28675 @code{display_hint} method. @xref{Pretty Printing API}.
28678 This is an integer attribute which is nonzero if there are children
28679 remaining after the end of the selected range.
28682 @subsubheading Example
28686 -var-list-children n
28687 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
28688 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
28690 -var-list-children --all-values n
28691 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
28692 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
28696 @subheading The @code{-var-info-type} Command
28697 @findex -var-info-type
28699 @subsubheading Synopsis
28702 -var-info-type @var{name}
28705 Returns the type of the specified variable @var{name}. The type is
28706 returned as a string in the same format as it is output by the
28710 type=@var{typename}
28714 @subheading The @code{-var-info-expression} Command
28715 @findex -var-info-expression
28717 @subsubheading Synopsis
28720 -var-info-expression @var{name}
28723 Returns a string that is suitable for presenting this
28724 variable object in user interface. The string is generally
28725 not valid expression in the current language, and cannot be evaluated.
28727 For example, if @code{a} is an array, and variable object
28728 @code{A} was created for @code{a}, then we'll get this output:
28731 (gdb) -var-info-expression A.1
28732 ^done,lang="C",exp="1"
28736 Here, the value of @code{lang} is the language name, which can be
28737 found in @ref{Supported Languages}.
28739 Note that the output of the @code{-var-list-children} command also
28740 includes those expressions, so the @code{-var-info-expression} command
28743 @subheading The @code{-var-info-path-expression} Command
28744 @findex -var-info-path-expression
28746 @subsubheading Synopsis
28749 -var-info-path-expression @var{name}
28752 Returns an expression that can be evaluated in the current
28753 context and will yield the same value that a variable object has.
28754 Compare this with the @code{-var-info-expression} command, which
28755 result can be used only for UI presentation. Typical use of
28756 the @code{-var-info-path-expression} command is creating a
28757 watchpoint from a variable object.
28759 This command is currently not valid for children of a dynamic varobj,
28760 and will give an error when invoked on one.
28762 For example, suppose @code{C} is a C@t{++} class, derived from class
28763 @code{Base}, and that the @code{Base} class has a member called
28764 @code{m_size}. Assume a variable @code{c} is has the type of
28765 @code{C} and a variable object @code{C} was created for variable
28766 @code{c}. Then, we'll get this output:
28768 (gdb) -var-info-path-expression C.Base.public.m_size
28769 ^done,path_expr=((Base)c).m_size)
28772 @subheading The @code{-var-show-attributes} Command
28773 @findex -var-show-attributes
28775 @subsubheading Synopsis
28778 -var-show-attributes @var{name}
28781 List attributes of the specified variable object @var{name}:
28784 status=@var{attr} [ ( ,@var{attr} )* ]
28788 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
28790 @subheading The @code{-var-evaluate-expression} Command
28791 @findex -var-evaluate-expression
28793 @subsubheading Synopsis
28796 -var-evaluate-expression [-f @var{format-spec}] @var{name}
28799 Evaluates the expression that is represented by the specified variable
28800 object and returns its value as a string. The format of the string
28801 can be specified with the @samp{-f} option. The possible values of
28802 this option are the same as for @code{-var-set-format}
28803 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
28804 the current display format will be used. The current display format
28805 can be changed using the @code{-var-set-format} command.
28811 Note that one must invoke @code{-var-list-children} for a variable
28812 before the value of a child variable can be evaluated.
28814 @subheading The @code{-var-assign} Command
28815 @findex -var-assign
28817 @subsubheading Synopsis
28820 -var-assign @var{name} @var{expression}
28823 Assigns the value of @var{expression} to the variable object specified
28824 by @var{name}. The object must be @samp{editable}. If the variable's
28825 value is altered by the assign, the variable will show up in any
28826 subsequent @code{-var-update} list.
28828 @subsubheading Example
28836 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
28840 @subheading The @code{-var-update} Command
28841 @findex -var-update
28843 @subsubheading Synopsis
28846 -var-update [@var{print-values}] @{@var{name} | "*"@}
28849 Reevaluate the expressions corresponding to the variable object
28850 @var{name} and all its direct and indirect children, and return the
28851 list of variable objects whose values have changed; @var{name} must
28852 be a root variable object. Here, ``changed'' means that the result of
28853 @code{-var-evaluate-expression} before and after the
28854 @code{-var-update} is different. If @samp{*} is used as the variable
28855 object names, all existing variable objects are updated, except
28856 for frozen ones (@pxref{-var-set-frozen}). The option
28857 @var{print-values} determines whether both names and values, or just
28858 names are printed. The possible values of this option are the same
28859 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
28860 recommended to use the @samp{--all-values} option, to reduce the
28861 number of MI commands needed on each program stop.
28863 With the @samp{*} parameter, if a variable object is bound to a
28864 currently running thread, it will not be updated, without any
28867 If @code{-var-set-update-range} was previously used on a varobj, then
28868 only the selected range of children will be reported.
28870 @code{-var-update} reports all the changed varobjs in a tuple named
28873 Each item in the change list is itself a tuple holding:
28877 The name of the varobj.
28880 If values were requested for this update, then this field will be
28881 present and will hold the value of the varobj.
28884 @anchor{-var-update}
28885 This field is a string which may take one of three values:
28889 The variable object's current value is valid.
28892 The variable object does not currently hold a valid value but it may
28893 hold one in the future if its associated expression comes back into
28897 The variable object no longer holds a valid value.
28898 This can occur when the executable file being debugged has changed,
28899 either through recompilation or by using the @value{GDBN} @code{file}
28900 command. The front end should normally choose to delete these variable
28904 In the future new values may be added to this list so the front should
28905 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
28908 This is only present if the varobj is still valid. If the type
28909 changed, then this will be the string @samp{true}; otherwise it will
28912 When a varobj's type changes, its children are also likely to have
28913 become incorrect. Therefore, the varobj's children are automatically
28914 deleted when this attribute is @samp{true}. Also, the varobj's update
28915 range, when set using the @code{-var-set-update-range} command, is
28919 If the varobj's type changed, then this field will be present and will
28922 @item new_num_children
28923 For a dynamic varobj, if the number of children changed, or if the
28924 type changed, this will be the new number of children.
28926 The @samp{numchild} field in other varobj responses is generally not
28927 valid for a dynamic varobj -- it will show the number of children that
28928 @value{GDBN} knows about, but because dynamic varobjs lazily
28929 instantiate their children, this will not reflect the number of
28930 children which may be available.
28932 The @samp{new_num_children} attribute only reports changes to the
28933 number of children known by @value{GDBN}. This is the only way to
28934 detect whether an update has removed children (which necessarily can
28935 only happen at the end of the update range).
28938 The display hint, if any.
28941 This is an integer value, which will be 1 if there are more children
28942 available outside the varobj's update range.
28945 This attribute will be present and have the value @samp{1} if the
28946 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
28947 then this attribute will not be present.
28950 If new children were added to a dynamic varobj within the selected
28951 update range (as set by @code{-var-set-update-range}), then they will
28952 be listed in this attribute.
28955 @subsubheading Example
28962 -var-update --all-values var1
28963 ^done,changelist=[@{name="var1",value="3",in_scope="true",
28964 type_changed="false"@}]
28968 @subheading The @code{-var-set-frozen} Command
28969 @findex -var-set-frozen
28970 @anchor{-var-set-frozen}
28972 @subsubheading Synopsis
28975 -var-set-frozen @var{name} @var{flag}
28978 Set the frozenness flag on the variable object @var{name}. The
28979 @var{flag} parameter should be either @samp{1} to make the variable
28980 frozen or @samp{0} to make it unfrozen. If a variable object is
28981 frozen, then neither itself, nor any of its children, are
28982 implicitly updated by @code{-var-update} of
28983 a parent variable or by @code{-var-update *}. Only
28984 @code{-var-update} of the variable itself will update its value and
28985 values of its children. After a variable object is unfrozen, it is
28986 implicitly updated by all subsequent @code{-var-update} operations.
28987 Unfreezing a variable does not update it, only subsequent
28988 @code{-var-update} does.
28990 @subsubheading Example
28994 -var-set-frozen V 1
28999 @subheading The @code{-var-set-update-range} command
29000 @findex -var-set-update-range
29001 @anchor{-var-set-update-range}
29003 @subsubheading Synopsis
29006 -var-set-update-range @var{name} @var{from} @var{to}
29009 Set the range of children to be returned by future invocations of
29010 @code{-var-update}.
29012 @var{from} and @var{to} indicate the range of children to report. If
29013 @var{from} or @var{to} is less than zero, the range is reset and all
29014 children will be reported. Otherwise, children starting at @var{from}
29015 (zero-based) and up to and excluding @var{to} will be reported.
29017 @subsubheading Example
29021 -var-set-update-range V 1 2
29025 @subheading The @code{-var-set-visualizer} command
29026 @findex -var-set-visualizer
29027 @anchor{-var-set-visualizer}
29029 @subsubheading Synopsis
29032 -var-set-visualizer @var{name} @var{visualizer}
29035 Set a visualizer for the variable object @var{name}.
29037 @var{visualizer} is the visualizer to use. The special value
29038 @samp{None} means to disable any visualizer in use.
29040 If not @samp{None}, @var{visualizer} must be a Python expression.
29041 This expression must evaluate to a callable object which accepts a
29042 single argument. @value{GDBN} will call this object with the value of
29043 the varobj @var{name} as an argument (this is done so that the same
29044 Python pretty-printing code can be used for both the CLI and MI).
29045 When called, this object must return an object which conforms to the
29046 pretty-printing interface (@pxref{Pretty Printing API}).
29048 The pre-defined function @code{gdb.default_visualizer} may be used to
29049 select a visualizer by following the built-in process
29050 (@pxref{Selecting Pretty-Printers}). This is done automatically when
29051 a varobj is created, and so ordinarily is not needed.
29053 This feature is only available if Python support is enabled. The MI
29054 command @code{-list-features} (@pxref{GDB/MI Support Commands})
29055 can be used to check this.
29057 @subsubheading Example
29059 Resetting the visualizer:
29063 -var-set-visualizer V None
29067 Reselecting the default (type-based) visualizer:
29071 -var-set-visualizer V gdb.default_visualizer
29075 Suppose @code{SomeClass} is a visualizer class. A lambda expression
29076 can be used to instantiate this class for a varobj:
29080 -var-set-visualizer V "lambda val: SomeClass()"
29084 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29085 @node GDB/MI Data Manipulation
29086 @section @sc{gdb/mi} Data Manipulation
29088 @cindex data manipulation, in @sc{gdb/mi}
29089 @cindex @sc{gdb/mi}, data manipulation
29090 This section describes the @sc{gdb/mi} commands that manipulate data:
29091 examine memory and registers, evaluate expressions, etc.
29093 @c REMOVED FROM THE INTERFACE.
29094 @c @subheading -data-assign
29095 @c Change the value of a program variable. Plenty of side effects.
29096 @c @subsubheading GDB Command
29098 @c @subsubheading Example
29101 @subheading The @code{-data-disassemble} Command
29102 @findex -data-disassemble
29104 @subsubheading Synopsis
29108 [ -s @var{start-addr} -e @var{end-addr} ]
29109 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
29117 @item @var{start-addr}
29118 is the beginning address (or @code{$pc})
29119 @item @var{end-addr}
29121 @item @var{filename}
29122 is the name of the file to disassemble
29123 @item @var{linenum}
29124 is the line number to disassemble around
29126 is the number of disassembly lines to be produced. If it is -1,
29127 the whole function will be disassembled, in case no @var{end-addr} is
29128 specified. If @var{end-addr} is specified as a non-zero value, and
29129 @var{lines} is lower than the number of disassembly lines between
29130 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
29131 displayed; if @var{lines} is higher than the number of lines between
29132 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
29135 is either 0 (meaning only disassembly), 1 (meaning mixed source and
29136 disassembly), 2 (meaning disassembly with raw opcodes), or 3 (meaning
29137 mixed source and disassembly with raw opcodes).
29140 @subsubheading Result
29142 The result of the @code{-data-disassemble} command will be a list named
29143 @samp{asm_insns}, the contents of this list depend on the @var{mode}
29144 used with the @code{-data-disassemble} command.
29146 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
29151 The address at which this instruction was disassembled.
29154 The name of the function this instruction is within.
29157 The decimal offset in bytes from the start of @samp{func-name}.
29160 The text disassembly for this @samp{address}.
29163 This field is only present for mode 2. This contains the raw opcode
29164 bytes for the @samp{inst} field.
29168 For modes 1 and 3 the @samp{asm_insns} list contains tuples named
29169 @samp{src_and_asm_line}, each of which has the following fields:
29173 The line number within @samp{file}.
29176 The file name from the compilation unit. This might be an absolute
29177 file name or a relative file name depending on the compile command
29181 Absolute file name of @samp{file}. It is converted to a canonical form
29182 using the source file search path
29183 (@pxref{Source Path, ,Specifying Source Directories})
29184 and after resolving all the symbolic links.
29186 If the source file is not found this field will contain the path as
29187 present in the debug information.
29189 @item line_asm_insn
29190 This is a list of tuples containing the disassembly for @samp{line} in
29191 @samp{file}. The fields of each tuple are the same as for
29192 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
29193 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
29198 Note that whatever included in the @samp{inst} field, is not
29199 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
29202 @subsubheading @value{GDBN} Command
29204 The corresponding @value{GDBN} command is @samp{disassemble}.
29206 @subsubheading Example
29208 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
29212 -data-disassemble -s $pc -e "$pc + 20" -- 0
29215 @{address="0x000107c0",func-name="main",offset="4",
29216 inst="mov 2, %o0"@},
29217 @{address="0x000107c4",func-name="main",offset="8",
29218 inst="sethi %hi(0x11800), %o2"@},
29219 @{address="0x000107c8",func-name="main",offset="12",
29220 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
29221 @{address="0x000107cc",func-name="main",offset="16",
29222 inst="sethi %hi(0x11800), %o2"@},
29223 @{address="0x000107d0",func-name="main",offset="20",
29224 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
29228 Disassemble the whole @code{main} function. Line 32 is part of
29232 -data-disassemble -f basics.c -l 32 -- 0
29234 @{address="0x000107bc",func-name="main",offset="0",
29235 inst="save %sp, -112, %sp"@},
29236 @{address="0x000107c0",func-name="main",offset="4",
29237 inst="mov 2, %o0"@},
29238 @{address="0x000107c4",func-name="main",offset="8",
29239 inst="sethi %hi(0x11800), %o2"@},
29241 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
29242 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
29246 Disassemble 3 instructions from the start of @code{main}:
29250 -data-disassemble -f basics.c -l 32 -n 3 -- 0
29252 @{address="0x000107bc",func-name="main",offset="0",
29253 inst="save %sp, -112, %sp"@},
29254 @{address="0x000107c0",func-name="main",offset="4",
29255 inst="mov 2, %o0"@},
29256 @{address="0x000107c4",func-name="main",offset="8",
29257 inst="sethi %hi(0x11800), %o2"@}]
29261 Disassemble 3 instructions from the start of @code{main} in mixed mode:
29265 -data-disassemble -f basics.c -l 32 -n 3 -- 1
29267 src_and_asm_line=@{line="31",
29268 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
29269 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
29270 line_asm_insn=[@{address="0x000107bc",
29271 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
29272 src_and_asm_line=@{line="32",
29273 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
29274 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
29275 line_asm_insn=[@{address="0x000107c0",
29276 func-name="main",offset="4",inst="mov 2, %o0"@},
29277 @{address="0x000107c4",func-name="main",offset="8",
29278 inst="sethi %hi(0x11800), %o2"@}]@}]
29283 @subheading The @code{-data-evaluate-expression} Command
29284 @findex -data-evaluate-expression
29286 @subsubheading Synopsis
29289 -data-evaluate-expression @var{expr}
29292 Evaluate @var{expr} as an expression. The expression could contain an
29293 inferior function call. The function call will execute synchronously.
29294 If the expression contains spaces, it must be enclosed in double quotes.
29296 @subsubheading @value{GDBN} Command
29298 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
29299 @samp{call}. In @code{gdbtk} only, there's a corresponding
29300 @samp{gdb_eval} command.
29302 @subsubheading Example
29304 In the following example, the numbers that precede the commands are the
29305 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
29306 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
29310 211-data-evaluate-expression A
29313 311-data-evaluate-expression &A
29314 311^done,value="0xefffeb7c"
29316 411-data-evaluate-expression A+3
29319 511-data-evaluate-expression "A + 3"
29325 @subheading The @code{-data-list-changed-registers} Command
29326 @findex -data-list-changed-registers
29328 @subsubheading Synopsis
29331 -data-list-changed-registers
29334 Display a list of the registers that have changed.
29336 @subsubheading @value{GDBN} Command
29338 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
29339 has the corresponding command @samp{gdb_changed_register_list}.
29341 @subsubheading Example
29343 On a PPC MBX board:
29351 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
29352 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
29355 -data-list-changed-registers
29356 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
29357 "10","11","13","14","15","16","17","18","19","20","21","22","23",
29358 "24","25","26","27","28","30","31","64","65","66","67","69"]
29363 @subheading The @code{-data-list-register-names} Command
29364 @findex -data-list-register-names
29366 @subsubheading Synopsis
29369 -data-list-register-names [ ( @var{regno} )+ ]
29372 Show a list of register names for the current target. If no arguments
29373 are given, it shows a list of the names of all the registers. If
29374 integer numbers are given as arguments, it will print a list of the
29375 names of the registers corresponding to the arguments. To ensure
29376 consistency between a register name and its number, the output list may
29377 include empty register names.
29379 @subsubheading @value{GDBN} Command
29381 @value{GDBN} does not have a command which corresponds to
29382 @samp{-data-list-register-names}. In @code{gdbtk} there is a
29383 corresponding command @samp{gdb_regnames}.
29385 @subsubheading Example
29387 For the PPC MBX board:
29390 -data-list-register-names
29391 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
29392 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
29393 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
29394 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
29395 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
29396 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
29397 "", "pc","ps","cr","lr","ctr","xer"]
29399 -data-list-register-names 1 2 3
29400 ^done,register-names=["r1","r2","r3"]
29404 @subheading The @code{-data-list-register-values} Command
29405 @findex -data-list-register-values
29407 @subsubheading Synopsis
29410 -data-list-register-values
29411 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
29414 Display the registers' contents. The format according to which the
29415 registers' contents are to be returned is given by @var{fmt}, followed
29416 by an optional list of numbers specifying the registers to display. A
29417 missing list of numbers indicates that the contents of all the
29418 registers must be returned. The @code{--skip-unavailable} option
29419 indicates that only the available registers are to be returned.
29421 Allowed formats for @var{fmt} are:
29438 @subsubheading @value{GDBN} Command
29440 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
29441 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
29443 @subsubheading Example
29445 For a PPC MBX board (note: line breaks are for readability only, they
29446 don't appear in the actual output):
29450 -data-list-register-values r 64 65
29451 ^done,register-values=[@{number="64",value="0xfe00a300"@},
29452 @{number="65",value="0x00029002"@}]
29454 -data-list-register-values x
29455 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
29456 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
29457 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
29458 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
29459 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
29460 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
29461 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
29462 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
29463 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
29464 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
29465 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
29466 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
29467 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
29468 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
29469 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
29470 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
29471 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
29472 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
29473 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
29474 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
29475 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
29476 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
29477 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
29478 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
29479 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
29480 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
29481 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
29482 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
29483 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
29484 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
29485 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
29486 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
29487 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
29488 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
29489 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
29490 @{number="69",value="0x20002b03"@}]
29495 @subheading The @code{-data-read-memory} Command
29496 @findex -data-read-memory
29498 This command is deprecated, use @code{-data-read-memory-bytes} instead.
29500 @subsubheading Synopsis
29503 -data-read-memory [ -o @var{byte-offset} ]
29504 @var{address} @var{word-format} @var{word-size}
29505 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
29512 @item @var{address}
29513 An expression specifying the address of the first memory word to be
29514 read. Complex expressions containing embedded white space should be
29515 quoted using the C convention.
29517 @item @var{word-format}
29518 The format to be used to print the memory words. The notation is the
29519 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
29522 @item @var{word-size}
29523 The size of each memory word in bytes.
29525 @item @var{nr-rows}
29526 The number of rows in the output table.
29528 @item @var{nr-cols}
29529 The number of columns in the output table.
29532 If present, indicates that each row should include an @sc{ascii} dump. The
29533 value of @var{aschar} is used as a padding character when a byte is not a
29534 member of the printable @sc{ascii} character set (printable @sc{ascii}
29535 characters are those whose code is between 32 and 126, inclusively).
29537 @item @var{byte-offset}
29538 An offset to add to the @var{address} before fetching memory.
29541 This command displays memory contents as a table of @var{nr-rows} by
29542 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
29543 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
29544 (returned as @samp{total-bytes}). Should less than the requested number
29545 of bytes be returned by the target, the missing words are identified
29546 using @samp{N/A}. The number of bytes read from the target is returned
29547 in @samp{nr-bytes} and the starting address used to read memory in
29550 The address of the next/previous row or page is available in
29551 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
29554 @subsubheading @value{GDBN} Command
29556 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
29557 @samp{gdb_get_mem} memory read command.
29559 @subsubheading Example
29561 Read six bytes of memory starting at @code{bytes+6} but then offset by
29562 @code{-6} bytes. Format as three rows of two columns. One byte per
29563 word. Display each word in hex.
29567 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
29568 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
29569 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
29570 prev-page="0x0000138a",memory=[
29571 @{addr="0x00001390",data=["0x00","0x01"]@},
29572 @{addr="0x00001392",data=["0x02","0x03"]@},
29573 @{addr="0x00001394",data=["0x04","0x05"]@}]
29577 Read two bytes of memory starting at address @code{shorts + 64} and
29578 display as a single word formatted in decimal.
29582 5-data-read-memory shorts+64 d 2 1 1
29583 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
29584 next-row="0x00001512",prev-row="0x0000150e",
29585 next-page="0x00001512",prev-page="0x0000150e",memory=[
29586 @{addr="0x00001510",data=["128"]@}]
29590 Read thirty two bytes of memory starting at @code{bytes+16} and format
29591 as eight rows of four columns. Include a string encoding with @samp{x}
29592 used as the non-printable character.
29596 4-data-read-memory bytes+16 x 1 8 4 x
29597 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
29598 next-row="0x000013c0",prev-row="0x0000139c",
29599 next-page="0x000013c0",prev-page="0x00001380",memory=[
29600 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
29601 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
29602 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
29603 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
29604 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
29605 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
29606 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
29607 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
29611 @subheading The @code{-data-read-memory-bytes} Command
29612 @findex -data-read-memory-bytes
29614 @subsubheading Synopsis
29617 -data-read-memory-bytes [ -o @var{byte-offset} ]
29618 @var{address} @var{count}
29625 @item @var{address}
29626 An expression specifying the address of the first memory word to be
29627 read. Complex expressions containing embedded white space should be
29628 quoted using the C convention.
29631 The number of bytes to read. This should be an integer literal.
29633 @item @var{byte-offset}
29634 The offsets in bytes relative to @var{address} at which to start
29635 reading. This should be an integer literal. This option is provided
29636 so that a frontend is not required to first evaluate address and then
29637 perform address arithmetics itself.
29641 This command attempts to read all accessible memory regions in the
29642 specified range. First, all regions marked as unreadable in the memory
29643 map (if one is defined) will be skipped. @xref{Memory Region
29644 Attributes}. Second, @value{GDBN} will attempt to read the remaining
29645 regions. For each one, if reading full region results in an errors,
29646 @value{GDBN} will try to read a subset of the region.
29648 In general, every single byte in the region may be readable or not,
29649 and the only way to read every readable byte is to try a read at
29650 every address, which is not practical. Therefore, @value{GDBN} will
29651 attempt to read all accessible bytes at either beginning or the end
29652 of the region, using a binary division scheme. This heuristic works
29653 well for reading accross a memory map boundary. Note that if a region
29654 has a readable range that is neither at the beginning or the end,
29655 @value{GDBN} will not read it.
29657 The result record (@pxref{GDB/MI Result Records}) that is output of
29658 the command includes a field named @samp{memory} whose content is a
29659 list of tuples. Each tuple represent a successfully read memory block
29660 and has the following fields:
29664 The start address of the memory block, as hexadecimal literal.
29667 The end address of the memory block, as hexadecimal literal.
29670 The offset of the memory block, as hexadecimal literal, relative to
29671 the start address passed to @code{-data-read-memory-bytes}.
29674 The contents of the memory block, in hex.
29680 @subsubheading @value{GDBN} Command
29682 The corresponding @value{GDBN} command is @samp{x}.
29684 @subsubheading Example
29688 -data-read-memory-bytes &a 10
29689 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
29691 contents="01000000020000000300"@}]
29696 @subheading The @code{-data-write-memory-bytes} Command
29697 @findex -data-write-memory-bytes
29699 @subsubheading Synopsis
29702 -data-write-memory-bytes @var{address} @var{contents}
29703 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
29710 @item @var{address}
29711 An expression specifying the address of the first memory word to be
29712 read. Complex expressions containing embedded white space should be
29713 quoted using the C convention.
29715 @item @var{contents}
29716 The hex-encoded bytes to write.
29719 Optional argument indicating the number of bytes to be written. If @var{count}
29720 is greater than @var{contents}' length, @value{GDBN} will repeatedly
29721 write @var{contents} until it fills @var{count} bytes.
29725 @subsubheading @value{GDBN} Command
29727 There's no corresponding @value{GDBN} command.
29729 @subsubheading Example
29733 -data-write-memory-bytes &a "aabbccdd"
29740 -data-write-memory-bytes &a "aabbccdd" 16e
29745 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29746 @node GDB/MI Tracepoint Commands
29747 @section @sc{gdb/mi} Tracepoint Commands
29749 The commands defined in this section implement MI support for
29750 tracepoints. For detailed introduction, see @ref{Tracepoints}.
29752 @subheading The @code{-trace-find} Command
29753 @findex -trace-find
29755 @subsubheading Synopsis
29758 -trace-find @var{mode} [@var{parameters}@dots{}]
29761 Find a trace frame using criteria defined by @var{mode} and
29762 @var{parameters}. The following table lists permissible
29763 modes and their parameters. For details of operation, see @ref{tfind}.
29768 No parameters are required. Stops examining trace frames.
29771 An integer is required as parameter. Selects tracepoint frame with
29774 @item tracepoint-number
29775 An integer is required as parameter. Finds next
29776 trace frame that corresponds to tracepoint with the specified number.
29779 An address is required as parameter. Finds
29780 next trace frame that corresponds to any tracepoint at the specified
29783 @item pc-inside-range
29784 Two addresses are required as parameters. Finds next trace
29785 frame that corresponds to a tracepoint at an address inside the
29786 specified range. Both bounds are considered to be inside the range.
29788 @item pc-outside-range
29789 Two addresses are required as parameters. Finds
29790 next trace frame that corresponds to a tracepoint at an address outside
29791 the specified range. Both bounds are considered to be inside the range.
29794 Line specification is required as parameter. @xref{Specify Location}.
29795 Finds next trace frame that corresponds to a tracepoint at
29796 the specified location.
29800 If @samp{none} was passed as @var{mode}, the response does not
29801 have fields. Otherwise, the response may have the following fields:
29805 This field has either @samp{0} or @samp{1} as the value, depending
29806 on whether a matching tracepoint was found.
29809 The index of the found traceframe. This field is present iff
29810 the @samp{found} field has value of @samp{1}.
29813 The index of the found tracepoint. This field is present iff
29814 the @samp{found} field has value of @samp{1}.
29817 The information about the frame corresponding to the found trace
29818 frame. This field is present only if a trace frame was found.
29819 @xref{GDB/MI Frame Information}, for description of this field.
29823 @subsubheading @value{GDBN} Command
29825 The corresponding @value{GDBN} command is @samp{tfind}.
29827 @subheading -trace-define-variable
29828 @findex -trace-define-variable
29830 @subsubheading Synopsis
29833 -trace-define-variable @var{name} [ @var{value} ]
29836 Create trace variable @var{name} if it does not exist. If
29837 @var{value} is specified, sets the initial value of the specified
29838 trace variable to that value. Note that the @var{name} should start
29839 with the @samp{$} character.
29841 @subsubheading @value{GDBN} Command
29843 The corresponding @value{GDBN} command is @samp{tvariable}.
29845 @subheading The @code{-trace-frame-collected} Command
29846 @findex -trace-frame-collected
29848 @subsubheading Synopsis
29851 -trace-frame-collected
29852 [--var-print-values @var{var_pval}]
29853 [--comp-print-values @var{comp_pval}]
29854 [--registers-format @var{regformat}]
29855 [--memory-contents]
29858 This command returns the set of collected objects, register names,
29859 trace state variable names, memory ranges and computed expressions
29860 that have been collected at a particular trace frame. The optional
29861 parameters to the command affect the output format in different ways.
29862 See the output description table below for more details.
29864 The reported names can be used in the normal manner to create
29865 varobjs and inspect the objects themselves. The items returned by
29866 this command are categorized so that it is clear which is a variable,
29867 which is a register, which is a trace state variable, which is a
29868 memory range and which is a computed expression.
29870 For instance, if the actions were
29872 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
29873 collect *(int*)0xaf02bef0@@40
29877 the object collected in its entirety would be @code{myVar}. The
29878 object @code{myArray} would be partially collected, because only the
29879 element at index @code{myIndex} would be collected. The remaining
29880 objects would be computed expressions.
29882 An example output would be:
29886 -trace-frame-collected
29888 explicit-variables=[@{name="myVar",value="1"@}],
29889 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
29890 @{name="myObj.field",value="0"@},
29891 @{name="myPtr->field",value="1"@},
29892 @{name="myCount + 2",value="3"@},
29893 @{name="$tvar1 + 1",value="43970027"@}],
29894 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
29895 @{number="1",value="0x0"@},
29896 @{number="2",value="0x4"@},
29898 @{number="125",value="0x0"@}],
29899 tvars=[@{name="$tvar1",current="43970026"@}],
29900 memory=[@{address="0x0000000000602264",length="4"@},
29901 @{address="0x0000000000615bc0",length="4"@}]
29908 @item explicit-variables
29909 The set of objects that have been collected in their entirety (as
29910 opposed to collecting just a few elements of an array or a few struct
29911 members). For each object, its name and value are printed.
29912 The @code{--var-print-values} option affects how or whether the value
29913 field is output. If @var{var_pval} is 0, then print only the names;
29914 if it is 1, print also their values; and if it is 2, print the name,
29915 type and value for simple data types, and the name and type for
29916 arrays, structures and unions.
29918 @item computed-expressions
29919 The set of computed expressions that have been collected at the
29920 current trace frame. The @code{--comp-print-values} option affects
29921 this set like the @code{--var-print-values} option affects the
29922 @code{explicit-variables} set. See above.
29925 The registers that have been collected at the current trace frame.
29926 For each register collected, the name and current value are returned.
29927 The value is formatted according to the @code{--registers-format}
29928 option. See the @command{-data-list-register-values} command for a
29929 list of the allowed formats. The default is @samp{x}.
29932 The trace state variables that have been collected at the current
29933 trace frame. For each trace state variable collected, the name and
29934 current value are returned.
29937 The set of memory ranges that have been collected at the current trace
29938 frame. Its content is a list of tuples. Each tuple represents a
29939 collected memory range and has the following fields:
29943 The start address of the memory range, as hexadecimal literal.
29946 The length of the memory range, as decimal literal.
29949 The contents of the memory block, in hex. This field is only present
29950 if the @code{--memory-contents} option is specified.
29956 @subsubheading @value{GDBN} Command
29958 There is no corresponding @value{GDBN} command.
29960 @subsubheading Example
29962 @subheading -trace-list-variables
29963 @findex -trace-list-variables
29965 @subsubheading Synopsis
29968 -trace-list-variables
29971 Return a table of all defined trace variables. Each element of the
29972 table has the following fields:
29976 The name of the trace variable. This field is always present.
29979 The initial value. This is a 64-bit signed integer. This
29980 field is always present.
29983 The value the trace variable has at the moment. This is a 64-bit
29984 signed integer. This field is absent iff current value is
29985 not defined, for example if the trace was never run, or is
29990 @subsubheading @value{GDBN} Command
29992 The corresponding @value{GDBN} command is @samp{tvariables}.
29994 @subsubheading Example
29998 -trace-list-variables
29999 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
30000 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
30001 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
30002 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
30003 body=[variable=@{name="$trace_timestamp",initial="0"@}
30004 variable=@{name="$foo",initial="10",current="15"@}]@}
30008 @subheading -trace-save
30009 @findex -trace-save
30011 @subsubheading Synopsis
30014 -trace-save [-r ] @var{filename}
30017 Saves the collected trace data to @var{filename}. Without the
30018 @samp{-r} option, the data is downloaded from the target and saved
30019 in a local file. With the @samp{-r} option the target is asked
30020 to perform the save.
30022 @subsubheading @value{GDBN} Command
30024 The corresponding @value{GDBN} command is @samp{tsave}.
30027 @subheading -trace-start
30028 @findex -trace-start
30030 @subsubheading Synopsis
30036 Starts a tracing experiments. The result of this command does not
30039 @subsubheading @value{GDBN} Command
30041 The corresponding @value{GDBN} command is @samp{tstart}.
30043 @subheading -trace-status
30044 @findex -trace-status
30046 @subsubheading Synopsis
30052 Obtains the status of a tracing experiment. The result may include
30053 the following fields:
30058 May have a value of either @samp{0}, when no tracing operations are
30059 supported, @samp{1}, when all tracing operations are supported, or
30060 @samp{file} when examining trace file. In the latter case, examining
30061 of trace frame is possible but new tracing experiement cannot be
30062 started. This field is always present.
30065 May have a value of either @samp{0} or @samp{1} depending on whether
30066 tracing experiement is in progress on target. This field is present
30067 if @samp{supported} field is not @samp{0}.
30070 Report the reason why the tracing was stopped last time. This field
30071 may be absent iff tracing was never stopped on target yet. The
30072 value of @samp{request} means the tracing was stopped as result of
30073 the @code{-trace-stop} command. The value of @samp{overflow} means
30074 the tracing buffer is full. The value of @samp{disconnection} means
30075 tracing was automatically stopped when @value{GDBN} has disconnected.
30076 The value of @samp{passcount} means tracing was stopped when a
30077 tracepoint was passed a maximal number of times for that tracepoint.
30078 This field is present if @samp{supported} field is not @samp{0}.
30080 @item stopping-tracepoint
30081 The number of tracepoint whose passcount as exceeded. This field is
30082 present iff the @samp{stop-reason} field has the value of
30086 @itemx frames-created
30087 The @samp{frames} field is a count of the total number of trace frames
30088 in the trace buffer, while @samp{frames-created} is the total created
30089 during the run, including ones that were discarded, such as when a
30090 circular trace buffer filled up. Both fields are optional.
30094 These fields tell the current size of the tracing buffer and the
30095 remaining space. These fields are optional.
30098 The value of the circular trace buffer flag. @code{1} means that the
30099 trace buffer is circular and old trace frames will be discarded if
30100 necessary to make room, @code{0} means that the trace buffer is linear
30104 The value of the disconnected tracing flag. @code{1} means that
30105 tracing will continue after @value{GDBN} disconnects, @code{0} means
30106 that the trace run will stop.
30109 The filename of the trace file being examined. This field is
30110 optional, and only present when examining a trace file.
30114 @subsubheading @value{GDBN} Command
30116 The corresponding @value{GDBN} command is @samp{tstatus}.
30118 @subheading -trace-stop
30119 @findex -trace-stop
30121 @subsubheading Synopsis
30127 Stops a tracing experiment. The result of this command has the same
30128 fields as @code{-trace-status}, except that the @samp{supported} and
30129 @samp{running} fields are not output.
30131 @subsubheading @value{GDBN} Command
30133 The corresponding @value{GDBN} command is @samp{tstop}.
30136 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30137 @node GDB/MI Symbol Query
30138 @section @sc{gdb/mi} Symbol Query Commands
30142 @subheading The @code{-symbol-info-address} Command
30143 @findex -symbol-info-address
30145 @subsubheading Synopsis
30148 -symbol-info-address @var{symbol}
30151 Describe where @var{symbol} is stored.
30153 @subsubheading @value{GDBN} Command
30155 The corresponding @value{GDBN} command is @samp{info address}.
30157 @subsubheading Example
30161 @subheading The @code{-symbol-info-file} Command
30162 @findex -symbol-info-file
30164 @subsubheading Synopsis
30170 Show the file for the symbol.
30172 @subsubheading @value{GDBN} Command
30174 There's no equivalent @value{GDBN} command. @code{gdbtk} has
30175 @samp{gdb_find_file}.
30177 @subsubheading Example
30181 @subheading The @code{-symbol-info-function} Command
30182 @findex -symbol-info-function
30184 @subsubheading Synopsis
30187 -symbol-info-function
30190 Show which function the symbol lives in.
30192 @subsubheading @value{GDBN} Command
30194 @samp{gdb_get_function} in @code{gdbtk}.
30196 @subsubheading Example
30200 @subheading The @code{-symbol-info-line} Command
30201 @findex -symbol-info-line
30203 @subsubheading Synopsis
30209 Show the core addresses of the code for a source line.
30211 @subsubheading @value{GDBN} Command
30213 The corresponding @value{GDBN} command is @samp{info line}.
30214 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
30216 @subsubheading Example
30220 @subheading The @code{-symbol-info-symbol} Command
30221 @findex -symbol-info-symbol
30223 @subsubheading Synopsis
30226 -symbol-info-symbol @var{addr}
30229 Describe what symbol is at location @var{addr}.
30231 @subsubheading @value{GDBN} Command
30233 The corresponding @value{GDBN} command is @samp{info symbol}.
30235 @subsubheading Example
30239 @subheading The @code{-symbol-list-functions} Command
30240 @findex -symbol-list-functions
30242 @subsubheading Synopsis
30245 -symbol-list-functions
30248 List the functions in the executable.
30250 @subsubheading @value{GDBN} Command
30252 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
30253 @samp{gdb_search} in @code{gdbtk}.
30255 @subsubheading Example
30260 @subheading The @code{-symbol-list-lines} Command
30261 @findex -symbol-list-lines
30263 @subsubheading Synopsis
30266 -symbol-list-lines @var{filename}
30269 Print the list of lines that contain code and their associated program
30270 addresses for the given source filename. The entries are sorted in
30271 ascending PC order.
30273 @subsubheading @value{GDBN} Command
30275 There is no corresponding @value{GDBN} command.
30277 @subsubheading Example
30280 -symbol-list-lines basics.c
30281 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
30287 @subheading The @code{-symbol-list-types} Command
30288 @findex -symbol-list-types
30290 @subsubheading Synopsis
30296 List all the type names.
30298 @subsubheading @value{GDBN} Command
30300 The corresponding commands are @samp{info types} in @value{GDBN},
30301 @samp{gdb_search} in @code{gdbtk}.
30303 @subsubheading Example
30307 @subheading The @code{-symbol-list-variables} Command
30308 @findex -symbol-list-variables
30310 @subsubheading Synopsis
30313 -symbol-list-variables
30316 List all the global and static variable names.
30318 @subsubheading @value{GDBN} Command
30320 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
30322 @subsubheading Example
30326 @subheading The @code{-symbol-locate} Command
30327 @findex -symbol-locate
30329 @subsubheading Synopsis
30335 @subsubheading @value{GDBN} Command
30337 @samp{gdb_loc} in @code{gdbtk}.
30339 @subsubheading Example
30343 @subheading The @code{-symbol-type} Command
30344 @findex -symbol-type
30346 @subsubheading Synopsis
30349 -symbol-type @var{variable}
30352 Show type of @var{variable}.
30354 @subsubheading @value{GDBN} Command
30356 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
30357 @samp{gdb_obj_variable}.
30359 @subsubheading Example
30364 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30365 @node GDB/MI File Commands
30366 @section @sc{gdb/mi} File Commands
30368 This section describes the GDB/MI commands to specify executable file names
30369 and to read in and obtain symbol table information.
30371 @subheading The @code{-file-exec-and-symbols} Command
30372 @findex -file-exec-and-symbols
30374 @subsubheading Synopsis
30377 -file-exec-and-symbols @var{file}
30380 Specify the executable file to be debugged. This file is the one from
30381 which the symbol table is also read. If no file is specified, the
30382 command clears the executable and symbol information. If breakpoints
30383 are set when using this command with no arguments, @value{GDBN} will produce
30384 error messages. Otherwise, no output is produced, except a completion
30387 @subsubheading @value{GDBN} Command
30389 The corresponding @value{GDBN} command is @samp{file}.
30391 @subsubheading Example
30395 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
30401 @subheading The @code{-file-exec-file} Command
30402 @findex -file-exec-file
30404 @subsubheading Synopsis
30407 -file-exec-file @var{file}
30410 Specify the executable file to be debugged. Unlike
30411 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
30412 from this file. If used without argument, @value{GDBN} clears the information
30413 about the executable file. No output is produced, except a completion
30416 @subsubheading @value{GDBN} Command
30418 The corresponding @value{GDBN} command is @samp{exec-file}.
30420 @subsubheading Example
30424 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
30431 @subheading The @code{-file-list-exec-sections} Command
30432 @findex -file-list-exec-sections
30434 @subsubheading Synopsis
30437 -file-list-exec-sections
30440 List the sections of the current executable file.
30442 @subsubheading @value{GDBN} Command
30444 The @value{GDBN} command @samp{info file} shows, among the rest, the same
30445 information as this command. @code{gdbtk} has a corresponding command
30446 @samp{gdb_load_info}.
30448 @subsubheading Example
30453 @subheading The @code{-file-list-exec-source-file} Command
30454 @findex -file-list-exec-source-file
30456 @subsubheading Synopsis
30459 -file-list-exec-source-file
30462 List the line number, the current source file, and the absolute path
30463 to the current source file for the current executable. The macro
30464 information field has a value of @samp{1} or @samp{0} depending on
30465 whether or not the file includes preprocessor macro information.
30467 @subsubheading @value{GDBN} Command
30469 The @value{GDBN} equivalent is @samp{info source}
30471 @subsubheading Example
30475 123-file-list-exec-source-file
30476 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
30481 @subheading The @code{-file-list-exec-source-files} Command
30482 @findex -file-list-exec-source-files
30484 @subsubheading Synopsis
30487 -file-list-exec-source-files
30490 List the source files for the current executable.
30492 It will always output both the filename and fullname (absolute file
30493 name) of a source file.
30495 @subsubheading @value{GDBN} Command
30497 The @value{GDBN} equivalent is @samp{info sources}.
30498 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
30500 @subsubheading Example
30503 -file-list-exec-source-files
30505 @{file=foo.c,fullname=/home/foo.c@},
30506 @{file=/home/bar.c,fullname=/home/bar.c@},
30507 @{file=gdb_could_not_find_fullpath.c@}]
30512 @subheading The @code{-file-list-shared-libraries} Command
30513 @findex -file-list-shared-libraries
30515 @subsubheading Synopsis
30518 -file-list-shared-libraries
30521 List the shared libraries in the program.
30523 @subsubheading @value{GDBN} Command
30525 The corresponding @value{GDBN} command is @samp{info shared}.
30527 @subsubheading Example
30531 @subheading The @code{-file-list-symbol-files} Command
30532 @findex -file-list-symbol-files
30534 @subsubheading Synopsis
30537 -file-list-symbol-files
30542 @subsubheading @value{GDBN} Command
30544 The corresponding @value{GDBN} command is @samp{info file} (part of it).
30546 @subsubheading Example
30551 @subheading The @code{-file-symbol-file} Command
30552 @findex -file-symbol-file
30554 @subsubheading Synopsis
30557 -file-symbol-file @var{file}
30560 Read symbol table info from the specified @var{file} argument. When
30561 used without arguments, clears @value{GDBN}'s symbol table info. No output is
30562 produced, except for a completion notification.
30564 @subsubheading @value{GDBN} Command
30566 The corresponding @value{GDBN} command is @samp{symbol-file}.
30568 @subsubheading Example
30572 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
30578 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30579 @node GDB/MI Memory Overlay Commands
30580 @section @sc{gdb/mi} Memory Overlay Commands
30582 The memory overlay commands are not implemented.
30584 @c @subheading -overlay-auto
30586 @c @subheading -overlay-list-mapping-state
30588 @c @subheading -overlay-list-overlays
30590 @c @subheading -overlay-map
30592 @c @subheading -overlay-off
30594 @c @subheading -overlay-on
30596 @c @subheading -overlay-unmap
30598 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30599 @node GDB/MI Signal Handling Commands
30600 @section @sc{gdb/mi} Signal Handling Commands
30602 Signal handling commands are not implemented.
30604 @c @subheading -signal-handle
30606 @c @subheading -signal-list-handle-actions
30608 @c @subheading -signal-list-signal-types
30612 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30613 @node GDB/MI Target Manipulation
30614 @section @sc{gdb/mi} Target Manipulation Commands
30617 @subheading The @code{-target-attach} Command
30618 @findex -target-attach
30620 @subsubheading Synopsis
30623 -target-attach @var{pid} | @var{gid} | @var{file}
30626 Attach to a process @var{pid} or a file @var{file} outside of
30627 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
30628 group, the id previously returned by
30629 @samp{-list-thread-groups --available} must be used.
30631 @subsubheading @value{GDBN} Command
30633 The corresponding @value{GDBN} command is @samp{attach}.
30635 @subsubheading Example
30639 =thread-created,id="1"
30640 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
30646 @subheading The @code{-target-compare-sections} Command
30647 @findex -target-compare-sections
30649 @subsubheading Synopsis
30652 -target-compare-sections [ @var{section} ]
30655 Compare data of section @var{section} on target to the exec file.
30656 Without the argument, all sections are compared.
30658 @subsubheading @value{GDBN} Command
30660 The @value{GDBN} equivalent is @samp{compare-sections}.
30662 @subsubheading Example
30667 @subheading The @code{-target-detach} Command
30668 @findex -target-detach
30670 @subsubheading Synopsis
30673 -target-detach [ @var{pid} | @var{gid} ]
30676 Detach from the remote target which normally resumes its execution.
30677 If either @var{pid} or @var{gid} is specified, detaches from either
30678 the specified process, or specified thread group. There's no output.
30680 @subsubheading @value{GDBN} Command
30682 The corresponding @value{GDBN} command is @samp{detach}.
30684 @subsubheading Example
30694 @subheading The @code{-target-disconnect} Command
30695 @findex -target-disconnect
30697 @subsubheading Synopsis
30703 Disconnect from the remote target. There's no output and the target is
30704 generally not resumed.
30706 @subsubheading @value{GDBN} Command
30708 The corresponding @value{GDBN} command is @samp{disconnect}.
30710 @subsubheading Example
30720 @subheading The @code{-target-download} Command
30721 @findex -target-download
30723 @subsubheading Synopsis
30729 Loads the executable onto the remote target.
30730 It prints out an update message every half second, which includes the fields:
30734 The name of the section.
30736 The size of what has been sent so far for that section.
30738 The size of the section.
30740 The total size of what was sent so far (the current and the previous sections).
30742 The size of the overall executable to download.
30746 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
30747 @sc{gdb/mi} Output Syntax}).
30749 In addition, it prints the name and size of the sections, as they are
30750 downloaded. These messages include the following fields:
30754 The name of the section.
30756 The size of the section.
30758 The size of the overall executable to download.
30762 At the end, a summary is printed.
30764 @subsubheading @value{GDBN} Command
30766 The corresponding @value{GDBN} command is @samp{load}.
30768 @subsubheading Example
30770 Note: each status message appears on a single line. Here the messages
30771 have been broken down so that they can fit onto a page.
30776 +download,@{section=".text",section-size="6668",total-size="9880"@}
30777 +download,@{section=".text",section-sent="512",section-size="6668",
30778 total-sent="512",total-size="9880"@}
30779 +download,@{section=".text",section-sent="1024",section-size="6668",
30780 total-sent="1024",total-size="9880"@}
30781 +download,@{section=".text",section-sent="1536",section-size="6668",
30782 total-sent="1536",total-size="9880"@}
30783 +download,@{section=".text",section-sent="2048",section-size="6668",
30784 total-sent="2048",total-size="9880"@}
30785 +download,@{section=".text",section-sent="2560",section-size="6668",
30786 total-sent="2560",total-size="9880"@}
30787 +download,@{section=".text",section-sent="3072",section-size="6668",
30788 total-sent="3072",total-size="9880"@}
30789 +download,@{section=".text",section-sent="3584",section-size="6668",
30790 total-sent="3584",total-size="9880"@}
30791 +download,@{section=".text",section-sent="4096",section-size="6668",
30792 total-sent="4096",total-size="9880"@}
30793 +download,@{section=".text",section-sent="4608",section-size="6668",
30794 total-sent="4608",total-size="9880"@}
30795 +download,@{section=".text",section-sent="5120",section-size="6668",
30796 total-sent="5120",total-size="9880"@}
30797 +download,@{section=".text",section-sent="5632",section-size="6668",
30798 total-sent="5632",total-size="9880"@}
30799 +download,@{section=".text",section-sent="6144",section-size="6668",
30800 total-sent="6144",total-size="9880"@}
30801 +download,@{section=".text",section-sent="6656",section-size="6668",
30802 total-sent="6656",total-size="9880"@}
30803 +download,@{section=".init",section-size="28",total-size="9880"@}
30804 +download,@{section=".fini",section-size="28",total-size="9880"@}
30805 +download,@{section=".data",section-size="3156",total-size="9880"@}
30806 +download,@{section=".data",section-sent="512",section-size="3156",
30807 total-sent="7236",total-size="9880"@}
30808 +download,@{section=".data",section-sent="1024",section-size="3156",
30809 total-sent="7748",total-size="9880"@}
30810 +download,@{section=".data",section-sent="1536",section-size="3156",
30811 total-sent="8260",total-size="9880"@}
30812 +download,@{section=".data",section-sent="2048",section-size="3156",
30813 total-sent="8772",total-size="9880"@}
30814 +download,@{section=".data",section-sent="2560",section-size="3156",
30815 total-sent="9284",total-size="9880"@}
30816 +download,@{section=".data",section-sent="3072",section-size="3156",
30817 total-sent="9796",total-size="9880"@}
30818 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
30825 @subheading The @code{-target-exec-status} Command
30826 @findex -target-exec-status
30828 @subsubheading Synopsis
30831 -target-exec-status
30834 Provide information on the state of the target (whether it is running or
30835 not, for instance).
30837 @subsubheading @value{GDBN} Command
30839 There's no equivalent @value{GDBN} command.
30841 @subsubheading Example
30845 @subheading The @code{-target-list-available-targets} Command
30846 @findex -target-list-available-targets
30848 @subsubheading Synopsis
30851 -target-list-available-targets
30854 List the possible targets to connect to.
30856 @subsubheading @value{GDBN} Command
30858 The corresponding @value{GDBN} command is @samp{help target}.
30860 @subsubheading Example
30864 @subheading The @code{-target-list-current-targets} Command
30865 @findex -target-list-current-targets
30867 @subsubheading Synopsis
30870 -target-list-current-targets
30873 Describe the current target.
30875 @subsubheading @value{GDBN} Command
30877 The corresponding information is printed by @samp{info file} (among
30880 @subsubheading Example
30884 @subheading The @code{-target-list-parameters} Command
30885 @findex -target-list-parameters
30887 @subsubheading Synopsis
30890 -target-list-parameters
30896 @subsubheading @value{GDBN} Command
30900 @subsubheading Example
30904 @subheading The @code{-target-select} Command
30905 @findex -target-select
30907 @subsubheading Synopsis
30910 -target-select @var{type} @var{parameters @dots{}}
30913 Connect @value{GDBN} to the remote target. This command takes two args:
30917 The type of target, for instance @samp{remote}, etc.
30918 @item @var{parameters}
30919 Device names, host names and the like. @xref{Target Commands, ,
30920 Commands for Managing Targets}, for more details.
30923 The output is a connection notification, followed by the address at
30924 which the target program is, in the following form:
30927 ^connected,addr="@var{address}",func="@var{function name}",
30928 args=[@var{arg list}]
30931 @subsubheading @value{GDBN} Command
30933 The corresponding @value{GDBN} command is @samp{target}.
30935 @subsubheading Example
30939 -target-select remote /dev/ttya
30940 ^connected,addr="0xfe00a300",func="??",args=[]
30944 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30945 @node GDB/MI File Transfer Commands
30946 @section @sc{gdb/mi} File Transfer Commands
30949 @subheading The @code{-target-file-put} Command
30950 @findex -target-file-put
30952 @subsubheading Synopsis
30955 -target-file-put @var{hostfile} @var{targetfile}
30958 Copy file @var{hostfile} from the host system (the machine running
30959 @value{GDBN}) to @var{targetfile} on the target system.
30961 @subsubheading @value{GDBN} Command
30963 The corresponding @value{GDBN} command is @samp{remote put}.
30965 @subsubheading Example
30969 -target-file-put localfile remotefile
30975 @subheading The @code{-target-file-get} Command
30976 @findex -target-file-get
30978 @subsubheading Synopsis
30981 -target-file-get @var{targetfile} @var{hostfile}
30984 Copy file @var{targetfile} from the target system to @var{hostfile}
30985 on the host system.
30987 @subsubheading @value{GDBN} Command
30989 The corresponding @value{GDBN} command is @samp{remote get}.
30991 @subsubheading Example
30995 -target-file-get remotefile localfile
31001 @subheading The @code{-target-file-delete} Command
31002 @findex -target-file-delete
31004 @subsubheading Synopsis
31007 -target-file-delete @var{targetfile}
31010 Delete @var{targetfile} from the target system.
31012 @subsubheading @value{GDBN} Command
31014 The corresponding @value{GDBN} command is @samp{remote delete}.
31016 @subsubheading Example
31020 -target-file-delete remotefile
31026 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31027 @node GDB/MI Ada Exceptions Commands
31028 @section Ada Exceptions @sc{gdb/mi} Commands
31030 @subheading The @code{-info-ada-exceptions} Command
31031 @findex -info-ada-exceptions
31033 @subsubheading Synopsis
31036 -info-ada-exceptions [ @var{regexp}]
31039 List all Ada exceptions defined within the program being debugged.
31040 With a regular expression @var{regexp}, only those exceptions whose
31041 names match @var{regexp} are listed.
31043 @subsubheading @value{GDBN} Command
31045 The corresponding @value{GDBN} command is @samp{info exceptions}.
31047 @subsubheading Result
31049 The result is a table of Ada exceptions. The following columns are
31050 defined for each exception:
31054 The name of the exception.
31057 The address of the exception.
31061 @subsubheading Example
31064 -info-ada-exceptions aint
31065 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
31066 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
31067 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
31068 body=[@{name="constraint_error",address="0x0000000000613da0"@},
31069 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
31072 @subheading Catching Ada Exceptions
31074 The commands describing how to ask @value{GDBN} to stop when a program
31075 raises an exception are described at @ref{Ada Exception GDB/MI
31076 Catchpoint Commands}.
31079 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31080 @node GDB/MI Support Commands
31081 @section @sc{gdb/mi} Support Commands
31083 Since new commands and features get regularly added to @sc{gdb/mi},
31084 some commands are available to help front-ends query the debugger
31085 about support for these capabilities. Similarly, it is also possible
31086 to query @value{GDBN} about target support of certain features.
31088 @subheading The @code{-info-gdb-mi-command} Command
31089 @cindex @code{-info-gdb-mi-command}
31090 @findex -info-gdb-mi-command
31092 @subsubheading Synopsis
31095 -info-gdb-mi-command @var{cmd_name}
31098 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
31100 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
31101 is technically not part of the command name (@pxref{GDB/MI Input
31102 Syntax}), and thus should be omitted in @var{cmd_name}. However,
31103 for ease of use, this command also accepts the form with the leading
31106 @subsubheading @value{GDBN} Command
31108 There is no corresponding @value{GDBN} command.
31110 @subsubheading Result
31112 The result is a tuple. There is currently only one field:
31116 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
31117 @code{"false"} otherwise.
31121 @subsubheading Example
31123 Here is an example where the @sc{gdb/mi} command does not exist:
31126 -info-gdb-mi-command unsupported-command
31127 ^done,command=@{exists="false"@}
31131 And here is an example where the @sc{gdb/mi} command is known
31135 -info-gdb-mi-command symbol-list-lines
31136 ^done,command=@{exists="true"@}
31139 @subheading The @code{-list-features} Command
31140 @findex -list-features
31141 @cindex supported @sc{gdb/mi} features, list
31143 Returns a list of particular features of the MI protocol that
31144 this version of gdb implements. A feature can be a command,
31145 or a new field in an output of some command, or even an
31146 important bugfix. While a frontend can sometimes detect presence
31147 of a feature at runtime, it is easier to perform detection at debugger
31150 The command returns a list of strings, with each string naming an
31151 available feature. Each returned string is just a name, it does not
31152 have any internal structure. The list of possible feature names
31158 (gdb) -list-features
31159 ^done,result=["feature1","feature2"]
31162 The current list of features is:
31165 @item frozen-varobjs
31166 Indicates support for the @code{-var-set-frozen} command, as well
31167 as possible presense of the @code{frozen} field in the output
31168 of @code{-varobj-create}.
31169 @item pending-breakpoints
31170 Indicates support for the @option{-f} option to the @code{-break-insert}
31173 Indicates Python scripting support, Python-based
31174 pretty-printing commands, and possible presence of the
31175 @samp{display_hint} field in the output of @code{-var-list-children}
31177 Indicates support for the @code{-thread-info} command.
31178 @item data-read-memory-bytes
31179 Indicates support for the @code{-data-read-memory-bytes} and the
31180 @code{-data-write-memory-bytes} commands.
31181 @item breakpoint-notifications
31182 Indicates that changes to breakpoints and breakpoints created via the
31183 CLI will be announced via async records.
31184 @item ada-task-info
31185 Indicates support for the @code{-ada-task-info} command.
31186 @item language-option
31187 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
31188 option (@pxref{Context management}).
31189 @item info-gdb-mi-command
31190 Indicates support for the @code{-info-gdb-mi-command} command.
31191 @item undefined-command-error-code
31192 Indicates support for the "undefined-command" error code in error result
31193 records, produced when trying to execute an undefined @sc{gdb/mi} command
31194 (@pxref{GDB/MI Result Records}).
31195 @item exec-run-start-option
31196 Indicates that the @code{-exec-run} command supports the @option{--start}
31197 option (@pxref{GDB/MI Program Execution}).
31200 @subheading The @code{-list-target-features} Command
31201 @findex -list-target-features
31203 Returns a list of particular features that are supported by the
31204 target. Those features affect the permitted MI commands, but
31205 unlike the features reported by the @code{-list-features} command, the
31206 features depend on which target GDB is using at the moment. Whenever
31207 a target can change, due to commands such as @code{-target-select},
31208 @code{-target-attach} or @code{-exec-run}, the list of target features
31209 may change, and the frontend should obtain it again.
31213 (gdb) -list-target-features
31214 ^done,result=["async"]
31217 The current list of features is:
31221 Indicates that the target is capable of asynchronous command
31222 execution, which means that @value{GDBN} will accept further commands
31223 while the target is running.
31226 Indicates that the target is capable of reverse execution.
31227 @xref{Reverse Execution}, for more information.
31231 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31232 @node GDB/MI Miscellaneous Commands
31233 @section Miscellaneous @sc{gdb/mi} Commands
31235 @c @subheading -gdb-complete
31237 @subheading The @code{-gdb-exit} Command
31240 @subsubheading Synopsis
31246 Exit @value{GDBN} immediately.
31248 @subsubheading @value{GDBN} Command
31250 Approximately corresponds to @samp{quit}.
31252 @subsubheading Example
31262 @subheading The @code{-exec-abort} Command
31263 @findex -exec-abort
31265 @subsubheading Synopsis
31271 Kill the inferior running program.
31273 @subsubheading @value{GDBN} Command
31275 The corresponding @value{GDBN} command is @samp{kill}.
31277 @subsubheading Example
31282 @subheading The @code{-gdb-set} Command
31285 @subsubheading Synopsis
31291 Set an internal @value{GDBN} variable.
31292 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
31294 @subsubheading @value{GDBN} Command
31296 The corresponding @value{GDBN} command is @samp{set}.
31298 @subsubheading Example
31308 @subheading The @code{-gdb-show} Command
31311 @subsubheading Synopsis
31317 Show the current value of a @value{GDBN} variable.
31319 @subsubheading @value{GDBN} Command
31321 The corresponding @value{GDBN} command is @samp{show}.
31323 @subsubheading Example
31332 @c @subheading -gdb-source
31335 @subheading The @code{-gdb-version} Command
31336 @findex -gdb-version
31338 @subsubheading Synopsis
31344 Show version information for @value{GDBN}. Used mostly in testing.
31346 @subsubheading @value{GDBN} Command
31348 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
31349 default shows this information when you start an interactive session.
31351 @subsubheading Example
31353 @c This example modifies the actual output from GDB to avoid overfull
31359 ~Copyright 2000 Free Software Foundation, Inc.
31360 ~GDB is free software, covered by the GNU General Public License, and
31361 ~you are welcome to change it and/or distribute copies of it under
31362 ~ certain conditions.
31363 ~Type "show copying" to see the conditions.
31364 ~There is absolutely no warranty for GDB. Type "show warranty" for
31366 ~This GDB was configured as
31367 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
31372 @subheading The @code{-list-thread-groups} Command
31373 @findex -list-thread-groups
31375 @subheading Synopsis
31378 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
31381 Lists thread groups (@pxref{Thread groups}). When a single thread
31382 group is passed as the argument, lists the children of that group.
31383 When several thread group are passed, lists information about those
31384 thread groups. Without any parameters, lists information about all
31385 top-level thread groups.
31387 Normally, thread groups that are being debugged are reported.
31388 With the @samp{--available} option, @value{GDBN} reports thread groups
31389 available on the target.
31391 The output of this command may have either a @samp{threads} result or
31392 a @samp{groups} result. The @samp{thread} result has a list of tuples
31393 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
31394 Information}). The @samp{groups} result has a list of tuples as value,
31395 each tuple describing a thread group. If top-level groups are
31396 requested (that is, no parameter is passed), or when several groups
31397 are passed, the output always has a @samp{groups} result. The format
31398 of the @samp{group} result is described below.
31400 To reduce the number of roundtrips it's possible to list thread groups
31401 together with their children, by passing the @samp{--recurse} option
31402 and the recursion depth. Presently, only recursion depth of 1 is
31403 permitted. If this option is present, then every reported thread group
31404 will also include its children, either as @samp{group} or
31405 @samp{threads} field.
31407 In general, any combination of option and parameters is permitted, with
31408 the following caveats:
31412 When a single thread group is passed, the output will typically
31413 be the @samp{threads} result. Because threads may not contain
31414 anything, the @samp{recurse} option will be ignored.
31417 When the @samp{--available} option is passed, limited information may
31418 be available. In particular, the list of threads of a process might
31419 be inaccessible. Further, specifying specific thread groups might
31420 not give any performance advantage over listing all thread groups.
31421 The frontend should assume that @samp{-list-thread-groups --available}
31422 is always an expensive operation and cache the results.
31426 The @samp{groups} result is a list of tuples, where each tuple may
31427 have the following fields:
31431 Identifier of the thread group. This field is always present.
31432 The identifier is an opaque string; frontends should not try to
31433 convert it to an integer, even though it might look like one.
31436 The type of the thread group. At present, only @samp{process} is a
31440 The target-specific process identifier. This field is only present
31441 for thread groups of type @samp{process} and only if the process exists.
31444 The exit code of this group's last exited thread, formatted in octal.
31445 This field is only present for thread groups of type @samp{process} and
31446 only if the process is not running.
31449 The number of children this thread group has. This field may be
31450 absent for an available thread group.
31453 This field has a list of tuples as value, each tuple describing a
31454 thread. It may be present if the @samp{--recurse} option is
31455 specified, and it's actually possible to obtain the threads.
31458 This field is a list of integers, each identifying a core that one
31459 thread of the group is running on. This field may be absent if
31460 such information is not available.
31463 The name of the executable file that corresponds to this thread group.
31464 The field is only present for thread groups of type @samp{process},
31465 and only if there is a corresponding executable file.
31469 @subheading Example
31473 -list-thread-groups
31474 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
31475 -list-thread-groups 17
31476 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
31477 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
31478 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
31479 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
31480 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
31481 -list-thread-groups --available
31482 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
31483 -list-thread-groups --available --recurse 1
31484 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
31485 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
31486 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
31487 -list-thread-groups --available --recurse 1 17 18
31488 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
31489 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
31490 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
31493 @subheading The @code{-info-os} Command
31496 @subsubheading Synopsis
31499 -info-os [ @var{type} ]
31502 If no argument is supplied, the command returns a table of available
31503 operating-system-specific information types. If one of these types is
31504 supplied as an argument @var{type}, then the command returns a table
31505 of data of that type.
31507 The types of information available depend on the target operating
31510 @subsubheading @value{GDBN} Command
31512 The corresponding @value{GDBN} command is @samp{info os}.
31514 @subsubheading Example
31516 When run on a @sc{gnu}/Linux system, the output will look something
31522 ^done,OSDataTable=@{nr_rows="9",nr_cols="3",
31523 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
31524 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
31525 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
31526 body=[item=@{col0="processes",col1="Listing of all processes",
31527 col2="Processes"@},
31528 item=@{col0="procgroups",col1="Listing of all process groups",
31529 col2="Process groups"@},
31530 item=@{col0="threads",col1="Listing of all threads",
31532 item=@{col0="files",col1="Listing of all file descriptors",
31533 col2="File descriptors"@},
31534 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
31536 item=@{col0="shm",col1="Listing of all shared-memory regions",
31537 col2="Shared-memory regions"@},
31538 item=@{col0="semaphores",col1="Listing of all semaphores",
31539 col2="Semaphores"@},
31540 item=@{col0="msg",col1="Listing of all message queues",
31541 col2="Message queues"@},
31542 item=@{col0="modules",col1="Listing of all loaded kernel modules",
31543 col2="Kernel modules"@}]@}
31546 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
31547 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
31548 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
31549 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
31550 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
31551 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
31552 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
31553 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
31555 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
31556 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
31560 (Note that the MI output here includes a @code{"Title"} column that
31561 does not appear in command-line @code{info os}; this column is useful
31562 for MI clients that want to enumerate the types of data, such as in a
31563 popup menu, but is needless clutter on the command line, and
31564 @code{info os} omits it.)
31566 @subheading The @code{-add-inferior} Command
31567 @findex -add-inferior
31569 @subheading Synopsis
31575 Creates a new inferior (@pxref{Inferiors and Programs}). The created
31576 inferior is not associated with any executable. Such association may
31577 be established with the @samp{-file-exec-and-symbols} command
31578 (@pxref{GDB/MI File Commands}). The command response has a single
31579 field, @samp{inferior}, whose value is the identifier of the
31580 thread group corresponding to the new inferior.
31582 @subheading Example
31587 ^done,inferior="i3"
31590 @subheading The @code{-interpreter-exec} Command
31591 @findex -interpreter-exec
31593 @subheading Synopsis
31596 -interpreter-exec @var{interpreter} @var{command}
31598 @anchor{-interpreter-exec}
31600 Execute the specified @var{command} in the given @var{interpreter}.
31602 @subheading @value{GDBN} Command
31604 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
31606 @subheading Example
31610 -interpreter-exec console "break main"
31611 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
31612 &"During symbol reading, bad structure-type format.\n"
31613 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
31618 @subheading The @code{-inferior-tty-set} Command
31619 @findex -inferior-tty-set
31621 @subheading Synopsis
31624 -inferior-tty-set /dev/pts/1
31627 Set terminal for future runs of the program being debugged.
31629 @subheading @value{GDBN} Command
31631 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
31633 @subheading Example
31637 -inferior-tty-set /dev/pts/1
31642 @subheading The @code{-inferior-tty-show} Command
31643 @findex -inferior-tty-show
31645 @subheading Synopsis
31651 Show terminal for future runs of program being debugged.
31653 @subheading @value{GDBN} Command
31655 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
31657 @subheading Example
31661 -inferior-tty-set /dev/pts/1
31665 ^done,inferior_tty_terminal="/dev/pts/1"
31669 @subheading The @code{-enable-timings} Command
31670 @findex -enable-timings
31672 @subheading Synopsis
31675 -enable-timings [yes | no]
31678 Toggle the printing of the wallclock, user and system times for an MI
31679 command as a field in its output. This command is to help frontend
31680 developers optimize the performance of their code. No argument is
31681 equivalent to @samp{yes}.
31683 @subheading @value{GDBN} Command
31687 @subheading Example
31695 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
31696 addr="0x080484ed",func="main",file="myprog.c",
31697 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
31699 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
31707 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
31708 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
31709 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
31710 fullname="/home/nickrob/myprog.c",line="73"@}
31715 @chapter @value{GDBN} Annotations
31717 This chapter describes annotations in @value{GDBN}. Annotations were
31718 designed to interface @value{GDBN} to graphical user interfaces or other
31719 similar programs which want to interact with @value{GDBN} at a
31720 relatively high level.
31722 The annotation mechanism has largely been superseded by @sc{gdb/mi}
31726 This is Edition @value{EDITION}, @value{DATE}.
31730 * Annotations Overview:: What annotations are; the general syntax.
31731 * Server Prefix:: Issuing a command without affecting user state.
31732 * Prompting:: Annotations marking @value{GDBN}'s need for input.
31733 * Errors:: Annotations for error messages.
31734 * Invalidation:: Some annotations describe things now invalid.
31735 * Annotations for Running::
31736 Whether the program is running, how it stopped, etc.
31737 * Source Annotations:: Annotations describing source code.
31740 @node Annotations Overview
31741 @section What is an Annotation?
31742 @cindex annotations
31744 Annotations start with a newline character, two @samp{control-z}
31745 characters, and the name of the annotation. If there is no additional
31746 information associated with this annotation, the name of the annotation
31747 is followed immediately by a newline. If there is additional
31748 information, the name of the annotation is followed by a space, the
31749 additional information, and a newline. The additional information
31750 cannot contain newline characters.
31752 Any output not beginning with a newline and two @samp{control-z}
31753 characters denotes literal output from @value{GDBN}. Currently there is
31754 no need for @value{GDBN} to output a newline followed by two
31755 @samp{control-z} characters, but if there was such a need, the
31756 annotations could be extended with an @samp{escape} annotation which
31757 means those three characters as output.
31759 The annotation @var{level}, which is specified using the
31760 @option{--annotate} command line option (@pxref{Mode Options}), controls
31761 how much information @value{GDBN} prints together with its prompt,
31762 values of expressions, source lines, and other types of output. Level 0
31763 is for no annotations, level 1 is for use when @value{GDBN} is run as a
31764 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
31765 for programs that control @value{GDBN}, and level 2 annotations have
31766 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
31767 Interface, annotate, GDB's Obsolete Annotations}).
31770 @kindex set annotate
31771 @item set annotate @var{level}
31772 The @value{GDBN} command @code{set annotate} sets the level of
31773 annotations to the specified @var{level}.
31775 @item show annotate
31776 @kindex show annotate
31777 Show the current annotation level.
31780 This chapter describes level 3 annotations.
31782 A simple example of starting up @value{GDBN} with annotations is:
31785 $ @kbd{gdb --annotate=3}
31787 Copyright 2003 Free Software Foundation, Inc.
31788 GDB is free software, covered by the GNU General Public License,
31789 and you are welcome to change it and/or distribute copies of it
31790 under certain conditions.
31791 Type "show copying" to see the conditions.
31792 There is absolutely no warranty for GDB. Type "show warranty"
31794 This GDB was configured as "i386-pc-linux-gnu"
31805 Here @samp{quit} is input to @value{GDBN}; the rest is output from
31806 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
31807 denotes a @samp{control-z} character) are annotations; the rest is
31808 output from @value{GDBN}.
31810 @node Server Prefix
31811 @section The Server Prefix
31812 @cindex server prefix
31814 If you prefix a command with @samp{server } then it will not affect
31815 the command history, nor will it affect @value{GDBN}'s notion of which
31816 command to repeat if @key{RET} is pressed on a line by itself. This
31817 means that commands can be run behind a user's back by a front-end in
31818 a transparent manner.
31820 The @code{server } prefix does not affect the recording of values into
31821 the value history; to print a value without recording it into the
31822 value history, use the @code{output} command instead of the
31823 @code{print} command.
31825 Using this prefix also disables confirmation requests
31826 (@pxref{confirmation requests}).
31829 @section Annotation for @value{GDBN} Input
31831 @cindex annotations for prompts
31832 When @value{GDBN} prompts for input, it annotates this fact so it is possible
31833 to know when to send output, when the output from a given command is
31836 Different kinds of input each have a different @dfn{input type}. Each
31837 input type has three annotations: a @code{pre-} annotation, which
31838 denotes the beginning of any prompt which is being output, a plain
31839 annotation, which denotes the end of the prompt, and then a @code{post-}
31840 annotation which denotes the end of any echo which may (or may not) be
31841 associated with the input. For example, the @code{prompt} input type
31842 features the following annotations:
31850 The input types are
31853 @findex pre-prompt annotation
31854 @findex prompt annotation
31855 @findex post-prompt annotation
31857 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
31859 @findex pre-commands annotation
31860 @findex commands annotation
31861 @findex post-commands annotation
31863 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
31864 command. The annotations are repeated for each command which is input.
31866 @findex pre-overload-choice annotation
31867 @findex overload-choice annotation
31868 @findex post-overload-choice annotation
31869 @item overload-choice
31870 When @value{GDBN} wants the user to select between various overloaded functions.
31872 @findex pre-query annotation
31873 @findex query annotation
31874 @findex post-query annotation
31876 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
31878 @findex pre-prompt-for-continue annotation
31879 @findex prompt-for-continue annotation
31880 @findex post-prompt-for-continue annotation
31881 @item prompt-for-continue
31882 When @value{GDBN} is asking the user to press return to continue. Note: Don't
31883 expect this to work well; instead use @code{set height 0} to disable
31884 prompting. This is because the counting of lines is buggy in the
31885 presence of annotations.
31890 @cindex annotations for errors, warnings and interrupts
31892 @findex quit annotation
31897 This annotation occurs right before @value{GDBN} responds to an interrupt.
31899 @findex error annotation
31904 This annotation occurs right before @value{GDBN} responds to an error.
31906 Quit and error annotations indicate that any annotations which @value{GDBN} was
31907 in the middle of may end abruptly. For example, if a
31908 @code{value-history-begin} annotation is followed by a @code{error}, one
31909 cannot expect to receive the matching @code{value-history-end}. One
31910 cannot expect not to receive it either, however; an error annotation
31911 does not necessarily mean that @value{GDBN} is immediately returning all the way
31914 @findex error-begin annotation
31915 A quit or error annotation may be preceded by
31921 Any output between that and the quit or error annotation is the error
31924 Warning messages are not yet annotated.
31925 @c If we want to change that, need to fix warning(), type_error(),
31926 @c range_error(), and possibly other places.
31929 @section Invalidation Notices
31931 @cindex annotations for invalidation messages
31932 The following annotations say that certain pieces of state may have
31936 @findex frames-invalid annotation
31937 @item ^Z^Zframes-invalid
31939 The frames (for example, output from the @code{backtrace} command) may
31942 @findex breakpoints-invalid annotation
31943 @item ^Z^Zbreakpoints-invalid
31945 The breakpoints may have changed. For example, the user just added or
31946 deleted a breakpoint.
31949 @node Annotations for Running
31950 @section Running the Program
31951 @cindex annotations for running programs
31953 @findex starting annotation
31954 @findex stopping annotation
31955 When the program starts executing due to a @value{GDBN} command such as
31956 @code{step} or @code{continue},
31962 is output. When the program stops,
31968 is output. Before the @code{stopped} annotation, a variety of
31969 annotations describe how the program stopped.
31972 @findex exited annotation
31973 @item ^Z^Zexited @var{exit-status}
31974 The program exited, and @var{exit-status} is the exit status (zero for
31975 successful exit, otherwise nonzero).
31977 @findex signalled annotation
31978 @findex signal-name annotation
31979 @findex signal-name-end annotation
31980 @findex signal-string annotation
31981 @findex signal-string-end annotation
31982 @item ^Z^Zsignalled
31983 The program exited with a signal. After the @code{^Z^Zsignalled}, the
31984 annotation continues:
31990 ^Z^Zsignal-name-end
31994 ^Z^Zsignal-string-end
31999 where @var{name} is the name of the signal, such as @code{SIGILL} or
32000 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
32001 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
32002 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
32003 user's benefit and have no particular format.
32005 @findex signal annotation
32007 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
32008 just saying that the program received the signal, not that it was
32009 terminated with it.
32011 @findex breakpoint annotation
32012 @item ^Z^Zbreakpoint @var{number}
32013 The program hit breakpoint number @var{number}.
32015 @findex watchpoint annotation
32016 @item ^Z^Zwatchpoint @var{number}
32017 The program hit watchpoint number @var{number}.
32020 @node Source Annotations
32021 @section Displaying Source
32022 @cindex annotations for source display
32024 @findex source annotation
32025 The following annotation is used instead of displaying source code:
32028 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
32031 where @var{filename} is an absolute file name indicating which source
32032 file, @var{line} is the line number within that file (where 1 is the
32033 first line in the file), @var{character} is the character position
32034 within the file (where 0 is the first character in the file) (for most
32035 debug formats this will necessarily point to the beginning of a line),
32036 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
32037 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
32038 @var{addr} is the address in the target program associated with the
32039 source which is being displayed. The @var{addr} is in the form @samp{0x}
32040 followed by one or more lowercase hex digits (note that this does not
32041 depend on the language).
32043 @node JIT Interface
32044 @chapter JIT Compilation Interface
32045 @cindex just-in-time compilation
32046 @cindex JIT compilation interface
32048 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
32049 interface. A JIT compiler is a program or library that generates native
32050 executable code at runtime and executes it, usually in order to achieve good
32051 performance while maintaining platform independence.
32053 Programs that use JIT compilation are normally difficult to debug because
32054 portions of their code are generated at runtime, instead of being loaded from
32055 object files, which is where @value{GDBN} normally finds the program's symbols
32056 and debug information. In order to debug programs that use JIT compilation,
32057 @value{GDBN} has an interface that allows the program to register in-memory
32058 symbol files with @value{GDBN} at runtime.
32060 If you are using @value{GDBN} to debug a program that uses this interface, then
32061 it should work transparently so long as you have not stripped the binary. If
32062 you are developing a JIT compiler, then the interface is documented in the rest
32063 of this chapter. At this time, the only known client of this interface is the
32066 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
32067 JIT compiler communicates with @value{GDBN} by writing data into a global
32068 variable and calling a fuction at a well-known symbol. When @value{GDBN}
32069 attaches, it reads a linked list of symbol files from the global variable to
32070 find existing code, and puts a breakpoint in the function so that it can find
32071 out about additional code.
32074 * Declarations:: Relevant C struct declarations
32075 * Registering Code:: Steps to register code
32076 * Unregistering Code:: Steps to unregister code
32077 * Custom Debug Info:: Emit debug information in a custom format
32081 @section JIT Declarations
32083 These are the relevant struct declarations that a C program should include to
32084 implement the interface:
32094 struct jit_code_entry
32096 struct jit_code_entry *next_entry;
32097 struct jit_code_entry *prev_entry;
32098 const char *symfile_addr;
32099 uint64_t symfile_size;
32102 struct jit_descriptor
32105 /* This type should be jit_actions_t, but we use uint32_t
32106 to be explicit about the bitwidth. */
32107 uint32_t action_flag;
32108 struct jit_code_entry *relevant_entry;
32109 struct jit_code_entry *first_entry;
32112 /* GDB puts a breakpoint in this function. */
32113 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
32115 /* Make sure to specify the version statically, because the
32116 debugger may check the version before we can set it. */
32117 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
32120 If the JIT is multi-threaded, then it is important that the JIT synchronize any
32121 modifications to this global data properly, which can easily be done by putting
32122 a global mutex around modifications to these structures.
32124 @node Registering Code
32125 @section Registering Code
32127 To register code with @value{GDBN}, the JIT should follow this protocol:
32131 Generate an object file in memory with symbols and other desired debug
32132 information. The file must include the virtual addresses of the sections.
32135 Create a code entry for the file, which gives the start and size of the symbol
32139 Add it to the linked list in the JIT descriptor.
32142 Point the relevant_entry field of the descriptor at the entry.
32145 Set @code{action_flag} to @code{JIT_REGISTER} and call
32146 @code{__jit_debug_register_code}.
32149 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
32150 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
32151 new code. However, the linked list must still be maintained in order to allow
32152 @value{GDBN} to attach to a running process and still find the symbol files.
32154 @node Unregistering Code
32155 @section Unregistering Code
32157 If code is freed, then the JIT should use the following protocol:
32161 Remove the code entry corresponding to the code from the linked list.
32164 Point the @code{relevant_entry} field of the descriptor at the code entry.
32167 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
32168 @code{__jit_debug_register_code}.
32171 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
32172 and the JIT will leak the memory used for the associated symbol files.
32174 @node Custom Debug Info
32175 @section Custom Debug Info
32176 @cindex custom JIT debug info
32177 @cindex JIT debug info reader
32179 Generating debug information in platform-native file formats (like ELF
32180 or COFF) may be an overkill for JIT compilers; especially if all the
32181 debug info is used for is displaying a meaningful backtrace. The
32182 issue can be resolved by having the JIT writers decide on a debug info
32183 format and also provide a reader that parses the debug info generated
32184 by the JIT compiler. This section gives a brief overview on writing
32185 such a parser. More specific details can be found in the source file
32186 @file{gdb/jit-reader.in}, which is also installed as a header at
32187 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
32189 The reader is implemented as a shared object (so this functionality is
32190 not available on platforms which don't allow loading shared objects at
32191 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
32192 @code{jit-reader-unload} are provided, to be used to load and unload
32193 the readers from a preconfigured directory. Once loaded, the shared
32194 object is used the parse the debug information emitted by the JIT
32198 * Using JIT Debug Info Readers:: How to use supplied readers correctly
32199 * Writing JIT Debug Info Readers:: Creating a debug-info reader
32202 @node Using JIT Debug Info Readers
32203 @subsection Using JIT Debug Info Readers
32204 @kindex jit-reader-load
32205 @kindex jit-reader-unload
32207 Readers can be loaded and unloaded using the @code{jit-reader-load}
32208 and @code{jit-reader-unload} commands.
32211 @item jit-reader-load @var{reader}
32212 Load the JIT reader named @var{reader}, which is a shared
32213 object specified as either an absolute or a relative file name. In
32214 the latter case, @value{GDBN} will try to load the reader from a
32215 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
32216 system (here @var{libdir} is the system library directory, often
32217 @file{/usr/local/lib}).
32219 Only one reader can be active at a time; trying to load a second
32220 reader when one is already loaded will result in @value{GDBN}
32221 reporting an error. A new JIT reader can be loaded by first unloading
32222 the current one using @code{jit-reader-unload} and then invoking
32223 @code{jit-reader-load}.
32225 @item jit-reader-unload
32226 Unload the currently loaded JIT reader.
32230 @node Writing JIT Debug Info Readers
32231 @subsection Writing JIT Debug Info Readers
32232 @cindex writing JIT debug info readers
32234 As mentioned, a reader is essentially a shared object conforming to a
32235 certain ABI. This ABI is described in @file{jit-reader.h}.
32237 @file{jit-reader.h} defines the structures, macros and functions
32238 required to write a reader. It is installed (along with
32239 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
32240 the system include directory.
32242 Readers need to be released under a GPL compatible license. A reader
32243 can be declared as released under such a license by placing the macro
32244 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
32246 The entry point for readers is the symbol @code{gdb_init_reader},
32247 which is expected to be a function with the prototype
32249 @findex gdb_init_reader
32251 extern struct gdb_reader_funcs *gdb_init_reader (void);
32254 @cindex @code{struct gdb_reader_funcs}
32256 @code{struct gdb_reader_funcs} contains a set of pointers to callback
32257 functions. These functions are executed to read the debug info
32258 generated by the JIT compiler (@code{read}), to unwind stack frames
32259 (@code{unwind}) and to create canonical frame IDs
32260 (@code{get_Frame_id}). It also has a callback that is called when the
32261 reader is being unloaded (@code{destroy}). The struct looks like this
32264 struct gdb_reader_funcs
32266 /* Must be set to GDB_READER_INTERFACE_VERSION. */
32267 int reader_version;
32269 /* For use by the reader. */
32272 gdb_read_debug_info *read;
32273 gdb_unwind_frame *unwind;
32274 gdb_get_frame_id *get_frame_id;
32275 gdb_destroy_reader *destroy;
32279 @cindex @code{struct gdb_symbol_callbacks}
32280 @cindex @code{struct gdb_unwind_callbacks}
32282 The callbacks are provided with another set of callbacks by
32283 @value{GDBN} to do their job. For @code{read}, these callbacks are
32284 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
32285 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
32286 @code{struct gdb_symbol_callbacks} has callbacks to create new object
32287 files and new symbol tables inside those object files. @code{struct
32288 gdb_unwind_callbacks} has callbacks to read registers off the current
32289 frame and to write out the values of the registers in the previous
32290 frame. Both have a callback (@code{target_read}) to read bytes off the
32291 target's address space.
32293 @node In-Process Agent
32294 @chapter In-Process Agent
32295 @cindex debugging agent
32296 The traditional debugging model is conceptually low-speed, but works fine,
32297 because most bugs can be reproduced in debugging-mode execution. However,
32298 as multi-core or many-core processors are becoming mainstream, and
32299 multi-threaded programs become more and more popular, there should be more
32300 and more bugs that only manifest themselves at normal-mode execution, for
32301 example, thread races, because debugger's interference with the program's
32302 timing may conceal the bugs. On the other hand, in some applications,
32303 it is not feasible for the debugger to interrupt the program's execution
32304 long enough for the developer to learn anything helpful about its behavior.
32305 If the program's correctness depends on its real-time behavior, delays
32306 introduced by a debugger might cause the program to fail, even when the
32307 code itself is correct. It is useful to be able to observe the program's
32308 behavior without interrupting it.
32310 Therefore, traditional debugging model is too intrusive to reproduce
32311 some bugs. In order to reduce the interference with the program, we can
32312 reduce the number of operations performed by debugger. The
32313 @dfn{In-Process Agent}, a shared library, is running within the same
32314 process with inferior, and is able to perform some debugging operations
32315 itself. As a result, debugger is only involved when necessary, and
32316 performance of debugging can be improved accordingly. Note that
32317 interference with program can be reduced but can't be removed completely,
32318 because the in-process agent will still stop or slow down the program.
32320 The in-process agent can interpret and execute Agent Expressions
32321 (@pxref{Agent Expressions}) during performing debugging operations. The
32322 agent expressions can be used for different purposes, such as collecting
32323 data in tracepoints, and condition evaluation in breakpoints.
32325 @anchor{Control Agent}
32326 You can control whether the in-process agent is used as an aid for
32327 debugging with the following commands:
32330 @kindex set agent on
32332 Causes the in-process agent to perform some operations on behalf of the
32333 debugger. Just which operations requested by the user will be done
32334 by the in-process agent depends on the its capabilities. For example,
32335 if you request to evaluate breakpoint conditions in the in-process agent,
32336 and the in-process agent has such capability as well, then breakpoint
32337 conditions will be evaluated in the in-process agent.
32339 @kindex set agent off
32340 @item set agent off
32341 Disables execution of debugging operations by the in-process agent. All
32342 of the operations will be performed by @value{GDBN}.
32346 Display the current setting of execution of debugging operations by
32347 the in-process agent.
32351 * In-Process Agent Protocol::
32354 @node In-Process Agent Protocol
32355 @section In-Process Agent Protocol
32356 @cindex in-process agent protocol
32358 The in-process agent is able to communicate with both @value{GDBN} and
32359 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
32360 used for communications between @value{GDBN} or GDBserver and the IPA.
32361 In general, @value{GDBN} or GDBserver sends commands
32362 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
32363 in-process agent replies back with the return result of the command, or
32364 some other information. The data sent to in-process agent is composed
32365 of primitive data types, such as 4-byte or 8-byte type, and composite
32366 types, which are called objects (@pxref{IPA Protocol Objects}).
32369 * IPA Protocol Objects::
32370 * IPA Protocol Commands::
32373 @node IPA Protocol Objects
32374 @subsection IPA Protocol Objects
32375 @cindex ipa protocol objects
32377 The commands sent to and results received from agent may contain some
32378 complex data types called @dfn{objects}.
32380 The in-process agent is running on the same machine with @value{GDBN}
32381 or GDBserver, so it doesn't have to handle as much differences between
32382 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
32383 However, there are still some differences of two ends in two processes:
32387 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
32388 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
32390 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
32391 GDBserver is compiled with one, and in-process agent is compiled with
32395 Here are the IPA Protocol Objects:
32399 agent expression object. It represents an agent expression
32400 (@pxref{Agent Expressions}).
32401 @anchor{agent expression object}
32403 tracepoint action object. It represents a tracepoint action
32404 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
32405 memory, static trace data and to evaluate expression.
32406 @anchor{tracepoint action object}
32408 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
32409 @anchor{tracepoint object}
32413 The following table describes important attributes of each IPA protocol
32416 @multitable @columnfractions .30 .20 .50
32417 @headitem Name @tab Size @tab Description
32418 @item @emph{agent expression object} @tab @tab
32419 @item length @tab 4 @tab length of bytes code
32420 @item byte code @tab @var{length} @tab contents of byte code
32421 @item @emph{tracepoint action for collecting memory} @tab @tab
32422 @item 'M' @tab 1 @tab type of tracepoint action
32423 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
32424 address of the lowest byte to collect, otherwise @var{addr} is the offset
32425 of @var{basereg} for memory collecting.
32426 @item len @tab 8 @tab length of memory for collecting
32427 @item basereg @tab 4 @tab the register number containing the starting
32428 memory address for collecting.
32429 @item @emph{tracepoint action for collecting registers} @tab @tab
32430 @item 'R' @tab 1 @tab type of tracepoint action
32431 @item @emph{tracepoint action for collecting static trace data} @tab @tab
32432 @item 'L' @tab 1 @tab type of tracepoint action
32433 @item @emph{tracepoint action for expression evaluation} @tab @tab
32434 @item 'X' @tab 1 @tab type of tracepoint action
32435 @item agent expression @tab length of @tab @ref{agent expression object}
32436 @item @emph{tracepoint object} @tab @tab
32437 @item number @tab 4 @tab number of tracepoint
32438 @item address @tab 8 @tab address of tracepoint inserted on
32439 @item type @tab 4 @tab type of tracepoint
32440 @item enabled @tab 1 @tab enable or disable of tracepoint
32441 @item step_count @tab 8 @tab step
32442 @item pass_count @tab 8 @tab pass
32443 @item numactions @tab 4 @tab number of tracepoint actions
32444 @item hit count @tab 8 @tab hit count
32445 @item trace frame usage @tab 8 @tab trace frame usage
32446 @item compiled_cond @tab 8 @tab compiled condition
32447 @item orig_size @tab 8 @tab orig size
32448 @item condition @tab 4 if condition is NULL otherwise length of
32449 @ref{agent expression object}
32450 @tab zero if condition is NULL, otherwise is
32451 @ref{agent expression object}
32452 @item actions @tab variable
32453 @tab numactions number of @ref{tracepoint action object}
32456 @node IPA Protocol Commands
32457 @subsection IPA Protocol Commands
32458 @cindex ipa protocol commands
32460 The spaces in each command are delimiters to ease reading this commands
32461 specification. They don't exist in real commands.
32465 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
32466 Installs a new fast tracepoint described by @var{tracepoint_object}
32467 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
32468 head of @dfn{jumppad}, which is used to jump to data collection routine
32473 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
32474 @var{target_address} is address of tracepoint in the inferior.
32475 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
32476 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
32477 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
32478 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
32485 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
32486 is about to kill inferiors.
32494 @item probe_marker_at:@var{address}
32495 Asks in-process agent to probe the marker at @var{address}.
32502 @item unprobe_marker_at:@var{address}
32503 Asks in-process agent to unprobe the marker at @var{address}.
32507 @chapter Reporting Bugs in @value{GDBN}
32508 @cindex bugs in @value{GDBN}
32509 @cindex reporting bugs in @value{GDBN}
32511 Your bug reports play an essential role in making @value{GDBN} reliable.
32513 Reporting a bug may help you by bringing a solution to your problem, or it
32514 may not. But in any case the principal function of a bug report is to help
32515 the entire community by making the next version of @value{GDBN} work better. Bug
32516 reports are your contribution to the maintenance of @value{GDBN}.
32518 In order for a bug report to serve its purpose, you must include the
32519 information that enables us to fix the bug.
32522 * Bug Criteria:: Have you found a bug?
32523 * Bug Reporting:: How to report bugs
32527 @section Have You Found a Bug?
32528 @cindex bug criteria
32530 If you are not sure whether you have found a bug, here are some guidelines:
32533 @cindex fatal signal
32534 @cindex debugger crash
32535 @cindex crash of debugger
32537 If the debugger gets a fatal signal, for any input whatever, that is a
32538 @value{GDBN} bug. Reliable debuggers never crash.
32540 @cindex error on valid input
32542 If @value{GDBN} produces an error message for valid input, that is a
32543 bug. (Note that if you're cross debugging, the problem may also be
32544 somewhere in the connection to the target.)
32546 @cindex invalid input
32548 If @value{GDBN} does not produce an error message for invalid input,
32549 that is a bug. However, you should note that your idea of
32550 ``invalid input'' might be our idea of ``an extension'' or ``support
32551 for traditional practice''.
32554 If you are an experienced user of debugging tools, your suggestions
32555 for improvement of @value{GDBN} are welcome in any case.
32558 @node Bug Reporting
32559 @section How to Report Bugs
32560 @cindex bug reports
32561 @cindex @value{GDBN} bugs, reporting
32563 A number of companies and individuals offer support for @sc{gnu} products.
32564 If you obtained @value{GDBN} from a support organization, we recommend you
32565 contact that organization first.
32567 You can find contact information for many support companies and
32568 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
32570 @c should add a web page ref...
32573 @ifset BUGURL_DEFAULT
32574 In any event, we also recommend that you submit bug reports for
32575 @value{GDBN}. The preferred method is to submit them directly using
32576 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
32577 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
32580 @strong{Do not send bug reports to @samp{info-gdb}, or to
32581 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
32582 not want to receive bug reports. Those that do have arranged to receive
32585 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
32586 serves as a repeater. The mailing list and the newsgroup carry exactly
32587 the same messages. Often people think of posting bug reports to the
32588 newsgroup instead of mailing them. This appears to work, but it has one
32589 problem which can be crucial: a newsgroup posting often lacks a mail
32590 path back to the sender. Thus, if we need to ask for more information,
32591 we may be unable to reach you. For this reason, it is better to send
32592 bug reports to the mailing list.
32594 @ifclear BUGURL_DEFAULT
32595 In any event, we also recommend that you submit bug reports for
32596 @value{GDBN} to @value{BUGURL}.
32600 The fundamental principle of reporting bugs usefully is this:
32601 @strong{report all the facts}. If you are not sure whether to state a
32602 fact or leave it out, state it!
32604 Often people omit facts because they think they know what causes the
32605 problem and assume that some details do not matter. Thus, you might
32606 assume that the name of the variable you use in an example does not matter.
32607 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
32608 stray memory reference which happens to fetch from the location where that
32609 name is stored in memory; perhaps, if the name were different, the contents
32610 of that location would fool the debugger into doing the right thing despite
32611 the bug. Play it safe and give a specific, complete example. That is the
32612 easiest thing for you to do, and the most helpful.
32614 Keep in mind that the purpose of a bug report is to enable us to fix the
32615 bug. It may be that the bug has been reported previously, but neither
32616 you nor we can know that unless your bug report is complete and
32619 Sometimes people give a few sketchy facts and ask, ``Does this ring a
32620 bell?'' Those bug reports are useless, and we urge everyone to
32621 @emph{refuse to respond to them} except to chide the sender to report
32624 To enable us to fix the bug, you should include all these things:
32628 The version of @value{GDBN}. @value{GDBN} announces it if you start
32629 with no arguments; you can also print it at any time using @code{show
32632 Without this, we will not know whether there is any point in looking for
32633 the bug in the current version of @value{GDBN}.
32636 The type of machine you are using, and the operating system name and
32640 The details of the @value{GDBN} build-time configuration.
32641 @value{GDBN} shows these details if you invoke it with the
32642 @option{--configuration} command-line option, or if you type
32643 @code{show configuration} at @value{GDBN}'s prompt.
32646 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
32647 ``@value{GCC}--2.8.1''.
32650 What compiler (and its version) was used to compile the program you are
32651 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
32652 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
32653 to get this information; for other compilers, see the documentation for
32657 The command arguments you gave the compiler to compile your example and
32658 observe the bug. For example, did you use @samp{-O}? To guarantee
32659 you will not omit something important, list them all. A copy of the
32660 Makefile (or the output from make) is sufficient.
32662 If we were to try to guess the arguments, we would probably guess wrong
32663 and then we might not encounter the bug.
32666 A complete input script, and all necessary source files, that will
32670 A description of what behavior you observe that you believe is
32671 incorrect. For example, ``It gets a fatal signal.''
32673 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
32674 will certainly notice it. But if the bug is incorrect output, we might
32675 not notice unless it is glaringly wrong. You might as well not give us
32676 a chance to make a mistake.
32678 Even if the problem you experience is a fatal signal, you should still
32679 say so explicitly. Suppose something strange is going on, such as, your
32680 copy of @value{GDBN} is out of synch, or you have encountered a bug in
32681 the C library on your system. (This has happened!) Your copy might
32682 crash and ours would not. If you told us to expect a crash, then when
32683 ours fails to crash, we would know that the bug was not happening for
32684 us. If you had not told us to expect a crash, then we would not be able
32685 to draw any conclusion from our observations.
32688 @cindex recording a session script
32689 To collect all this information, you can use a session recording program
32690 such as @command{script}, which is available on many Unix systems.
32691 Just run your @value{GDBN} session inside @command{script} and then
32692 include the @file{typescript} file with your bug report.
32694 Another way to record a @value{GDBN} session is to run @value{GDBN}
32695 inside Emacs and then save the entire buffer to a file.
32698 If you wish to suggest changes to the @value{GDBN} source, send us context
32699 diffs. If you even discuss something in the @value{GDBN} source, refer to
32700 it by context, not by line number.
32702 The line numbers in our development sources will not match those in your
32703 sources. Your line numbers would convey no useful information to us.
32707 Here are some things that are not necessary:
32711 A description of the envelope of the bug.
32713 Often people who encounter a bug spend a lot of time investigating
32714 which changes to the input file will make the bug go away and which
32715 changes will not affect it.
32717 This is often time consuming and not very useful, because the way we
32718 will find the bug is by running a single example under the debugger
32719 with breakpoints, not by pure deduction from a series of examples.
32720 We recommend that you save your time for something else.
32722 Of course, if you can find a simpler example to report @emph{instead}
32723 of the original one, that is a convenience for us. Errors in the
32724 output will be easier to spot, running under the debugger will take
32725 less time, and so on.
32727 However, simplification is not vital; if you do not want to do this,
32728 report the bug anyway and send us the entire test case you used.
32731 A patch for the bug.
32733 A patch for the bug does help us if it is a good one. But do not omit
32734 the necessary information, such as the test case, on the assumption that
32735 a patch is all we need. We might see problems with your patch and decide
32736 to fix the problem another way, or we might not understand it at all.
32738 Sometimes with a program as complicated as @value{GDBN} it is very hard to
32739 construct an example that will make the program follow a certain path
32740 through the code. If you do not send us the example, we will not be able
32741 to construct one, so we will not be able to verify that the bug is fixed.
32743 And if we cannot understand what bug you are trying to fix, or why your
32744 patch should be an improvement, we will not install it. A test case will
32745 help us to understand.
32748 A guess about what the bug is or what it depends on.
32750 Such guesses are usually wrong. Even we cannot guess right about such
32751 things without first using the debugger to find the facts.
32754 @c The readline documentation is distributed with the readline code
32755 @c and consists of the two following files:
32758 @c Use -I with makeinfo to point to the appropriate directory,
32759 @c environment var TEXINPUTS with TeX.
32760 @ifclear SYSTEM_READLINE
32761 @include rluser.texi
32762 @include hsuser.texi
32766 @appendix In Memoriam
32768 The @value{GDBN} project mourns the loss of the following long-time
32773 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
32774 to Free Software in general. Outside of @value{GDBN}, he was known in
32775 the Amiga world for his series of Fish Disks, and the GeekGadget project.
32777 @item Michael Snyder
32778 Michael was one of the Global Maintainers of the @value{GDBN} project,
32779 with contributions recorded as early as 1996, until 2011. In addition
32780 to his day to day participation, he was a large driving force behind
32781 adding Reverse Debugging to @value{GDBN}.
32784 Beyond their technical contributions to the project, they were also
32785 enjoyable members of the Free Software Community. We will miss them.
32787 @node Formatting Documentation
32788 @appendix Formatting Documentation
32790 @cindex @value{GDBN} reference card
32791 @cindex reference card
32792 The @value{GDBN} 4 release includes an already-formatted reference card, ready
32793 for printing with PostScript or Ghostscript, in the @file{gdb}
32794 subdirectory of the main source directory@footnote{In
32795 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
32796 release.}. If you can use PostScript or Ghostscript with your printer,
32797 you can print the reference card immediately with @file{refcard.ps}.
32799 The release also includes the source for the reference card. You
32800 can format it, using @TeX{}, by typing:
32806 The @value{GDBN} reference card is designed to print in @dfn{landscape}
32807 mode on US ``letter'' size paper;
32808 that is, on a sheet 11 inches wide by 8.5 inches
32809 high. You will need to specify this form of printing as an option to
32810 your @sc{dvi} output program.
32812 @cindex documentation
32814 All the documentation for @value{GDBN} comes as part of the machine-readable
32815 distribution. The documentation is written in Texinfo format, which is
32816 a documentation system that uses a single source file to produce both
32817 on-line information and a printed manual. You can use one of the Info
32818 formatting commands to create the on-line version of the documentation
32819 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
32821 @value{GDBN} includes an already formatted copy of the on-line Info
32822 version of this manual in the @file{gdb} subdirectory. The main Info
32823 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
32824 subordinate files matching @samp{gdb.info*} in the same directory. If
32825 necessary, you can print out these files, or read them with any editor;
32826 but they are easier to read using the @code{info} subsystem in @sc{gnu}
32827 Emacs or the standalone @code{info} program, available as part of the
32828 @sc{gnu} Texinfo distribution.
32830 If you want to format these Info files yourself, you need one of the
32831 Info formatting programs, such as @code{texinfo-format-buffer} or
32834 If you have @code{makeinfo} installed, and are in the top level
32835 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
32836 version @value{GDBVN}), you can make the Info file by typing:
32843 If you want to typeset and print copies of this manual, you need @TeX{},
32844 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
32845 Texinfo definitions file.
32847 @TeX{} is a typesetting program; it does not print files directly, but
32848 produces output files called @sc{dvi} files. To print a typeset
32849 document, you need a program to print @sc{dvi} files. If your system
32850 has @TeX{} installed, chances are it has such a program. The precise
32851 command to use depends on your system; @kbd{lpr -d} is common; another
32852 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
32853 require a file name without any extension or a @samp{.dvi} extension.
32855 @TeX{} also requires a macro definitions file called
32856 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
32857 written in Texinfo format. On its own, @TeX{} cannot either read or
32858 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
32859 and is located in the @file{gdb-@var{version-number}/texinfo}
32862 If you have @TeX{} and a @sc{dvi} printer program installed, you can
32863 typeset and print this manual. First switch to the @file{gdb}
32864 subdirectory of the main source directory (for example, to
32865 @file{gdb-@value{GDBVN}/gdb}) and type:
32871 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
32873 @node Installing GDB
32874 @appendix Installing @value{GDBN}
32875 @cindex installation
32878 * Requirements:: Requirements for building @value{GDBN}
32879 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
32880 * Separate Objdir:: Compiling @value{GDBN} in another directory
32881 * Config Names:: Specifying names for hosts and targets
32882 * Configure Options:: Summary of options for configure
32883 * System-wide configuration:: Having a system-wide init file
32887 @section Requirements for Building @value{GDBN}
32888 @cindex building @value{GDBN}, requirements for
32890 Building @value{GDBN} requires various tools and packages to be available.
32891 Other packages will be used only if they are found.
32893 @heading Tools/Packages Necessary for Building @value{GDBN}
32895 @item ISO C90 compiler
32896 @value{GDBN} is written in ISO C90. It should be buildable with any
32897 working C90 compiler, e.g.@: GCC.
32901 @heading Tools/Packages Optional for Building @value{GDBN}
32905 @value{GDBN} can use the Expat XML parsing library. This library may be
32906 included with your operating system distribution; if it is not, you
32907 can get the latest version from @url{http://expat.sourceforge.net}.
32908 The @file{configure} script will search for this library in several
32909 standard locations; if it is installed in an unusual path, you can
32910 use the @option{--with-libexpat-prefix} option to specify its location.
32916 Remote protocol memory maps (@pxref{Memory Map Format})
32918 Target descriptions (@pxref{Target Descriptions})
32920 Remote shared library lists (@xref{Library List Format},
32921 or alternatively @pxref{Library List Format for SVR4 Targets})
32923 MS-Windows shared libraries (@pxref{Shared Libraries})
32925 Traceframe info (@pxref{Traceframe Info Format})
32927 Branch trace (@pxref{Branch Trace Format})
32931 @cindex compressed debug sections
32932 @value{GDBN} will use the @samp{zlib} library, if available, to read
32933 compressed debug sections. Some linkers, such as GNU gold, are capable
32934 of producing binaries with compressed debug sections. If @value{GDBN}
32935 is compiled with @samp{zlib}, it will be able to read the debug
32936 information in such binaries.
32938 The @samp{zlib} library is likely included with your operating system
32939 distribution; if it is not, you can get the latest version from
32940 @url{http://zlib.net}.
32943 @value{GDBN}'s features related to character sets (@pxref{Character
32944 Sets}) require a functioning @code{iconv} implementation. If you are
32945 on a GNU system, then this is provided by the GNU C Library. Some
32946 other systems also provide a working @code{iconv}.
32948 If @value{GDBN} is using the @code{iconv} program which is installed
32949 in a non-standard place, you will need to tell @value{GDBN} where to find it.
32950 This is done with @option{--with-iconv-bin} which specifies the
32951 directory that contains the @code{iconv} program.
32953 On systems without @code{iconv}, you can install GNU Libiconv. If you
32954 have previously installed Libiconv, you can use the
32955 @option{--with-libiconv-prefix} option to configure.
32957 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
32958 arrange to build Libiconv if a directory named @file{libiconv} appears
32959 in the top-most source directory. If Libiconv is built this way, and
32960 if the operating system does not provide a suitable @code{iconv}
32961 implementation, then the just-built library will automatically be used
32962 by @value{GDBN}. One easy way to set this up is to download GNU
32963 Libiconv, unpack it, and then rename the directory holding the
32964 Libiconv source code to @samp{libiconv}.
32967 @node Running Configure
32968 @section Invoking the @value{GDBN} @file{configure} Script
32969 @cindex configuring @value{GDBN}
32970 @value{GDBN} comes with a @file{configure} script that automates the process
32971 of preparing @value{GDBN} for installation; you can then use @code{make} to
32972 build the @code{gdb} program.
32974 @c irrelevant in info file; it's as current as the code it lives with.
32975 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
32976 look at the @file{README} file in the sources; we may have improved the
32977 installation procedures since publishing this manual.}
32980 The @value{GDBN} distribution includes all the source code you need for
32981 @value{GDBN} in a single directory, whose name is usually composed by
32982 appending the version number to @samp{gdb}.
32984 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
32985 @file{gdb-@value{GDBVN}} directory. That directory contains:
32988 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
32989 script for configuring @value{GDBN} and all its supporting libraries
32991 @item gdb-@value{GDBVN}/gdb
32992 the source specific to @value{GDBN} itself
32994 @item gdb-@value{GDBVN}/bfd
32995 source for the Binary File Descriptor library
32997 @item gdb-@value{GDBVN}/include
32998 @sc{gnu} include files
33000 @item gdb-@value{GDBVN}/libiberty
33001 source for the @samp{-liberty} free software library
33003 @item gdb-@value{GDBVN}/opcodes
33004 source for the library of opcode tables and disassemblers
33006 @item gdb-@value{GDBVN}/readline
33007 source for the @sc{gnu} command-line interface
33009 @item gdb-@value{GDBVN}/glob
33010 source for the @sc{gnu} filename pattern-matching subroutine
33012 @item gdb-@value{GDBVN}/mmalloc
33013 source for the @sc{gnu} memory-mapped malloc package
33016 The simplest way to configure and build @value{GDBN} is to run @file{configure}
33017 from the @file{gdb-@var{version-number}} source directory, which in
33018 this example is the @file{gdb-@value{GDBVN}} directory.
33020 First switch to the @file{gdb-@var{version-number}} source directory
33021 if you are not already in it; then run @file{configure}. Pass the
33022 identifier for the platform on which @value{GDBN} will run as an
33028 cd gdb-@value{GDBVN}
33029 ./configure @var{host}
33034 where @var{host} is an identifier such as @samp{sun4} or
33035 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
33036 (You can often leave off @var{host}; @file{configure} tries to guess the
33037 correct value by examining your system.)
33039 Running @samp{configure @var{host}} and then running @code{make} builds the
33040 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
33041 libraries, then @code{gdb} itself. The configured source files, and the
33042 binaries, are left in the corresponding source directories.
33045 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
33046 system does not recognize this automatically when you run a different
33047 shell, you may need to run @code{sh} on it explicitly:
33050 sh configure @var{host}
33053 If you run @file{configure} from a directory that contains source
33054 directories for multiple libraries or programs, such as the
33055 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
33057 creates configuration files for every directory level underneath (unless
33058 you tell it not to, with the @samp{--norecursion} option).
33060 You should run the @file{configure} script from the top directory in the
33061 source tree, the @file{gdb-@var{version-number}} directory. If you run
33062 @file{configure} from one of the subdirectories, you will configure only
33063 that subdirectory. That is usually not what you want. In particular,
33064 if you run the first @file{configure} from the @file{gdb} subdirectory
33065 of the @file{gdb-@var{version-number}} directory, you will omit the
33066 configuration of @file{bfd}, @file{readline}, and other sibling
33067 directories of the @file{gdb} subdirectory. This leads to build errors
33068 about missing include files such as @file{bfd/bfd.h}.
33070 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
33071 However, you should make sure that the shell on your path (named by
33072 the @samp{SHELL} environment variable) is publicly readable. Remember
33073 that @value{GDBN} uses the shell to start your program---some systems refuse to
33074 let @value{GDBN} debug child processes whose programs are not readable.
33076 @node Separate Objdir
33077 @section Compiling @value{GDBN} in Another Directory
33079 If you want to run @value{GDBN} versions for several host or target machines,
33080 you need a different @code{gdb} compiled for each combination of
33081 host and target. @file{configure} is designed to make this easy by
33082 allowing you to generate each configuration in a separate subdirectory,
33083 rather than in the source directory. If your @code{make} program
33084 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
33085 @code{make} in each of these directories builds the @code{gdb}
33086 program specified there.
33088 To build @code{gdb} in a separate directory, run @file{configure}
33089 with the @samp{--srcdir} option to specify where to find the source.
33090 (You also need to specify a path to find @file{configure}
33091 itself from your working directory. If the path to @file{configure}
33092 would be the same as the argument to @samp{--srcdir}, you can leave out
33093 the @samp{--srcdir} option; it is assumed.)
33095 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
33096 separate directory for a Sun 4 like this:
33100 cd gdb-@value{GDBVN}
33103 ../gdb-@value{GDBVN}/configure sun4
33108 When @file{configure} builds a configuration using a remote source
33109 directory, it creates a tree for the binaries with the same structure
33110 (and using the same names) as the tree under the source directory. In
33111 the example, you'd find the Sun 4 library @file{libiberty.a} in the
33112 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
33113 @file{gdb-sun4/gdb}.
33115 Make sure that your path to the @file{configure} script has just one
33116 instance of @file{gdb} in it. If your path to @file{configure} looks
33117 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
33118 one subdirectory of @value{GDBN}, not the whole package. This leads to
33119 build errors about missing include files such as @file{bfd/bfd.h}.
33121 One popular reason to build several @value{GDBN} configurations in separate
33122 directories is to configure @value{GDBN} for cross-compiling (where
33123 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
33124 programs that run on another machine---the @dfn{target}).
33125 You specify a cross-debugging target by
33126 giving the @samp{--target=@var{target}} option to @file{configure}.
33128 When you run @code{make} to build a program or library, you must run
33129 it in a configured directory---whatever directory you were in when you
33130 called @file{configure} (or one of its subdirectories).
33132 The @code{Makefile} that @file{configure} generates in each source
33133 directory also runs recursively. If you type @code{make} in a source
33134 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
33135 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
33136 will build all the required libraries, and then build GDB.
33138 When you have multiple hosts or targets configured in separate
33139 directories, you can run @code{make} on them in parallel (for example,
33140 if they are NFS-mounted on each of the hosts); they will not interfere
33144 @section Specifying Names for Hosts and Targets
33146 The specifications used for hosts and targets in the @file{configure}
33147 script are based on a three-part naming scheme, but some short predefined
33148 aliases are also supported. The full naming scheme encodes three pieces
33149 of information in the following pattern:
33152 @var{architecture}-@var{vendor}-@var{os}
33155 For example, you can use the alias @code{sun4} as a @var{host} argument,
33156 or as the value for @var{target} in a @code{--target=@var{target}}
33157 option. The equivalent full name is @samp{sparc-sun-sunos4}.
33159 The @file{configure} script accompanying @value{GDBN} does not provide
33160 any query facility to list all supported host and target names or
33161 aliases. @file{configure} calls the Bourne shell script
33162 @code{config.sub} to map abbreviations to full names; you can read the
33163 script, if you wish, or you can use it to test your guesses on
33164 abbreviations---for example:
33167 % sh config.sub i386-linux
33169 % sh config.sub alpha-linux
33170 alpha-unknown-linux-gnu
33171 % sh config.sub hp9k700
33173 % sh config.sub sun4
33174 sparc-sun-sunos4.1.1
33175 % sh config.sub sun3
33176 m68k-sun-sunos4.1.1
33177 % sh config.sub i986v
33178 Invalid configuration `i986v': machine `i986v' not recognized
33182 @code{config.sub} is also distributed in the @value{GDBN} source
33183 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
33185 @node Configure Options
33186 @section @file{configure} Options
33188 Here is a summary of the @file{configure} options and arguments that
33189 are most often useful for building @value{GDBN}. @file{configure} also has
33190 several other options not listed here. @inforef{What Configure
33191 Does,,configure.info}, for a full explanation of @file{configure}.
33194 configure @r{[}--help@r{]}
33195 @r{[}--prefix=@var{dir}@r{]}
33196 @r{[}--exec-prefix=@var{dir}@r{]}
33197 @r{[}--srcdir=@var{dirname}@r{]}
33198 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
33199 @r{[}--target=@var{target}@r{]}
33204 You may introduce options with a single @samp{-} rather than
33205 @samp{--} if you prefer; but you may abbreviate option names if you use
33210 Display a quick summary of how to invoke @file{configure}.
33212 @item --prefix=@var{dir}
33213 Configure the source to install programs and files under directory
33216 @item --exec-prefix=@var{dir}
33217 Configure the source to install programs under directory
33220 @c avoid splitting the warning from the explanation:
33222 @item --srcdir=@var{dirname}
33223 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
33224 @code{make} that implements the @code{VPATH} feature.}@*
33225 Use this option to make configurations in directories separate from the
33226 @value{GDBN} source directories. Among other things, you can use this to
33227 build (or maintain) several configurations simultaneously, in separate
33228 directories. @file{configure} writes configuration-specific files in
33229 the current directory, but arranges for them to use the source in the
33230 directory @var{dirname}. @file{configure} creates directories under
33231 the working directory in parallel to the source directories below
33234 @item --norecursion
33235 Configure only the directory level where @file{configure} is executed; do not
33236 propagate configuration to subdirectories.
33238 @item --target=@var{target}
33239 Configure @value{GDBN} for cross-debugging programs running on the specified
33240 @var{target}. Without this option, @value{GDBN} is configured to debug
33241 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
33243 There is no convenient way to generate a list of all available targets.
33245 @item @var{host} @dots{}
33246 Configure @value{GDBN} to run on the specified @var{host}.
33248 There is no convenient way to generate a list of all available hosts.
33251 There are many other options available as well, but they are generally
33252 needed for special purposes only.
33254 @node System-wide configuration
33255 @section System-wide configuration and settings
33256 @cindex system-wide init file
33258 @value{GDBN} can be configured to have a system-wide init file;
33259 this file will be read and executed at startup (@pxref{Startup, , What
33260 @value{GDBN} does during startup}).
33262 Here is the corresponding configure option:
33265 @item --with-system-gdbinit=@var{file}
33266 Specify that the default location of the system-wide init file is
33270 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
33271 it may be subject to relocation. Two possible cases:
33275 If the default location of this init file contains @file{$prefix},
33276 it will be subject to relocation. Suppose that the configure options
33277 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
33278 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
33279 init file is looked for as @file{$install/etc/gdbinit} instead of
33280 @file{$prefix/etc/gdbinit}.
33283 By contrast, if the default location does not contain the prefix,
33284 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
33285 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
33286 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
33287 wherever @value{GDBN} is installed.
33290 If the configured location of the system-wide init file (as given by the
33291 @option{--with-system-gdbinit} option at configure time) is in the
33292 data-directory (as specified by @option{--with-gdb-datadir} at configure
33293 time) or in one of its subdirectories, then @value{GDBN} will look for the
33294 system-wide init file in the directory specified by the
33295 @option{--data-directory} command-line option.
33296 Note that the system-wide init file is only read once, during @value{GDBN}
33297 initialization. If the data-directory is changed after @value{GDBN} has
33298 started with the @code{set data-directory} command, the file will not be
33302 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
33305 @node System-wide Configuration Scripts
33306 @subsection Installed System-wide Configuration Scripts
33307 @cindex system-wide configuration scripts
33309 The @file{system-gdbinit} directory, located inside the data-directory
33310 (as specified by @option{--with-gdb-datadir} at configure time) contains
33311 a number of scripts which can be used as system-wide init files. To
33312 automatically source those scripts at startup, @value{GDBN} should be
33313 configured with @option{--with-system-gdbinit}. Otherwise, any user
33314 should be able to source them by hand as needed.
33316 The following scripts are currently available:
33319 @item @file{elinos.py}
33321 @cindex ELinOS system-wide configuration script
33322 This script is useful when debugging a program on an ELinOS target.
33323 It takes advantage of the environment variables defined in a standard
33324 ELinOS environment in order to determine the location of the system
33325 shared libraries, and then sets the @samp{solib-absolute-prefix}
33326 and @samp{solib-search-path} variables appropriately.
33328 @item @file{wrs-linux.py}
33329 @pindex wrs-linux.py
33330 @cindex Wind River Linux system-wide configuration script
33331 This script is useful when debugging a program on a target running
33332 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
33333 the host-side sysroot used by the target system.
33337 @node Maintenance Commands
33338 @appendix Maintenance Commands
33339 @cindex maintenance commands
33340 @cindex internal commands
33342 In addition to commands intended for @value{GDBN} users, @value{GDBN}
33343 includes a number of commands intended for @value{GDBN} developers,
33344 that are not documented elsewhere in this manual. These commands are
33345 provided here for reference. (For commands that turn on debugging
33346 messages, see @ref{Debugging Output}.)
33349 @kindex maint agent
33350 @kindex maint agent-eval
33351 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
33352 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
33353 Translate the given @var{expression} into remote agent bytecodes.
33354 This command is useful for debugging the Agent Expression mechanism
33355 (@pxref{Agent Expressions}). The @samp{agent} version produces an
33356 expression useful for data collection, such as by tracepoints, while
33357 @samp{maint agent-eval} produces an expression that evaluates directly
33358 to a result. For instance, a collection expression for @code{globa +
33359 globb} will include bytecodes to record four bytes of memory at each
33360 of the addresses of @code{globa} and @code{globb}, while discarding
33361 the result of the addition, while an evaluation expression will do the
33362 addition and return the sum.
33363 If @code{-at} is given, generate remote agent bytecode for @var{location}.
33364 If not, generate remote agent bytecode for current frame PC address.
33366 @kindex maint agent-printf
33367 @item maint agent-printf @var{format},@var{expr},...
33368 Translate the given format string and list of argument expressions
33369 into remote agent bytecodes and display them as a disassembled list.
33370 This command is useful for debugging the agent version of dynamic
33371 printf (@pxref{Dynamic Printf}).
33373 @kindex maint info breakpoints
33374 @item @anchor{maint info breakpoints}maint info breakpoints
33375 Using the same format as @samp{info breakpoints}, display both the
33376 breakpoints you've set explicitly, and those @value{GDBN} is using for
33377 internal purposes. Internal breakpoints are shown with negative
33378 breakpoint numbers. The type column identifies what kind of breakpoint
33383 Normal, explicitly set breakpoint.
33386 Normal, explicitly set watchpoint.
33389 Internal breakpoint, used to handle correctly stepping through
33390 @code{longjmp} calls.
33392 @item longjmp resume
33393 Internal breakpoint at the target of a @code{longjmp}.
33396 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
33399 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
33402 Shared library events.
33406 @kindex maint info bfds
33407 @item maint info bfds
33408 This prints information about each @code{bfd} object that is known to
33409 @value{GDBN}. @xref{Top, , BFD, bfd, The Binary File Descriptor Library}.
33411 @kindex set displaced-stepping
33412 @kindex show displaced-stepping
33413 @cindex displaced stepping support
33414 @cindex out-of-line single-stepping
33415 @item set displaced-stepping
33416 @itemx show displaced-stepping
33417 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
33418 if the target supports it. Displaced stepping is a way to single-step
33419 over breakpoints without removing them from the inferior, by executing
33420 an out-of-line copy of the instruction that was originally at the
33421 breakpoint location. It is also known as out-of-line single-stepping.
33424 @item set displaced-stepping on
33425 If the target architecture supports it, @value{GDBN} will use
33426 displaced stepping to step over breakpoints.
33428 @item set displaced-stepping off
33429 @value{GDBN} will not use displaced stepping to step over breakpoints,
33430 even if such is supported by the target architecture.
33432 @cindex non-stop mode, and @samp{set displaced-stepping}
33433 @item set displaced-stepping auto
33434 This is the default mode. @value{GDBN} will use displaced stepping
33435 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
33436 architecture supports displaced stepping.
33439 @kindex maint check-psymtabs
33440 @item maint check-psymtabs
33441 Check the consistency of currently expanded psymtabs versus symtabs.
33442 Use this to check, for example, whether a symbol is in one but not the other.
33444 @kindex maint check-symtabs
33445 @item maint check-symtabs
33446 Check the consistency of currently expanded symtabs.
33448 @kindex maint expand-symtabs
33449 @item maint expand-symtabs [@var{regexp}]
33450 Expand symbol tables.
33451 If @var{regexp} is specified, only expand symbol tables for file
33452 names matching @var{regexp}.
33454 @kindex maint set catch-demangler-crashes
33455 @kindex maint show catch-demangler-crashes
33456 @cindex demangler crashes
33457 @item maint set catch-demangler-crashes [on|off]
33458 @itemx maint show catch-demangler-crashes
33459 Control whether @value{GDBN} should attempt to catch crashes in the
33460 symbol name demangler. The default is to attempt to catch crashes.
33461 If enabled, the first time a crash is caught, a core file is created,
33462 the offending symbol is displayed and the user is presented with the
33463 option to terminate the current session.
33465 @kindex maint cplus first_component
33466 @item maint cplus first_component @var{name}
33467 Print the first C@t{++} class/namespace component of @var{name}.
33469 @kindex maint cplus namespace
33470 @item maint cplus namespace
33471 Print the list of possible C@t{++} namespaces.
33473 @kindex maint demangle
33474 @item maint demangle @var{name}
33475 Demangle a C@t{++} or Objective-C mangled @var{name}.
33477 @kindex maint deprecate
33478 @kindex maint undeprecate
33479 @cindex deprecated commands
33480 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
33481 @itemx maint undeprecate @var{command}
33482 Deprecate or undeprecate the named @var{command}. Deprecated commands
33483 cause @value{GDBN} to issue a warning when you use them. The optional
33484 argument @var{replacement} says which newer command should be used in
33485 favor of the deprecated one; if it is given, @value{GDBN} will mention
33486 the replacement as part of the warning.
33488 @kindex maint dump-me
33489 @item maint dump-me
33490 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
33491 Cause a fatal signal in the debugger and force it to dump its core.
33492 This is supported only on systems which support aborting a program
33493 with the @code{SIGQUIT} signal.
33495 @kindex maint internal-error
33496 @kindex maint internal-warning
33497 @kindex maint demangler-warning
33498 @cindex demangler crashes
33499 @item maint internal-error @r{[}@var{message-text}@r{]}
33500 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
33501 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
33503 Cause @value{GDBN} to call the internal function @code{internal_error},
33504 @code{internal_warning} or @code{demangler_warning} and hence behave
33505 as though an internal problam has been detected. In addition to
33506 reporting the internal problem, these functions give the user the
33507 opportunity to either quit @value{GDBN} or (for @code{internal_error}
33508 and @code{internal_warning}) create a core file of the current
33509 @value{GDBN} session.
33511 These commands take an optional parameter @var{message-text} that is
33512 used as the text of the error or warning message.
33514 Here's an example of using @code{internal-error}:
33517 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
33518 @dots{}/maint.c:121: internal-error: testing, 1, 2
33519 A problem internal to GDB has been detected. Further
33520 debugging may prove unreliable.
33521 Quit this debugging session? (y or n) @kbd{n}
33522 Create a core file? (y or n) @kbd{n}
33526 @cindex @value{GDBN} internal error
33527 @cindex internal errors, control of @value{GDBN} behavior
33528 @cindex demangler crashes
33530 @kindex maint set internal-error
33531 @kindex maint show internal-error
33532 @kindex maint set internal-warning
33533 @kindex maint show internal-warning
33534 @kindex maint set demangler-warning
33535 @kindex maint show demangler-warning
33536 @item maint set internal-error @var{action} [ask|yes|no]
33537 @itemx maint show internal-error @var{action}
33538 @itemx maint set internal-warning @var{action} [ask|yes|no]
33539 @itemx maint show internal-warning @var{action}
33540 @itemx maint set demangler-warning @var{action} [ask|yes|no]
33541 @itemx maint show demangler-warning @var{action}
33542 When @value{GDBN} reports an internal problem (error or warning) it
33543 gives the user the opportunity to both quit @value{GDBN} and create a
33544 core file of the current @value{GDBN} session. These commands let you
33545 override the default behaviour for each particular @var{action},
33546 described in the table below.
33550 You can specify that @value{GDBN} should always (yes) or never (no)
33551 quit. The default is to ask the user what to do.
33554 You can specify that @value{GDBN} should always (yes) or never (no)
33555 create a core file. The default is to ask the user what to do. Note
33556 that there is no @code{corefile} option for @code{demangler-warning}:
33557 demangler warnings always create a core file and this cannot be
33561 @kindex maint packet
33562 @item maint packet @var{text}
33563 If @value{GDBN} is talking to an inferior via the serial protocol,
33564 then this command sends the string @var{text} to the inferior, and
33565 displays the response packet. @value{GDBN} supplies the initial
33566 @samp{$} character, the terminating @samp{#} character, and the
33569 @kindex maint print architecture
33570 @item maint print architecture @r{[}@var{file}@r{]}
33571 Print the entire architecture configuration. The optional argument
33572 @var{file} names the file where the output goes.
33574 @kindex maint print c-tdesc
33575 @item maint print c-tdesc
33576 Print the current target description (@pxref{Target Descriptions}) as
33577 a C source file. The created source file can be used in @value{GDBN}
33578 when an XML parser is not available to parse the description.
33580 @kindex maint print dummy-frames
33581 @item maint print dummy-frames
33582 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
33585 (@value{GDBP}) @kbd{b add}
33587 (@value{GDBP}) @kbd{print add(2,3)}
33588 Breakpoint 2, add (a=2, b=3) at @dots{}
33590 The program being debugged stopped while in a function called from GDB.
33592 (@value{GDBP}) @kbd{maint print dummy-frames}
33593 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
33597 Takes an optional file parameter.
33599 @kindex maint print registers
33600 @kindex maint print raw-registers
33601 @kindex maint print cooked-registers
33602 @kindex maint print register-groups
33603 @kindex maint print remote-registers
33604 @item maint print registers @r{[}@var{file}@r{]}
33605 @itemx maint print raw-registers @r{[}@var{file}@r{]}
33606 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
33607 @itemx maint print register-groups @r{[}@var{file}@r{]}
33608 @itemx maint print remote-registers @r{[}@var{file}@r{]}
33609 Print @value{GDBN}'s internal register data structures.
33611 The command @code{maint print raw-registers} includes the contents of
33612 the raw register cache; the command @code{maint print
33613 cooked-registers} includes the (cooked) value of all registers,
33614 including registers which aren't available on the target nor visible
33615 to user; the command @code{maint print register-groups} includes the
33616 groups that each register is a member of; and the command @code{maint
33617 print remote-registers} includes the remote target's register numbers
33618 and offsets in the `G' packets.
33620 These commands take an optional parameter, a file name to which to
33621 write the information.
33623 @kindex maint print reggroups
33624 @item maint print reggroups @r{[}@var{file}@r{]}
33625 Print @value{GDBN}'s internal register group data structures. The
33626 optional argument @var{file} tells to what file to write the
33629 The register groups info looks like this:
33632 (@value{GDBP}) @kbd{maint print reggroups}
33645 This command forces @value{GDBN} to flush its internal register cache.
33647 @kindex maint print objfiles
33648 @cindex info for known object files
33649 @item maint print objfiles @r{[}@var{regexp}@r{]}
33650 Print a dump of all known object files.
33651 If @var{regexp} is specified, only print object files whose names
33652 match @var{regexp}. For each object file, this command prints its name,
33653 address in memory, and all of its psymtabs and symtabs.
33655 @kindex maint print user-registers
33656 @cindex user registers
33657 @item maint print user-registers
33658 List all currently available @dfn{user registers}. User registers
33659 typically provide alternate names for actual hardware registers. They
33660 include the four ``standard'' registers @code{$fp}, @code{$pc},
33661 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
33662 registers can be used in expressions in the same way as the canonical
33663 register names, but only the latter are listed by the @code{info
33664 registers} and @code{maint print registers} commands.
33666 @kindex maint print section-scripts
33667 @cindex info for known .debug_gdb_scripts-loaded scripts
33668 @item maint print section-scripts [@var{regexp}]
33669 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
33670 If @var{regexp} is specified, only print scripts loaded by object files
33671 matching @var{regexp}.
33672 For each script, this command prints its name as specified in the objfile,
33673 and the full path if known.
33674 @xref{dotdebug_gdb_scripts section}.
33676 @kindex maint print statistics
33677 @cindex bcache statistics
33678 @item maint print statistics
33679 This command prints, for each object file in the program, various data
33680 about that object file followed by the byte cache (@dfn{bcache})
33681 statistics for the object file. The objfile data includes the number
33682 of minimal, partial, full, and stabs symbols, the number of types
33683 defined by the objfile, the number of as yet unexpanded psym tables,
33684 the number of line tables and string tables, and the amount of memory
33685 used by the various tables. The bcache statistics include the counts,
33686 sizes, and counts of duplicates of all and unique objects, max,
33687 average, and median entry size, total memory used and its overhead and
33688 savings, and various measures of the hash table size and chain
33691 @kindex maint print target-stack
33692 @cindex target stack description
33693 @item maint print target-stack
33694 A @dfn{target} is an interface between the debugger and a particular
33695 kind of file or process. Targets can be stacked in @dfn{strata},
33696 so that more than one target can potentially respond to a request.
33697 In particular, memory accesses will walk down the stack of targets
33698 until they find a target that is interested in handling that particular
33701 This command prints a short description of each layer that was pushed on
33702 the @dfn{target stack}, starting from the top layer down to the bottom one.
33704 @kindex maint print type
33705 @cindex type chain of a data type
33706 @item maint print type @var{expr}
33707 Print the type chain for a type specified by @var{expr}. The argument
33708 can be either a type name or a symbol. If it is a symbol, the type of
33709 that symbol is described. The type chain produced by this command is
33710 a recursive definition of the data type as stored in @value{GDBN}'s
33711 data structures, including its flags and contained types.
33713 @kindex maint set dwarf2 always-disassemble
33714 @kindex maint show dwarf2 always-disassemble
33715 @item maint set dwarf2 always-disassemble
33716 @item maint show dwarf2 always-disassemble
33717 Control the behavior of @code{info address} when using DWARF debugging
33720 The default is @code{off}, which means that @value{GDBN} should try to
33721 describe a variable's location in an easily readable format. When
33722 @code{on}, @value{GDBN} will instead display the DWARF location
33723 expression in an assembly-like format. Note that some locations are
33724 too complex for @value{GDBN} to describe simply; in this case you will
33725 always see the disassembly form.
33727 Here is an example of the resulting disassembly:
33730 (gdb) info addr argc
33731 Symbol "argc" is a complex DWARF expression:
33735 For more information on these expressions, see
33736 @uref{http://www.dwarfstd.org/, the DWARF standard}.
33738 @kindex maint set dwarf2 max-cache-age
33739 @kindex maint show dwarf2 max-cache-age
33740 @item maint set dwarf2 max-cache-age
33741 @itemx maint show dwarf2 max-cache-age
33742 Control the DWARF 2 compilation unit cache.
33744 @cindex DWARF 2 compilation units cache
33745 In object files with inter-compilation-unit references, such as those
33746 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
33747 reader needs to frequently refer to previously read compilation units.
33748 This setting controls how long a compilation unit will remain in the
33749 cache if it is not referenced. A higher limit means that cached
33750 compilation units will be stored in memory longer, and more total
33751 memory will be used. Setting it to zero disables caching, which will
33752 slow down @value{GDBN} startup, but reduce memory consumption.
33754 @kindex maint set profile
33755 @kindex maint show profile
33756 @cindex profiling GDB
33757 @item maint set profile
33758 @itemx maint show profile
33759 Control profiling of @value{GDBN}.
33761 Profiling will be disabled until you use the @samp{maint set profile}
33762 command to enable it. When you enable profiling, the system will begin
33763 collecting timing and execution count data; when you disable profiling or
33764 exit @value{GDBN}, the results will be written to a log file. Remember that
33765 if you use profiling, @value{GDBN} will overwrite the profiling log file
33766 (often called @file{gmon.out}). If you have a record of important profiling
33767 data in a @file{gmon.out} file, be sure to move it to a safe location.
33769 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
33770 compiled with the @samp{-pg} compiler option.
33772 @kindex maint set show-debug-regs
33773 @kindex maint show show-debug-regs
33774 @cindex hardware debug registers
33775 @item maint set show-debug-regs
33776 @itemx maint show show-debug-regs
33777 Control whether to show variables that mirror the hardware debug
33778 registers. Use @code{on} to enable, @code{off} to disable. If
33779 enabled, the debug registers values are shown when @value{GDBN} inserts or
33780 removes a hardware breakpoint or watchpoint, and when the inferior
33781 triggers a hardware-assisted breakpoint or watchpoint.
33783 @kindex maint set show-all-tib
33784 @kindex maint show show-all-tib
33785 @item maint set show-all-tib
33786 @itemx maint show show-all-tib
33787 Control whether to show all non zero areas within a 1k block starting
33788 at thread local base, when using the @samp{info w32 thread-information-block}
33791 @kindex maint set target-async
33792 @kindex maint show target-async
33793 @item maint set target-async
33794 @itemx maint show target-async
33795 This controls whether @value{GDBN} targets operate in synchronous or
33796 asynchronous mode (@pxref{Background Execution}). Normally the
33797 default is asynchronous, if it is available; but this can be changed
33798 to more easily debug problems occurring only in synchronous mode.
33800 @kindex maint set per-command
33801 @kindex maint show per-command
33802 @item maint set per-command
33803 @itemx maint show per-command
33804 @cindex resources used by commands
33806 @value{GDBN} can display the resources used by each command.
33807 This is useful in debugging performance problems.
33810 @item maint set per-command space [on|off]
33811 @itemx maint show per-command space
33812 Enable or disable the printing of the memory used by GDB for each command.
33813 If enabled, @value{GDBN} will display how much memory each command
33814 took, following the command's own output.
33815 This can also be requested by invoking @value{GDBN} with the
33816 @option{--statistics} command-line switch (@pxref{Mode Options}).
33818 @item maint set per-command time [on|off]
33819 @itemx maint show per-command time
33820 Enable or disable the printing of the execution time of @value{GDBN}
33822 If enabled, @value{GDBN} will display how much time it
33823 took to execute each command, following the command's own output.
33824 Both CPU time and wallclock time are printed.
33825 Printing both is useful when trying to determine whether the cost is
33826 CPU or, e.g., disk/network latency.
33827 Note that the CPU time printed is for @value{GDBN} only, it does not include
33828 the execution time of the inferior because there's no mechanism currently
33829 to compute how much time was spent by @value{GDBN} and how much time was
33830 spent by the program been debugged.
33831 This can also be requested by invoking @value{GDBN} with the
33832 @option{--statistics} command-line switch (@pxref{Mode Options}).
33834 @item maint set per-command symtab [on|off]
33835 @itemx maint show per-command symtab
33836 Enable or disable the printing of basic symbol table statistics
33838 If enabled, @value{GDBN} will display the following information:
33842 number of symbol tables
33844 number of primary symbol tables
33846 number of blocks in the blockvector
33850 @kindex maint space
33851 @cindex memory used by commands
33852 @item maint space @var{value}
33853 An alias for @code{maint set per-command space}.
33854 A non-zero value enables it, zero disables it.
33857 @cindex time of command execution
33858 @item maint time @var{value}
33859 An alias for @code{maint set per-command time}.
33860 A non-zero value enables it, zero disables it.
33862 @kindex maint translate-address
33863 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
33864 Find the symbol stored at the location specified by the address
33865 @var{addr} and an optional section name @var{section}. If found,
33866 @value{GDBN} prints the name of the closest symbol and an offset from
33867 the symbol's location to the specified address. This is similar to
33868 the @code{info address} command (@pxref{Symbols}), except that this
33869 command also allows to find symbols in other sections.
33871 If section was not specified, the section in which the symbol was found
33872 is also printed. For dynamically linked executables, the name of
33873 executable or shared library containing the symbol is printed as well.
33877 The following command is useful for non-interactive invocations of
33878 @value{GDBN}, such as in the test suite.
33881 @item set watchdog @var{nsec}
33882 @kindex set watchdog
33883 @cindex watchdog timer
33884 @cindex timeout for commands
33885 Set the maximum number of seconds @value{GDBN} will wait for the
33886 target operation to finish. If this time expires, @value{GDBN}
33887 reports and error and the command is aborted.
33889 @item show watchdog
33890 Show the current setting of the target wait timeout.
33893 @node Remote Protocol
33894 @appendix @value{GDBN} Remote Serial Protocol
33899 * Stop Reply Packets::
33900 * General Query Packets::
33901 * Architecture-Specific Protocol Details::
33902 * Tracepoint Packets::
33903 * Host I/O Packets::
33905 * Notification Packets::
33906 * Remote Non-Stop::
33907 * Packet Acknowledgment::
33909 * File-I/O Remote Protocol Extension::
33910 * Library List Format::
33911 * Library List Format for SVR4 Targets::
33912 * Memory Map Format::
33913 * Thread List Format::
33914 * Traceframe Info Format::
33915 * Branch Trace Format::
33921 There may be occasions when you need to know something about the
33922 protocol---for example, if there is only one serial port to your target
33923 machine, you might want your program to do something special if it
33924 recognizes a packet meant for @value{GDBN}.
33926 In the examples below, @samp{->} and @samp{<-} are used to indicate
33927 transmitted and received data, respectively.
33929 @cindex protocol, @value{GDBN} remote serial
33930 @cindex serial protocol, @value{GDBN} remote
33931 @cindex remote serial protocol
33932 All @value{GDBN} commands and responses (other than acknowledgments
33933 and notifications, see @ref{Notification Packets}) are sent as a
33934 @var{packet}. A @var{packet} is introduced with the character
33935 @samp{$}, the actual @var{packet-data}, and the terminating character
33936 @samp{#} followed by a two-digit @var{checksum}:
33939 @code{$}@var{packet-data}@code{#}@var{checksum}
33943 @cindex checksum, for @value{GDBN} remote
33945 The two-digit @var{checksum} is computed as the modulo 256 sum of all
33946 characters between the leading @samp{$} and the trailing @samp{#} (an
33947 eight bit unsigned checksum).
33949 Implementors should note that prior to @value{GDBN} 5.0 the protocol
33950 specification also included an optional two-digit @var{sequence-id}:
33953 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
33956 @cindex sequence-id, for @value{GDBN} remote
33958 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
33959 has never output @var{sequence-id}s. Stubs that handle packets added
33960 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
33962 When either the host or the target machine receives a packet, the first
33963 response expected is an acknowledgment: either @samp{+} (to indicate
33964 the package was received correctly) or @samp{-} (to request
33968 -> @code{$}@var{packet-data}@code{#}@var{checksum}
33973 The @samp{+}/@samp{-} acknowledgments can be disabled
33974 once a connection is established.
33975 @xref{Packet Acknowledgment}, for details.
33977 The host (@value{GDBN}) sends @var{command}s, and the target (the
33978 debugging stub incorporated in your program) sends a @var{response}. In
33979 the case of step and continue @var{command}s, the response is only sent
33980 when the operation has completed, and the target has again stopped all
33981 threads in all attached processes. This is the default all-stop mode
33982 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
33983 execution mode; see @ref{Remote Non-Stop}, for details.
33985 @var{packet-data} consists of a sequence of characters with the
33986 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
33989 @cindex remote protocol, field separator
33990 Fields within the packet should be separated using @samp{,} @samp{;} or
33991 @samp{:}. Except where otherwise noted all numbers are represented in
33992 @sc{hex} with leading zeros suppressed.
33994 Implementors should note that prior to @value{GDBN} 5.0, the character
33995 @samp{:} could not appear as the third character in a packet (as it
33996 would potentially conflict with the @var{sequence-id}).
33998 @cindex remote protocol, binary data
33999 @anchor{Binary Data}
34000 Binary data in most packets is encoded either as two hexadecimal
34001 digits per byte of binary data. This allowed the traditional remote
34002 protocol to work over connections which were only seven-bit clean.
34003 Some packets designed more recently assume an eight-bit clean
34004 connection, and use a more efficient encoding to send and receive
34007 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
34008 as an escape character. Any escaped byte is transmitted as the escape
34009 character followed by the original character XORed with @code{0x20}.
34010 For example, the byte @code{0x7d} would be transmitted as the two
34011 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
34012 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
34013 @samp{@}}) must always be escaped. Responses sent by the stub
34014 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
34015 is not interpreted as the start of a run-length encoded sequence
34018 Response @var{data} can be run-length encoded to save space.
34019 Run-length encoding replaces runs of identical characters with one
34020 instance of the repeated character, followed by a @samp{*} and a
34021 repeat count. The repeat count is itself sent encoded, to avoid
34022 binary characters in @var{data}: a value of @var{n} is sent as
34023 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
34024 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
34025 code 32) for a repeat count of 3. (This is because run-length
34026 encoding starts to win for counts 3 or more.) Thus, for example,
34027 @samp{0* } is a run-length encoding of ``0000'': the space character
34028 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
34031 The printable characters @samp{#} and @samp{$} or with a numeric value
34032 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
34033 seven repeats (@samp{$}) can be expanded using a repeat count of only
34034 five (@samp{"}). For example, @samp{00000000} can be encoded as
34037 The error response returned for some packets includes a two character
34038 error number. That number is not well defined.
34040 @cindex empty response, for unsupported packets
34041 For any @var{command} not supported by the stub, an empty response
34042 (@samp{$#00}) should be returned. That way it is possible to extend the
34043 protocol. A newer @value{GDBN} can tell if a packet is supported based
34046 At a minimum, a stub is required to support the @samp{g} and @samp{G}
34047 commands for register access, and the @samp{m} and @samp{M} commands
34048 for memory access. Stubs that only control single-threaded targets
34049 can implement run control with the @samp{c} (continue), and @samp{s}
34050 (step) commands. Stubs that support multi-threading targets should
34051 support the @samp{vCont} command. All other commands are optional.
34056 The following table provides a complete list of all currently defined
34057 @var{command}s and their corresponding response @var{data}.
34058 @xref{File-I/O Remote Protocol Extension}, for details about the File
34059 I/O extension of the remote protocol.
34061 Each packet's description has a template showing the packet's overall
34062 syntax, followed by an explanation of the packet's meaning. We
34063 include spaces in some of the templates for clarity; these are not
34064 part of the packet's syntax. No @value{GDBN} packet uses spaces to
34065 separate its components. For example, a template like @samp{foo
34066 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
34067 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
34068 @var{baz}. @value{GDBN} does not transmit a space character between the
34069 @samp{foo} and the @var{bar}, or between the @var{bar} and the
34072 @cindex @var{thread-id}, in remote protocol
34073 @anchor{thread-id syntax}
34074 Several packets and replies include a @var{thread-id} field to identify
34075 a thread. Normally these are positive numbers with a target-specific
34076 interpretation, formatted as big-endian hex strings. A @var{thread-id}
34077 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
34080 In addition, the remote protocol supports a multiprocess feature in
34081 which the @var{thread-id} syntax is extended to optionally include both
34082 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
34083 The @var{pid} (process) and @var{tid} (thread) components each have the
34084 format described above: a positive number with target-specific
34085 interpretation formatted as a big-endian hex string, literal @samp{-1}
34086 to indicate all processes or threads (respectively), or @samp{0} to
34087 indicate an arbitrary process or thread. Specifying just a process, as
34088 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
34089 error to specify all processes but a specific thread, such as
34090 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
34091 for those packets and replies explicitly documented to include a process
34092 ID, rather than a @var{thread-id}.
34094 The multiprocess @var{thread-id} syntax extensions are only used if both
34095 @value{GDBN} and the stub report support for the @samp{multiprocess}
34096 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
34099 Note that all packet forms beginning with an upper- or lower-case
34100 letter, other than those described here, are reserved for future use.
34102 Here are the packet descriptions.
34107 @cindex @samp{!} packet
34108 @anchor{extended mode}
34109 Enable extended mode. In extended mode, the remote server is made
34110 persistent. The @samp{R} packet is used to restart the program being
34116 The remote target both supports and has enabled extended mode.
34120 @cindex @samp{?} packet
34122 Indicate the reason the target halted. The reply is the same as for
34123 step and continue. This packet has a special interpretation when the
34124 target is in non-stop mode; see @ref{Remote Non-Stop}.
34127 @xref{Stop Reply Packets}, for the reply specifications.
34129 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
34130 @cindex @samp{A} packet
34131 Initialized @code{argv[]} array passed into program. @var{arglen}
34132 specifies the number of bytes in the hex encoded byte stream
34133 @var{arg}. See @code{gdbserver} for more details.
34138 The arguments were set.
34144 @cindex @samp{b} packet
34145 (Don't use this packet; its behavior is not well-defined.)
34146 Change the serial line speed to @var{baud}.
34148 JTC: @emph{When does the transport layer state change? When it's
34149 received, or after the ACK is transmitted. In either case, there are
34150 problems if the command or the acknowledgment packet is dropped.}
34152 Stan: @emph{If people really wanted to add something like this, and get
34153 it working for the first time, they ought to modify ser-unix.c to send
34154 some kind of out-of-band message to a specially-setup stub and have the
34155 switch happen "in between" packets, so that from remote protocol's point
34156 of view, nothing actually happened.}
34158 @item B @var{addr},@var{mode}
34159 @cindex @samp{B} packet
34160 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
34161 breakpoint at @var{addr}.
34163 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
34164 (@pxref{insert breakpoint or watchpoint packet}).
34166 @cindex @samp{bc} packet
34169 Backward continue. Execute the target system in reverse. No parameter.
34170 @xref{Reverse Execution}, for more information.
34173 @xref{Stop Reply Packets}, for the reply specifications.
34175 @cindex @samp{bs} packet
34178 Backward single step. Execute one instruction in reverse. No parameter.
34179 @xref{Reverse Execution}, for more information.
34182 @xref{Stop Reply Packets}, for the reply specifications.
34184 @item c @r{[}@var{addr}@r{]}
34185 @cindex @samp{c} packet
34186 Continue at @var{addr}, which is the address to resume. If @var{addr}
34187 is omitted, resume at current address.
34189 This packet is deprecated for multi-threading support. @xref{vCont
34193 @xref{Stop Reply Packets}, for the reply specifications.
34195 @item C @var{sig}@r{[};@var{addr}@r{]}
34196 @cindex @samp{C} packet
34197 Continue with signal @var{sig} (hex signal number). If
34198 @samp{;@var{addr}} is omitted, resume at same address.
34200 This packet is deprecated for multi-threading support. @xref{vCont
34204 @xref{Stop Reply Packets}, for the reply specifications.
34207 @cindex @samp{d} packet
34210 Don't use this packet; instead, define a general set packet
34211 (@pxref{General Query Packets}).
34215 @cindex @samp{D} packet
34216 The first form of the packet is used to detach @value{GDBN} from the
34217 remote system. It is sent to the remote target
34218 before @value{GDBN} disconnects via the @code{detach} command.
34220 The second form, including a process ID, is used when multiprocess
34221 protocol extensions are enabled (@pxref{multiprocess extensions}), to
34222 detach only a specific process. The @var{pid} is specified as a
34223 big-endian hex string.
34233 @item F @var{RC},@var{EE},@var{CF};@var{XX}
34234 @cindex @samp{F} packet
34235 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
34236 This is part of the File-I/O protocol extension. @xref{File-I/O
34237 Remote Protocol Extension}, for the specification.
34240 @anchor{read registers packet}
34241 @cindex @samp{g} packet
34242 Read general registers.
34246 @item @var{XX@dots{}}
34247 Each byte of register data is described by two hex digits. The bytes
34248 with the register are transmitted in target byte order. The size of
34249 each register and their position within the @samp{g} packet are
34250 determined by the @value{GDBN} internal gdbarch functions
34251 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
34252 specification of several standard @samp{g} packets is specified below.
34254 When reading registers from a trace frame (@pxref{Analyze Collected
34255 Data,,Using the Collected Data}), the stub may also return a string of
34256 literal @samp{x}'s in place of the register data digits, to indicate
34257 that the corresponding register has not been collected, thus its value
34258 is unavailable. For example, for an architecture with 4 registers of
34259 4 bytes each, the following reply indicates to @value{GDBN} that
34260 registers 0 and 2 have not been collected, while registers 1 and 3
34261 have been collected, and both have zero value:
34265 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
34272 @item G @var{XX@dots{}}
34273 @cindex @samp{G} packet
34274 Write general registers. @xref{read registers packet}, for a
34275 description of the @var{XX@dots{}} data.
34285 @item H @var{op} @var{thread-id}
34286 @cindex @samp{H} packet
34287 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
34288 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
34289 should be @samp{c} for step and continue operations (note that this
34290 is deprecated, supporting the @samp{vCont} command is a better
34291 option), and @samp{g} for other operations. The thread designator
34292 @var{thread-id} has the format and interpretation described in
34293 @ref{thread-id syntax}.
34304 @c 'H': How restrictive (or permissive) is the thread model. If a
34305 @c thread is selected and stopped, are other threads allowed
34306 @c to continue to execute? As I mentioned above, I think the
34307 @c semantics of each command when a thread is selected must be
34308 @c described. For example:
34310 @c 'g': If the stub supports threads and a specific thread is
34311 @c selected, returns the register block from that thread;
34312 @c otherwise returns current registers.
34314 @c 'G' If the stub supports threads and a specific thread is
34315 @c selected, sets the registers of the register block of
34316 @c that thread; otherwise sets current registers.
34318 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
34319 @anchor{cycle step packet}
34320 @cindex @samp{i} packet
34321 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
34322 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
34323 step starting at that address.
34326 @cindex @samp{I} packet
34327 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
34331 @cindex @samp{k} packet
34334 The exact effect of this packet is not specified.
34336 For a bare-metal target, it may power cycle or reset the target
34337 system. For that reason, the @samp{k} packet has no reply.
34339 For a single-process target, it may kill that process if possible.
34341 A multiple-process target may choose to kill just one process, or all
34342 that are under @value{GDBN}'s control. For more precise control, use
34343 the vKill packet (@pxref{vKill packet}).
34345 If the target system immediately closes the connection in response to
34346 @samp{k}, @value{GDBN} does not consider the lack of packet
34347 acknowledgment to be an error, and assumes the kill was successful.
34349 If connected using @kbd{target extended-remote}, and the target does
34350 not close the connection in response to a kill request, @value{GDBN}
34351 probes the target state as if a new connection was opened
34352 (@pxref{? packet}).
34354 @item m @var{addr},@var{length}
34355 @cindex @samp{m} packet
34356 Read @var{length} bytes of memory starting at address @var{addr}.
34357 Note that @var{addr} may not be aligned to any particular boundary.
34359 The stub need not use any particular size or alignment when gathering
34360 data from memory for the response; even if @var{addr} is word-aligned
34361 and @var{length} is a multiple of the word size, the stub is free to
34362 use byte accesses, or not. For this reason, this packet may not be
34363 suitable for accessing memory-mapped I/O devices.
34364 @cindex alignment of remote memory accesses
34365 @cindex size of remote memory accesses
34366 @cindex memory, alignment and size of remote accesses
34370 @item @var{XX@dots{}}
34371 Memory contents; each byte is transmitted as a two-digit hexadecimal
34372 number. The reply may contain fewer bytes than requested if the
34373 server was able to read only part of the region of memory.
34378 @item M @var{addr},@var{length}:@var{XX@dots{}}
34379 @cindex @samp{M} packet
34380 Write @var{length} bytes of memory starting at address @var{addr}.
34381 The data is given by @var{XX@dots{}}; each byte is transmitted as a two-digit
34382 hexadecimal number.
34389 for an error (this includes the case where only part of the data was
34394 @cindex @samp{p} packet
34395 Read the value of register @var{n}; @var{n} is in hex.
34396 @xref{read registers packet}, for a description of how the returned
34397 register value is encoded.
34401 @item @var{XX@dots{}}
34402 the register's value
34406 Indicating an unrecognized @var{query}.
34409 @item P @var{n@dots{}}=@var{r@dots{}}
34410 @anchor{write register packet}
34411 @cindex @samp{P} packet
34412 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
34413 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
34414 digits for each byte in the register (target byte order).
34424 @item q @var{name} @var{params}@dots{}
34425 @itemx Q @var{name} @var{params}@dots{}
34426 @cindex @samp{q} packet
34427 @cindex @samp{Q} packet
34428 General query (@samp{q}) and set (@samp{Q}). These packets are
34429 described fully in @ref{General Query Packets}.
34432 @cindex @samp{r} packet
34433 Reset the entire system.
34435 Don't use this packet; use the @samp{R} packet instead.
34438 @cindex @samp{R} packet
34439 Restart the program being debugged. The @var{XX}, while needed, is ignored.
34440 This packet is only available in extended mode (@pxref{extended mode}).
34442 The @samp{R} packet has no reply.
34444 @item s @r{[}@var{addr}@r{]}
34445 @cindex @samp{s} packet
34446 Single step, resuming at @var{addr}. If
34447 @var{addr} is omitted, resume at same address.
34449 This packet is deprecated for multi-threading support. @xref{vCont
34453 @xref{Stop Reply Packets}, for the reply specifications.
34455 @item S @var{sig}@r{[};@var{addr}@r{]}
34456 @anchor{step with signal packet}
34457 @cindex @samp{S} packet
34458 Step with signal. This is analogous to the @samp{C} packet, but
34459 requests a single-step, rather than a normal resumption of execution.
34461 This packet is deprecated for multi-threading support. @xref{vCont
34465 @xref{Stop Reply Packets}, for the reply specifications.
34467 @item t @var{addr}:@var{PP},@var{MM}
34468 @cindex @samp{t} packet
34469 Search backwards starting at address @var{addr} for a match with pattern
34470 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
34471 There must be at least 3 digits in @var{addr}.
34473 @item T @var{thread-id}
34474 @cindex @samp{T} packet
34475 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
34480 thread is still alive
34486 Packets starting with @samp{v} are identified by a multi-letter name,
34487 up to the first @samp{;} or @samp{?} (or the end of the packet).
34489 @item vAttach;@var{pid}
34490 @cindex @samp{vAttach} packet
34491 Attach to a new process with the specified process ID @var{pid}.
34492 The process ID is a
34493 hexadecimal integer identifying the process. In all-stop mode, all
34494 threads in the attached process are stopped; in non-stop mode, it may be
34495 attached without being stopped if that is supported by the target.
34497 @c In non-stop mode, on a successful vAttach, the stub should set the
34498 @c current thread to a thread of the newly-attached process. After
34499 @c attaching, GDB queries for the attached process's thread ID with qC.
34500 @c Also note that, from a user perspective, whether or not the
34501 @c target is stopped on attach in non-stop mode depends on whether you
34502 @c use the foreground or background version of the attach command, not
34503 @c on what vAttach does; GDB does the right thing with respect to either
34504 @c stopping or restarting threads.
34506 This packet is only available in extended mode (@pxref{extended mode}).
34512 @item @r{Any stop packet}
34513 for success in all-stop mode (@pxref{Stop Reply Packets})
34515 for success in non-stop mode (@pxref{Remote Non-Stop})
34518 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
34519 @cindex @samp{vCont} packet
34520 @anchor{vCont packet}
34521 Resume the inferior, specifying different actions for each thread.
34522 If an action is specified with no @var{thread-id}, then it is applied to any
34523 threads that don't have a specific action specified; if no default action is
34524 specified then other threads should remain stopped in all-stop mode and
34525 in their current state in non-stop mode.
34526 Specifying multiple
34527 default actions is an error; specifying no actions is also an error.
34528 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
34530 Currently supported actions are:
34536 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
34540 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
34543 @item r @var{start},@var{end}
34544 Step once, and then keep stepping as long as the thread stops at
34545 addresses between @var{start} (inclusive) and @var{end} (exclusive).
34546 The remote stub reports a stop reply when either the thread goes out
34547 of the range or is stopped due to an unrelated reason, such as hitting
34548 a breakpoint. @xref{range stepping}.
34550 If the range is empty (@var{start} == @var{end}), then the action
34551 becomes equivalent to the @samp{s} action. In other words,
34552 single-step once, and report the stop (even if the stepped instruction
34553 jumps to @var{start}).
34555 (A stop reply may be sent at any point even if the PC is still within
34556 the stepping range; for example, it is valid to implement this packet
34557 in a degenerate way as a single instruction step operation.)
34561 The optional argument @var{addr} normally associated with the
34562 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
34563 not supported in @samp{vCont}.
34565 The @samp{t} action is only relevant in non-stop mode
34566 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
34567 A stop reply should be generated for any affected thread not already stopped.
34568 When a thread is stopped by means of a @samp{t} action,
34569 the corresponding stop reply should indicate that the thread has stopped with
34570 signal @samp{0}, regardless of whether the target uses some other signal
34571 as an implementation detail.
34573 The stub must support @samp{vCont} if it reports support for
34574 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
34575 this case @samp{vCont} actions can be specified to apply to all threads
34576 in a process by using the @samp{p@var{pid}.-1} form of the
34580 @xref{Stop Reply Packets}, for the reply specifications.
34583 @cindex @samp{vCont?} packet
34584 Request a list of actions supported by the @samp{vCont} packet.
34588 @item vCont@r{[};@var{action}@dots{}@r{]}
34589 The @samp{vCont} packet is supported. Each @var{action} is a supported
34590 command in the @samp{vCont} packet.
34592 The @samp{vCont} packet is not supported.
34595 @item vFile:@var{operation}:@var{parameter}@dots{}
34596 @cindex @samp{vFile} packet
34597 Perform a file operation on the target system. For details,
34598 see @ref{Host I/O Packets}.
34600 @item vFlashErase:@var{addr},@var{length}
34601 @cindex @samp{vFlashErase} packet
34602 Direct the stub to erase @var{length} bytes of flash starting at
34603 @var{addr}. The region may enclose any number of flash blocks, but
34604 its start and end must fall on block boundaries, as indicated by the
34605 flash block size appearing in the memory map (@pxref{Memory Map
34606 Format}). @value{GDBN} groups flash memory programming operations
34607 together, and sends a @samp{vFlashDone} request after each group; the
34608 stub is allowed to delay erase operation until the @samp{vFlashDone}
34609 packet is received.
34619 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
34620 @cindex @samp{vFlashWrite} packet
34621 Direct the stub to write data to flash address @var{addr}. The data
34622 is passed in binary form using the same encoding as for the @samp{X}
34623 packet (@pxref{Binary Data}). The memory ranges specified by
34624 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
34625 not overlap, and must appear in order of increasing addresses
34626 (although @samp{vFlashErase} packets for higher addresses may already
34627 have been received; the ordering is guaranteed only between
34628 @samp{vFlashWrite} packets). If a packet writes to an address that was
34629 neither erased by a preceding @samp{vFlashErase} packet nor by some other
34630 target-specific method, the results are unpredictable.
34638 for vFlashWrite addressing non-flash memory
34644 @cindex @samp{vFlashDone} packet
34645 Indicate to the stub that flash programming operation is finished.
34646 The stub is permitted to delay or batch the effects of a group of
34647 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
34648 @samp{vFlashDone} packet is received. The contents of the affected
34649 regions of flash memory are unpredictable until the @samp{vFlashDone}
34650 request is completed.
34652 @item vKill;@var{pid}
34653 @cindex @samp{vKill} packet
34654 @anchor{vKill packet}
34655 Kill the process with the specified process ID @var{pid}, which is a
34656 hexadecimal integer identifying the process. This packet is used in
34657 preference to @samp{k} when multiprocess protocol extensions are
34658 supported; see @ref{multiprocess extensions}.
34668 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
34669 @cindex @samp{vRun} packet
34670 Run the program @var{filename}, passing it each @var{argument} on its
34671 command line. The file and arguments are hex-encoded strings. If
34672 @var{filename} is an empty string, the stub may use a default program
34673 (e.g.@: the last program run). The program is created in the stopped
34676 @c FIXME: What about non-stop mode?
34678 This packet is only available in extended mode (@pxref{extended mode}).
34684 @item @r{Any stop packet}
34685 for success (@pxref{Stop Reply Packets})
34689 @cindex @samp{vStopped} packet
34690 @xref{Notification Packets}.
34692 @item X @var{addr},@var{length}:@var{XX@dots{}}
34694 @cindex @samp{X} packet
34695 Write data to memory, where the data is transmitted in binary.
34696 Memory is specified by its address @var{addr} and number of bytes @var{length};
34697 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
34707 @item z @var{type},@var{addr},@var{kind}
34708 @itemx Z @var{type},@var{addr},@var{kind}
34709 @anchor{insert breakpoint or watchpoint packet}
34710 @cindex @samp{z} packet
34711 @cindex @samp{Z} packets
34712 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
34713 watchpoint starting at address @var{address} of kind @var{kind}.
34715 Each breakpoint and watchpoint packet @var{type} is documented
34718 @emph{Implementation notes: A remote target shall return an empty string
34719 for an unrecognized breakpoint or watchpoint packet @var{type}. A
34720 remote target shall support either both or neither of a given
34721 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
34722 avoid potential problems with duplicate packets, the operations should
34723 be implemented in an idempotent way.}
34725 @item z0,@var{addr},@var{kind}
34726 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
34727 @cindex @samp{z0} packet
34728 @cindex @samp{Z0} packet
34729 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
34730 @var{addr} of type @var{kind}.
34732 A memory breakpoint is implemented by replacing the instruction at
34733 @var{addr} with a software breakpoint or trap instruction. The
34734 @var{kind} is target-specific and typically indicates the size of
34735 the breakpoint in bytes that should be inserted. E.g., the @sc{arm}
34736 and @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
34737 architectures have additional meanings for @var{kind};
34738 @var{cond_list} is an optional list of conditional expressions in bytecode
34739 form that should be evaluated on the target's side. These are the
34740 conditions that should be taken into consideration when deciding if
34741 the breakpoint trigger should be reported back to @var{GDBN}.
34743 The @var{cond_list} parameter is comprised of a series of expressions,
34744 concatenated without separators. Each expression has the following form:
34748 @item X @var{len},@var{expr}
34749 @var{len} is the length of the bytecode expression and @var{expr} is the
34750 actual conditional expression in bytecode form.
34754 The optional @var{cmd_list} parameter introduces commands that may be
34755 run on the target, rather than being reported back to @value{GDBN}.
34756 The parameter starts with a numeric flag @var{persist}; if the flag is
34757 nonzero, then the breakpoint may remain active and the commands
34758 continue to be run even when @value{GDBN} disconnects from the target.
34759 Following this flag is a series of expressions concatenated with no
34760 separators. Each expression has the following form:
34764 @item X @var{len},@var{expr}
34765 @var{len} is the length of the bytecode expression and @var{expr} is the
34766 actual conditional expression in bytecode form.
34770 see @ref{Architecture-Specific Protocol Details}.
34772 @emph{Implementation note: It is possible for a target to copy or move
34773 code that contains memory breakpoints (e.g., when implementing
34774 overlays). The behavior of this packet, in the presence of such a
34775 target, is not defined.}
34787 @item z1,@var{addr},@var{kind}
34788 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}
34789 @cindex @samp{z1} packet
34790 @cindex @samp{Z1} packet
34791 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
34792 address @var{addr}.
34794 A hardware breakpoint is implemented using a mechanism that is not
34795 dependant on being able to modify the target's memory. The @var{kind}
34796 and @var{cond_list} have the same meaning as in @samp{Z0} packets.
34798 @emph{Implementation note: A hardware breakpoint is not affected by code
34811 @item z2,@var{addr},@var{kind}
34812 @itemx Z2,@var{addr},@var{kind}
34813 @cindex @samp{z2} packet
34814 @cindex @samp{Z2} packet
34815 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
34816 The number of bytes to watch is specified by @var{kind}.
34828 @item z3,@var{addr},@var{kind}
34829 @itemx Z3,@var{addr},@var{kind}
34830 @cindex @samp{z3} packet
34831 @cindex @samp{Z3} packet
34832 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
34833 The number of bytes to watch is specified by @var{kind}.
34845 @item z4,@var{addr},@var{kind}
34846 @itemx Z4,@var{addr},@var{kind}
34847 @cindex @samp{z4} packet
34848 @cindex @samp{Z4} packet
34849 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
34850 The number of bytes to watch is specified by @var{kind}.
34864 @node Stop Reply Packets
34865 @section Stop Reply Packets
34866 @cindex stop reply packets
34868 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
34869 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
34870 receive any of the below as a reply. Except for @samp{?}
34871 and @samp{vStopped}, that reply is only returned
34872 when the target halts. In the below the exact meaning of @dfn{signal
34873 number} is defined by the header @file{include/gdb/signals.h} in the
34874 @value{GDBN} source code.
34876 As in the description of request packets, we include spaces in the
34877 reply templates for clarity; these are not part of the reply packet's
34878 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
34884 The program received signal number @var{AA} (a two-digit hexadecimal
34885 number). This is equivalent to a @samp{T} response with no
34886 @var{n}:@var{r} pairs.
34888 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
34889 @cindex @samp{T} packet reply
34890 The program received signal number @var{AA} (a two-digit hexadecimal
34891 number). This is equivalent to an @samp{S} response, except that the
34892 @samp{@var{n}:@var{r}} pairs can carry values of important registers
34893 and other information directly in the stop reply packet, reducing
34894 round-trip latency. Single-step and breakpoint traps are reported
34895 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
34899 If @var{n} is a hexadecimal number, it is a register number, and the
34900 corresponding @var{r} gives that register's value. The data @var{r} is a
34901 series of bytes in target byte order, with each byte given by a
34902 two-digit hex number.
34905 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
34906 the stopped thread, as specified in @ref{thread-id syntax}.
34909 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
34910 the core on which the stop event was detected.
34913 If @var{n} is a recognized @dfn{stop reason}, it describes a more
34914 specific event that stopped the target. The currently defined stop
34915 reasons are listed below. The @var{aa} should be @samp{05}, the trap
34916 signal. At most one stop reason should be present.
34919 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
34920 and go on to the next; this allows us to extend the protocol in the
34924 The currently defined stop reasons are:
34930 The packet indicates a watchpoint hit, and @var{r} is the data address, in
34933 @cindex shared library events, remote reply
34935 The packet indicates that the loaded libraries have changed.
34936 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
34937 list of loaded libraries. The @var{r} part is ignored.
34939 @cindex replay log events, remote reply
34941 The packet indicates that the target cannot continue replaying
34942 logged execution events, because it has reached the end (or the
34943 beginning when executing backward) of the log. The value of @var{r}
34944 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
34945 for more information.
34949 @itemx W @var{AA} ; process:@var{pid}
34950 The process exited, and @var{AA} is the exit status. This is only
34951 applicable to certain targets.
34953 The second form of the response, including the process ID of the exited
34954 process, can be used only when @value{GDBN} has reported support for
34955 multiprocess protocol extensions; see @ref{multiprocess extensions}.
34956 The @var{pid} is formatted as a big-endian hex string.
34959 @itemx X @var{AA} ; process:@var{pid}
34960 The process terminated with signal @var{AA}.
34962 The second form of the response, including the process ID of the
34963 terminated process, can be used only when @value{GDBN} has reported
34964 support for multiprocess protocol extensions; see @ref{multiprocess
34965 extensions}. The @var{pid} is formatted as a big-endian hex string.
34967 @item O @var{XX}@dots{}
34968 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
34969 written as the program's console output. This can happen at any time
34970 while the program is running and the debugger should continue to wait
34971 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
34973 @item F @var{call-id},@var{parameter}@dots{}
34974 @var{call-id} is the identifier which says which host system call should
34975 be called. This is just the name of the function. Translation into the
34976 correct system call is only applicable as it's defined in @value{GDBN}.
34977 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
34980 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
34981 this very system call.
34983 The target replies with this packet when it expects @value{GDBN} to
34984 call a host system call on behalf of the target. @value{GDBN} replies
34985 with an appropriate @samp{F} packet and keeps up waiting for the next
34986 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
34987 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
34988 Protocol Extension}, for more details.
34992 @node General Query Packets
34993 @section General Query Packets
34994 @cindex remote query requests
34996 Packets starting with @samp{q} are @dfn{general query packets};
34997 packets starting with @samp{Q} are @dfn{general set packets}. General
34998 query and set packets are a semi-unified form for retrieving and
34999 sending information to and from the stub.
35001 The initial letter of a query or set packet is followed by a name
35002 indicating what sort of thing the packet applies to. For example,
35003 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
35004 definitions with the stub. These packet names follow some
35009 The name must not contain commas, colons or semicolons.
35011 Most @value{GDBN} query and set packets have a leading upper case
35014 The names of custom vendor packets should use a company prefix, in
35015 lower case, followed by a period. For example, packets designed at
35016 the Acme Corporation might begin with @samp{qacme.foo} (for querying
35017 foos) or @samp{Qacme.bar} (for setting bars).
35020 The name of a query or set packet should be separated from any
35021 parameters by a @samp{:}; the parameters themselves should be
35022 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
35023 full packet name, and check for a separator or the end of the packet,
35024 in case two packet names share a common prefix. New packets should not begin
35025 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
35026 packets predate these conventions, and have arguments without any terminator
35027 for the packet name; we suspect they are in widespread use in places that
35028 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
35029 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
35032 Like the descriptions of the other packets, each description here
35033 has a template showing the packet's overall syntax, followed by an
35034 explanation of the packet's meaning. We include spaces in some of the
35035 templates for clarity; these are not part of the packet's syntax. No
35036 @value{GDBN} packet uses spaces to separate its components.
35038 Here are the currently defined query and set packets:
35044 Turn on or off the agent as a helper to perform some debugging operations
35045 delegated from @value{GDBN} (@pxref{Control Agent}).
35047 @item QAllow:@var{op}:@var{val}@dots{}
35048 @cindex @samp{QAllow} packet
35049 Specify which operations @value{GDBN} expects to request of the
35050 target, as a semicolon-separated list of operation name and value
35051 pairs. Possible values for @var{op} include @samp{WriteReg},
35052 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
35053 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
35054 indicating that @value{GDBN} will not request the operation, or 1,
35055 indicating that it may. (The target can then use this to set up its
35056 own internals optimally, for instance if the debugger never expects to
35057 insert breakpoints, it may not need to install its own trap handler.)
35060 @cindex current thread, remote request
35061 @cindex @samp{qC} packet
35062 Return the current thread ID.
35066 @item QC @var{thread-id}
35067 Where @var{thread-id} is a thread ID as documented in
35068 @ref{thread-id syntax}.
35069 @item @r{(anything else)}
35070 Any other reply implies the old thread ID.
35073 @item qCRC:@var{addr},@var{length}
35074 @cindex CRC of memory block, remote request
35075 @cindex @samp{qCRC} packet
35076 @anchor{qCRC packet}
35077 Compute the CRC checksum of a block of memory using CRC-32 defined in
35078 IEEE 802.3. The CRC is computed byte at a time, taking the most
35079 significant bit of each byte first. The initial pattern code
35080 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
35082 @emph{Note:} This is the same CRC used in validating separate debug
35083 files (@pxref{Separate Debug Files, , Debugging Information in Separate
35084 Files}). However the algorithm is slightly different. When validating
35085 separate debug files, the CRC is computed taking the @emph{least}
35086 significant bit of each byte first, and the final result is inverted to
35087 detect trailing zeros.
35092 An error (such as memory fault)
35093 @item C @var{crc32}
35094 The specified memory region's checksum is @var{crc32}.
35097 @item QDisableRandomization:@var{value}
35098 @cindex disable address space randomization, remote request
35099 @cindex @samp{QDisableRandomization} packet
35100 Some target operating systems will randomize the virtual address space
35101 of the inferior process as a security feature, but provide a feature
35102 to disable such randomization, e.g.@: to allow for a more deterministic
35103 debugging experience. On such systems, this packet with a @var{value}
35104 of 1 directs the target to disable address space randomization for
35105 processes subsequently started via @samp{vRun} packets, while a packet
35106 with a @var{value} of 0 tells the target to enable address space
35109 This packet is only available in extended mode (@pxref{extended mode}).
35114 The request succeeded.
35117 An error occurred. The error number @var{nn} is given as hex digits.
35120 An empty reply indicates that @samp{QDisableRandomization} is not supported
35124 This packet is not probed by default; the remote stub must request it,
35125 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
35126 This should only be done on targets that actually support disabling
35127 address space randomization.
35130 @itemx qsThreadInfo
35131 @cindex list active threads, remote request
35132 @cindex @samp{qfThreadInfo} packet
35133 @cindex @samp{qsThreadInfo} packet
35134 Obtain a list of all active thread IDs from the target (OS). Since there
35135 may be too many active threads to fit into one reply packet, this query
35136 works iteratively: it may require more than one query/reply sequence to
35137 obtain the entire list of threads. The first query of the sequence will
35138 be the @samp{qfThreadInfo} query; subsequent queries in the
35139 sequence will be the @samp{qsThreadInfo} query.
35141 NOTE: This packet replaces the @samp{qL} query (see below).
35145 @item m @var{thread-id}
35147 @item m @var{thread-id},@var{thread-id}@dots{}
35148 a comma-separated list of thread IDs
35150 (lower case letter @samp{L}) denotes end of list.
35153 In response to each query, the target will reply with a list of one or
35154 more thread IDs, separated by commas.
35155 @value{GDBN} will respond to each reply with a request for more thread
35156 ids (using the @samp{qs} form of the query), until the target responds
35157 with @samp{l} (lower-case ell, for @dfn{last}).
35158 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
35161 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
35162 initial connection with the remote target, and the very first thread ID
35163 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
35164 message. Therefore, the stub should ensure that the first thread ID in
35165 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
35167 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
35168 @cindex get thread-local storage address, remote request
35169 @cindex @samp{qGetTLSAddr} packet
35170 Fetch the address associated with thread local storage specified
35171 by @var{thread-id}, @var{offset}, and @var{lm}.
35173 @var{thread-id} is the thread ID associated with the
35174 thread for which to fetch the TLS address. @xref{thread-id syntax}.
35176 @var{offset} is the (big endian, hex encoded) offset associated with the
35177 thread local variable. (This offset is obtained from the debug
35178 information associated with the variable.)
35180 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
35181 load module associated with the thread local storage. For example,
35182 a @sc{gnu}/Linux system will pass the link map address of the shared
35183 object associated with the thread local storage under consideration.
35184 Other operating environments may choose to represent the load module
35185 differently, so the precise meaning of this parameter will vary.
35189 @item @var{XX}@dots{}
35190 Hex encoded (big endian) bytes representing the address of the thread
35191 local storage requested.
35194 An error occurred. The error number @var{nn} is given as hex digits.
35197 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
35200 @item qGetTIBAddr:@var{thread-id}
35201 @cindex get thread information block address
35202 @cindex @samp{qGetTIBAddr} packet
35203 Fetch address of the Windows OS specific Thread Information Block.
35205 @var{thread-id} is the thread ID associated with the thread.
35209 @item @var{XX}@dots{}
35210 Hex encoded (big endian) bytes representing the linear address of the
35211 thread information block.
35214 An error occured. This means that either the thread was not found, or the
35215 address could not be retrieved.
35218 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
35221 @item qL @var{startflag} @var{threadcount} @var{nextthread}
35222 Obtain thread information from RTOS. Where: @var{startflag} (one hex
35223 digit) is one to indicate the first query and zero to indicate a
35224 subsequent query; @var{threadcount} (two hex digits) is the maximum
35225 number of threads the response packet can contain; and @var{nextthread}
35226 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
35227 returned in the response as @var{argthread}.
35229 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
35233 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
35234 Where: @var{count} (two hex digits) is the number of threads being
35235 returned; @var{done} (one hex digit) is zero to indicate more threads
35236 and one indicates no further threads; @var{argthreadid} (eight hex
35237 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
35238 is a sequence of thread IDs, @var{threadid} (eight hex
35239 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
35243 @cindex section offsets, remote request
35244 @cindex @samp{qOffsets} packet
35245 Get section offsets that the target used when relocating the downloaded
35250 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
35251 Relocate the @code{Text} section by @var{xxx} from its original address.
35252 Relocate the @code{Data} section by @var{yyy} from its original address.
35253 If the object file format provides segment information (e.g.@: @sc{elf}
35254 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
35255 segments by the supplied offsets.
35257 @emph{Note: while a @code{Bss} offset may be included in the response,
35258 @value{GDBN} ignores this and instead applies the @code{Data} offset
35259 to the @code{Bss} section.}
35261 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
35262 Relocate the first segment of the object file, which conventionally
35263 contains program code, to a starting address of @var{xxx}. If
35264 @samp{DataSeg} is specified, relocate the second segment, which
35265 conventionally contains modifiable data, to a starting address of
35266 @var{yyy}. @value{GDBN} will report an error if the object file
35267 does not contain segment information, or does not contain at least
35268 as many segments as mentioned in the reply. Extra segments are
35269 kept at fixed offsets relative to the last relocated segment.
35272 @item qP @var{mode} @var{thread-id}
35273 @cindex thread information, remote request
35274 @cindex @samp{qP} packet
35275 Returns information on @var{thread-id}. Where: @var{mode} is a hex
35276 encoded 32 bit mode; @var{thread-id} is a thread ID
35277 (@pxref{thread-id syntax}).
35279 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
35282 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
35286 @cindex non-stop mode, remote request
35287 @cindex @samp{QNonStop} packet
35289 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
35290 @xref{Remote Non-Stop}, for more information.
35295 The request succeeded.
35298 An error occurred. The error number @var{nn} is given as hex digits.
35301 An empty reply indicates that @samp{QNonStop} is not supported by
35305 This packet is not probed by default; the remote stub must request it,
35306 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
35307 Use of this packet is controlled by the @code{set non-stop} command;
35308 @pxref{Non-Stop Mode}.
35310 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
35311 @cindex pass signals to inferior, remote request
35312 @cindex @samp{QPassSignals} packet
35313 @anchor{QPassSignals}
35314 Each listed @var{signal} should be passed directly to the inferior process.
35315 Signals are numbered identically to continue packets and stop replies
35316 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
35317 strictly greater than the previous item. These signals do not need to stop
35318 the inferior, or be reported to @value{GDBN}. All other signals should be
35319 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
35320 combine; any earlier @samp{QPassSignals} list is completely replaced by the
35321 new list. This packet improves performance when using @samp{handle
35322 @var{signal} nostop noprint pass}.
35327 The request succeeded.
35330 An error occurred. The error number @var{nn} is given as hex digits.
35333 An empty reply indicates that @samp{QPassSignals} is not supported by
35337 Use of this packet is controlled by the @code{set remote pass-signals}
35338 command (@pxref{Remote Configuration, set remote pass-signals}).
35339 This packet is not probed by default; the remote stub must request it,
35340 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
35342 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
35343 @cindex signals the inferior may see, remote request
35344 @cindex @samp{QProgramSignals} packet
35345 @anchor{QProgramSignals}
35346 Each listed @var{signal} may be delivered to the inferior process.
35347 Others should be silently discarded.
35349 In some cases, the remote stub may need to decide whether to deliver a
35350 signal to the program or not without @value{GDBN} involvement. One
35351 example of that is while detaching --- the program's threads may have
35352 stopped for signals that haven't yet had a chance of being reported to
35353 @value{GDBN}, and so the remote stub can use the signal list specified
35354 by this packet to know whether to deliver or ignore those pending
35357 This does not influence whether to deliver a signal as requested by a
35358 resumption packet (@pxref{vCont packet}).
35360 Signals are numbered identically to continue packets and stop replies
35361 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
35362 strictly greater than the previous item. Multiple
35363 @samp{QProgramSignals} packets do not combine; any earlier
35364 @samp{QProgramSignals} list is completely replaced by the new list.
35369 The request succeeded.
35372 An error occurred. The error number @var{nn} is given as hex digits.
35375 An empty reply indicates that @samp{QProgramSignals} is not supported
35379 Use of this packet is controlled by the @code{set remote program-signals}
35380 command (@pxref{Remote Configuration, set remote program-signals}).
35381 This packet is not probed by default; the remote stub must request it,
35382 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
35384 @item qRcmd,@var{command}
35385 @cindex execute remote command, remote request
35386 @cindex @samp{qRcmd} packet
35387 @var{command} (hex encoded) is passed to the local interpreter for
35388 execution. Invalid commands should be reported using the output
35389 string. Before the final result packet, the target may also respond
35390 with a number of intermediate @samp{O@var{output}} console output
35391 packets. @emph{Implementors should note that providing access to a
35392 stubs's interpreter may have security implications}.
35397 A command response with no output.
35399 A command response with the hex encoded output string @var{OUTPUT}.
35401 Indicate a badly formed request.
35403 An empty reply indicates that @samp{qRcmd} is not recognized.
35406 (Note that the @code{qRcmd} packet's name is separated from the
35407 command by a @samp{,}, not a @samp{:}, contrary to the naming
35408 conventions above. Please don't use this packet as a model for new
35411 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
35412 @cindex searching memory, in remote debugging
35414 @cindex @samp{qSearch:memory} packet
35416 @cindex @samp{qSearch memory} packet
35417 @anchor{qSearch memory}
35418 Search @var{length} bytes at @var{address} for @var{search-pattern}.
35419 Both @var{address} and @var{length} are encoded in hex;
35420 @var{search-pattern} is a sequence of bytes, also hex encoded.
35425 The pattern was not found.
35427 The pattern was found at @var{address}.
35429 A badly formed request or an error was encountered while searching memory.
35431 An empty reply indicates that @samp{qSearch:memory} is not recognized.
35434 @item QStartNoAckMode
35435 @cindex @samp{QStartNoAckMode} packet
35436 @anchor{QStartNoAckMode}
35437 Request that the remote stub disable the normal @samp{+}/@samp{-}
35438 protocol acknowledgments (@pxref{Packet Acknowledgment}).
35443 The stub has switched to no-acknowledgment mode.
35444 @value{GDBN} acknowledges this reponse,
35445 but neither the stub nor @value{GDBN} shall send or expect further
35446 @samp{+}/@samp{-} acknowledgments in the current connection.
35448 An empty reply indicates that the stub does not support no-acknowledgment mode.
35451 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
35452 @cindex supported packets, remote query
35453 @cindex features of the remote protocol
35454 @cindex @samp{qSupported} packet
35455 @anchor{qSupported}
35456 Tell the remote stub about features supported by @value{GDBN}, and
35457 query the stub for features it supports. This packet allows
35458 @value{GDBN} and the remote stub to take advantage of each others'
35459 features. @samp{qSupported} also consolidates multiple feature probes
35460 at startup, to improve @value{GDBN} performance---a single larger
35461 packet performs better than multiple smaller probe packets on
35462 high-latency links. Some features may enable behavior which must not
35463 be on by default, e.g.@: because it would confuse older clients or
35464 stubs. Other features may describe packets which could be
35465 automatically probed for, but are not. These features must be
35466 reported before @value{GDBN} will use them. This ``default
35467 unsupported'' behavior is not appropriate for all packets, but it
35468 helps to keep the initial connection time under control with new
35469 versions of @value{GDBN} which support increasing numbers of packets.
35473 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
35474 The stub supports or does not support each returned @var{stubfeature},
35475 depending on the form of each @var{stubfeature} (see below for the
35478 An empty reply indicates that @samp{qSupported} is not recognized,
35479 or that no features needed to be reported to @value{GDBN}.
35482 The allowed forms for each feature (either a @var{gdbfeature} in the
35483 @samp{qSupported} packet, or a @var{stubfeature} in the response)
35487 @item @var{name}=@var{value}
35488 The remote protocol feature @var{name} is supported, and associated
35489 with the specified @var{value}. The format of @var{value} depends
35490 on the feature, but it must not include a semicolon.
35492 The remote protocol feature @var{name} is supported, and does not
35493 need an associated value.
35495 The remote protocol feature @var{name} is not supported.
35497 The remote protocol feature @var{name} may be supported, and
35498 @value{GDBN} should auto-detect support in some other way when it is
35499 needed. This form will not be used for @var{gdbfeature} notifications,
35500 but may be used for @var{stubfeature} responses.
35503 Whenever the stub receives a @samp{qSupported} request, the
35504 supplied set of @value{GDBN} features should override any previous
35505 request. This allows @value{GDBN} to put the stub in a known
35506 state, even if the stub had previously been communicating with
35507 a different version of @value{GDBN}.
35509 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
35514 This feature indicates whether @value{GDBN} supports multiprocess
35515 extensions to the remote protocol. @value{GDBN} does not use such
35516 extensions unless the stub also reports that it supports them by
35517 including @samp{multiprocess+} in its @samp{qSupported} reply.
35518 @xref{multiprocess extensions}, for details.
35521 This feature indicates that @value{GDBN} supports the XML target
35522 description. If the stub sees @samp{xmlRegisters=} with target
35523 specific strings separated by a comma, it will report register
35527 This feature indicates whether @value{GDBN} supports the
35528 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
35529 instruction reply packet}).
35532 Stubs should ignore any unknown values for
35533 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
35534 packet supports receiving packets of unlimited length (earlier
35535 versions of @value{GDBN} may reject overly long responses). Additional values
35536 for @var{gdbfeature} may be defined in the future to let the stub take
35537 advantage of new features in @value{GDBN}, e.g.@: incompatible
35538 improvements in the remote protocol---the @samp{multiprocess} feature is
35539 an example of such a feature. The stub's reply should be independent
35540 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
35541 describes all the features it supports, and then the stub replies with
35542 all the features it supports.
35544 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
35545 responses, as long as each response uses one of the standard forms.
35547 Some features are flags. A stub which supports a flag feature
35548 should respond with a @samp{+} form response. Other features
35549 require values, and the stub should respond with an @samp{=}
35552 Each feature has a default value, which @value{GDBN} will use if
35553 @samp{qSupported} is not available or if the feature is not mentioned
35554 in the @samp{qSupported} response. The default values are fixed; a
35555 stub is free to omit any feature responses that match the defaults.
35557 Not all features can be probed, but for those which can, the probing
35558 mechanism is useful: in some cases, a stub's internal
35559 architecture may not allow the protocol layer to know some information
35560 about the underlying target in advance. This is especially common in
35561 stubs which may be configured for multiple targets.
35563 These are the currently defined stub features and their properties:
35565 @multitable @columnfractions 0.35 0.2 0.12 0.2
35566 @c NOTE: The first row should be @headitem, but we do not yet require
35567 @c a new enough version of Texinfo (4.7) to use @headitem.
35569 @tab Value Required
35573 @item @samp{PacketSize}
35578 @item @samp{qXfer:auxv:read}
35583 @item @samp{qXfer:btrace:read}
35588 @item @samp{qXfer:features:read}
35593 @item @samp{qXfer:libraries:read}
35598 @item @samp{qXfer:libraries-svr4:read}
35603 @item @samp{augmented-libraries-svr4-read}
35608 @item @samp{qXfer:memory-map:read}
35613 @item @samp{qXfer:sdata:read}
35618 @item @samp{qXfer:spu:read}
35623 @item @samp{qXfer:spu:write}
35628 @item @samp{qXfer:siginfo:read}
35633 @item @samp{qXfer:siginfo:write}
35638 @item @samp{qXfer:threads:read}
35643 @item @samp{qXfer:traceframe-info:read}
35648 @item @samp{qXfer:uib:read}
35653 @item @samp{qXfer:fdpic:read}
35658 @item @samp{Qbtrace:off}
35663 @item @samp{Qbtrace:bts}
35668 @item @samp{QNonStop}
35673 @item @samp{QPassSignals}
35678 @item @samp{QStartNoAckMode}
35683 @item @samp{multiprocess}
35688 @item @samp{ConditionalBreakpoints}
35693 @item @samp{ConditionalTracepoints}
35698 @item @samp{ReverseContinue}
35703 @item @samp{ReverseStep}
35708 @item @samp{TracepointSource}
35713 @item @samp{QAgent}
35718 @item @samp{QAllow}
35723 @item @samp{QDisableRandomization}
35728 @item @samp{EnableDisableTracepoints}
35733 @item @samp{QTBuffer:size}
35738 @item @samp{tracenz}
35743 @item @samp{BreakpointCommands}
35750 These are the currently defined stub features, in more detail:
35753 @cindex packet size, remote protocol
35754 @item PacketSize=@var{bytes}
35755 The remote stub can accept packets up to at least @var{bytes} in
35756 length. @value{GDBN} will send packets up to this size for bulk
35757 transfers, and will never send larger packets. This is a limit on the
35758 data characters in the packet, including the frame and checksum.
35759 There is no trailing NUL byte in a remote protocol packet; if the stub
35760 stores packets in a NUL-terminated format, it should allow an extra
35761 byte in its buffer for the NUL. If this stub feature is not supported,
35762 @value{GDBN} guesses based on the size of the @samp{g} packet response.
35764 @item qXfer:auxv:read
35765 The remote stub understands the @samp{qXfer:auxv:read} packet
35766 (@pxref{qXfer auxiliary vector read}).
35768 @item qXfer:btrace:read
35769 The remote stub understands the @samp{qXfer:btrace:read}
35770 packet (@pxref{qXfer btrace read}).
35772 @item qXfer:features:read
35773 The remote stub understands the @samp{qXfer:features:read} packet
35774 (@pxref{qXfer target description read}).
35776 @item qXfer:libraries:read
35777 The remote stub understands the @samp{qXfer:libraries:read} packet
35778 (@pxref{qXfer library list read}).
35780 @item qXfer:libraries-svr4:read
35781 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
35782 (@pxref{qXfer svr4 library list read}).
35784 @item augmented-libraries-svr4-read
35785 The remote stub understands the augmented form of the
35786 @samp{qXfer:libraries-svr4:read} packet
35787 (@pxref{qXfer svr4 library list read}).
35789 @item qXfer:memory-map:read
35790 The remote stub understands the @samp{qXfer:memory-map:read} packet
35791 (@pxref{qXfer memory map read}).
35793 @item qXfer:sdata:read
35794 The remote stub understands the @samp{qXfer:sdata:read} packet
35795 (@pxref{qXfer sdata read}).
35797 @item qXfer:spu:read
35798 The remote stub understands the @samp{qXfer:spu:read} packet
35799 (@pxref{qXfer spu read}).
35801 @item qXfer:spu:write
35802 The remote stub understands the @samp{qXfer:spu:write} packet
35803 (@pxref{qXfer spu write}).
35805 @item qXfer:siginfo:read
35806 The remote stub understands the @samp{qXfer:siginfo:read} packet
35807 (@pxref{qXfer siginfo read}).
35809 @item qXfer:siginfo:write
35810 The remote stub understands the @samp{qXfer:siginfo:write} packet
35811 (@pxref{qXfer siginfo write}).
35813 @item qXfer:threads:read
35814 The remote stub understands the @samp{qXfer:threads:read} packet
35815 (@pxref{qXfer threads read}).
35817 @item qXfer:traceframe-info:read
35818 The remote stub understands the @samp{qXfer:traceframe-info:read}
35819 packet (@pxref{qXfer traceframe info read}).
35821 @item qXfer:uib:read
35822 The remote stub understands the @samp{qXfer:uib:read}
35823 packet (@pxref{qXfer unwind info block}).
35825 @item qXfer:fdpic:read
35826 The remote stub understands the @samp{qXfer:fdpic:read}
35827 packet (@pxref{qXfer fdpic loadmap read}).
35830 The remote stub understands the @samp{QNonStop} packet
35831 (@pxref{QNonStop}).
35834 The remote stub understands the @samp{QPassSignals} packet
35835 (@pxref{QPassSignals}).
35837 @item QStartNoAckMode
35838 The remote stub understands the @samp{QStartNoAckMode} packet and
35839 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
35842 @anchor{multiprocess extensions}
35843 @cindex multiprocess extensions, in remote protocol
35844 The remote stub understands the multiprocess extensions to the remote
35845 protocol syntax. The multiprocess extensions affect the syntax of
35846 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
35847 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
35848 replies. Note that reporting this feature indicates support for the
35849 syntactic extensions only, not that the stub necessarily supports
35850 debugging of more than one process at a time. The stub must not use
35851 multiprocess extensions in packet replies unless @value{GDBN} has also
35852 indicated it supports them in its @samp{qSupported} request.
35854 @item qXfer:osdata:read
35855 The remote stub understands the @samp{qXfer:osdata:read} packet
35856 ((@pxref{qXfer osdata read}).
35858 @item ConditionalBreakpoints
35859 The target accepts and implements evaluation of conditional expressions
35860 defined for breakpoints. The target will only report breakpoint triggers
35861 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
35863 @item ConditionalTracepoints
35864 The remote stub accepts and implements conditional expressions defined
35865 for tracepoints (@pxref{Tracepoint Conditions}).
35867 @item ReverseContinue
35868 The remote stub accepts and implements the reverse continue packet
35872 The remote stub accepts and implements the reverse step packet
35875 @item TracepointSource
35876 The remote stub understands the @samp{QTDPsrc} packet that supplies
35877 the source form of tracepoint definitions.
35880 The remote stub understands the @samp{QAgent} packet.
35883 The remote stub understands the @samp{QAllow} packet.
35885 @item QDisableRandomization
35886 The remote stub understands the @samp{QDisableRandomization} packet.
35888 @item StaticTracepoint
35889 @cindex static tracepoints, in remote protocol
35890 The remote stub supports static tracepoints.
35892 @item InstallInTrace
35893 @anchor{install tracepoint in tracing}
35894 The remote stub supports installing tracepoint in tracing.
35896 @item EnableDisableTracepoints
35897 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
35898 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
35899 to be enabled and disabled while a trace experiment is running.
35901 @item QTBuffer:size
35902 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
35903 packet that allows to change the size of the trace buffer.
35906 @cindex string tracing, in remote protocol
35907 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
35908 See @ref{Bytecode Descriptions} for details about the bytecode.
35910 @item BreakpointCommands
35911 @cindex breakpoint commands, in remote protocol
35912 The remote stub supports running a breakpoint's command list itself,
35913 rather than reporting the hit to @value{GDBN}.
35916 The remote stub understands the @samp{Qbtrace:off} packet.
35919 The remote stub understands the @samp{Qbtrace:bts} packet.
35924 @cindex symbol lookup, remote request
35925 @cindex @samp{qSymbol} packet
35926 Notify the target that @value{GDBN} is prepared to serve symbol lookup
35927 requests. Accept requests from the target for the values of symbols.
35932 The target does not need to look up any (more) symbols.
35933 @item qSymbol:@var{sym_name}
35934 The target requests the value of symbol @var{sym_name} (hex encoded).
35935 @value{GDBN} may provide the value by using the
35936 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
35940 @item qSymbol:@var{sym_value}:@var{sym_name}
35941 Set the value of @var{sym_name} to @var{sym_value}.
35943 @var{sym_name} (hex encoded) is the name of a symbol whose value the
35944 target has previously requested.
35946 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
35947 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
35953 The target does not need to look up any (more) symbols.
35954 @item qSymbol:@var{sym_name}
35955 The target requests the value of a new symbol @var{sym_name} (hex
35956 encoded). @value{GDBN} will continue to supply the values of symbols
35957 (if available), until the target ceases to request them.
35962 @itemx QTDisconnected
35969 @itemx qTMinFTPILen
35971 @xref{Tracepoint Packets}.
35973 @item qThreadExtraInfo,@var{thread-id}
35974 @cindex thread attributes info, remote request
35975 @cindex @samp{qThreadExtraInfo} packet
35976 Obtain from the target OS a printable string description of thread
35977 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
35978 for the forms of @var{thread-id}. This
35979 string may contain anything that the target OS thinks is interesting
35980 for @value{GDBN} to tell the user about the thread. The string is
35981 displayed in @value{GDBN}'s @code{info threads} display. Some
35982 examples of possible thread extra info strings are @samp{Runnable}, or
35983 @samp{Blocked on Mutex}.
35987 @item @var{XX}@dots{}
35988 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
35989 comprising the printable string containing the extra information about
35990 the thread's attributes.
35993 (Note that the @code{qThreadExtraInfo} packet's name is separated from
35994 the command by a @samp{,}, not a @samp{:}, contrary to the naming
35995 conventions above. Please don't use this packet as a model for new
36014 @xref{Tracepoint Packets}.
36016 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
36017 @cindex read special object, remote request
36018 @cindex @samp{qXfer} packet
36019 @anchor{qXfer read}
36020 Read uninterpreted bytes from the target's special data area
36021 identified by the keyword @var{object}. Request @var{length} bytes
36022 starting at @var{offset} bytes into the data. The content and
36023 encoding of @var{annex} is specific to @var{object}; it can supply
36024 additional details about what data to access.
36026 Here are the specific requests of this form defined so far. All
36027 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
36028 formats, listed below.
36031 @item qXfer:auxv:read::@var{offset},@var{length}
36032 @anchor{qXfer auxiliary vector read}
36033 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
36034 auxiliary vector}. Note @var{annex} must be empty.
36036 This packet is not probed by default; the remote stub must request it,
36037 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36039 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
36040 @anchor{qXfer btrace read}
36042 Return a description of the current branch trace.
36043 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
36044 packet may have one of the following values:
36048 Returns all available branch trace.
36051 Returns all available branch trace if the branch trace changed since
36052 the last read request.
36055 Returns the new branch trace since the last read request. Adds a new
36056 block to the end of the trace that begins at zero and ends at the source
36057 location of the first branch in the trace buffer. This extra block is
36058 used to stitch traces together.
36060 If the trace buffer overflowed, returns an error indicating the overflow.
36063 This packet is not probed by default; the remote stub must request it
36064 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36066 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
36067 @anchor{qXfer target description read}
36068 Access the @dfn{target description}. @xref{Target Descriptions}. The
36069 annex specifies which XML document to access. The main description is
36070 always loaded from the @samp{target.xml} annex.
36072 This packet is not probed by default; the remote stub must request it,
36073 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36075 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
36076 @anchor{qXfer library list read}
36077 Access the target's list of loaded libraries. @xref{Library List Format}.
36078 The annex part of the generic @samp{qXfer} packet must be empty
36079 (@pxref{qXfer read}).
36081 Targets which maintain a list of libraries in the program's memory do
36082 not need to implement this packet; it is designed for platforms where
36083 the operating system manages the list of loaded libraries.
36085 This packet is not probed by default; the remote stub must request it,
36086 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36088 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
36089 @anchor{qXfer svr4 library list read}
36090 Access the target's list of loaded libraries when the target is an SVR4
36091 platform. @xref{Library List Format for SVR4 Targets}. The annex part
36092 of the generic @samp{qXfer} packet must be empty unless the remote
36093 stub indicated it supports the augmented form of this packet
36094 by supplying an appropriate @samp{qSupported} response
36095 (@pxref{qXfer read}, @ref{qSupported}).
36097 This packet is optional for better performance on SVR4 targets.
36098 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
36100 This packet is not probed by default; the remote stub must request it,
36101 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36103 If the remote stub indicates it supports the augmented form of this
36104 packet then the annex part of the generic @samp{qXfer} packet may
36105 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
36106 arguments. The currently supported arguments are:
36109 @item start=@var{address}
36110 A hexadecimal number specifying the address of the @samp{struct
36111 link_map} to start reading the library list from. If unset or zero
36112 then the first @samp{struct link_map} in the library list will be
36113 chosen as the starting point.
36115 @item prev=@var{address}
36116 A hexadecimal number specifying the address of the @samp{struct
36117 link_map} immediately preceding the @samp{struct link_map}
36118 specified by the @samp{start} argument. If unset or zero then
36119 the remote stub will expect that no @samp{struct link_map}
36120 exists prior to the starting point.
36124 Arguments that are not understood by the remote stub will be silently
36127 @item qXfer:memory-map:read::@var{offset},@var{length}
36128 @anchor{qXfer memory map read}
36129 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
36130 annex part of the generic @samp{qXfer} packet must be empty
36131 (@pxref{qXfer read}).
36133 This packet is not probed by default; the remote stub must request it,
36134 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36136 @item qXfer:sdata:read::@var{offset},@var{length}
36137 @anchor{qXfer sdata read}
36139 Read contents of the extra collected static tracepoint marker
36140 information. The annex part of the generic @samp{qXfer} packet must
36141 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
36144 This packet is not probed by default; the remote stub must request it,
36145 by supplying an appropriate @samp{qSupported} response
36146 (@pxref{qSupported}).
36148 @item qXfer:siginfo:read::@var{offset},@var{length}
36149 @anchor{qXfer siginfo read}
36150 Read contents of the extra signal information on the target
36151 system. The annex part of the generic @samp{qXfer} packet must be
36152 empty (@pxref{qXfer read}).
36154 This packet is not probed by default; the remote stub must request it,
36155 by supplying an appropriate @samp{qSupported} response
36156 (@pxref{qSupported}).
36158 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
36159 @anchor{qXfer spu read}
36160 Read contents of an @code{spufs} file on the target system. The
36161 annex specifies which file to read; it must be of the form
36162 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
36163 in the target process, and @var{name} identifes the @code{spufs} file
36164 in that context to be accessed.
36166 This packet is not probed by default; the remote stub must request it,
36167 by supplying an appropriate @samp{qSupported} response
36168 (@pxref{qSupported}).
36170 @item qXfer:threads:read::@var{offset},@var{length}
36171 @anchor{qXfer threads read}
36172 Access the list of threads on target. @xref{Thread List Format}. The
36173 annex part of the generic @samp{qXfer} packet must be empty
36174 (@pxref{qXfer read}).
36176 This packet is not probed by default; the remote stub must request it,
36177 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36179 @item qXfer:traceframe-info:read::@var{offset},@var{length}
36180 @anchor{qXfer traceframe info read}
36182 Return a description of the current traceframe's contents.
36183 @xref{Traceframe Info Format}. The annex part of the generic
36184 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
36186 This packet is not probed by default; the remote stub must request it,
36187 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36189 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
36190 @anchor{qXfer unwind info block}
36192 Return the unwind information block for @var{pc}. This packet is used
36193 on OpenVMS/ia64 to ask the kernel unwind information.
36195 This packet is not probed by default.
36197 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
36198 @anchor{qXfer fdpic loadmap read}
36199 Read contents of @code{loadmap}s on the target system. The
36200 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
36201 executable @code{loadmap} or interpreter @code{loadmap} to read.
36203 This packet is not probed by default; the remote stub must request it,
36204 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36206 @item qXfer:osdata:read::@var{offset},@var{length}
36207 @anchor{qXfer osdata read}
36208 Access the target's @dfn{operating system information}.
36209 @xref{Operating System Information}.
36216 Data @var{data} (@pxref{Binary Data}) has been read from the
36217 target. There may be more data at a higher address (although
36218 it is permitted to return @samp{m} even for the last valid
36219 block of data, as long as at least one byte of data was read).
36220 It is possible for @var{data} to have fewer bytes than the @var{length} in the
36224 Data @var{data} (@pxref{Binary Data}) has been read from the target.
36225 There is no more data to be read. It is possible for @var{data} to
36226 have fewer bytes than the @var{length} in the request.
36229 The @var{offset} in the request is at the end of the data.
36230 There is no more data to be read.
36233 The request was malformed, or @var{annex} was invalid.
36236 The offset was invalid, or there was an error encountered reading the data.
36237 The @var{nn} part is a hex-encoded @code{errno} value.
36240 An empty reply indicates the @var{object} string was not recognized by
36241 the stub, or that the object does not support reading.
36244 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
36245 @cindex write data into object, remote request
36246 @anchor{qXfer write}
36247 Write uninterpreted bytes into the target's special data area
36248 identified by the keyword @var{object}, starting at @var{offset} bytes
36249 into the data. The binary-encoded data (@pxref{Binary Data}) to be
36250 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
36251 is specific to @var{object}; it can supply additional details about what data
36254 Here are the specific requests of this form defined so far. All
36255 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
36256 formats, listed below.
36259 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
36260 @anchor{qXfer siginfo write}
36261 Write @var{data} to the extra signal information on the target system.
36262 The annex part of the generic @samp{qXfer} packet must be
36263 empty (@pxref{qXfer write}).
36265 This packet is not probed by default; the remote stub must request it,
36266 by supplying an appropriate @samp{qSupported} response
36267 (@pxref{qSupported}).
36269 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
36270 @anchor{qXfer spu write}
36271 Write @var{data} to an @code{spufs} file on the target system. The
36272 annex specifies which file to write; it must be of the form
36273 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
36274 in the target process, and @var{name} identifes the @code{spufs} file
36275 in that context to be accessed.
36277 This packet is not probed by default; the remote stub must request it,
36278 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36284 @var{nn} (hex encoded) is the number of bytes written.
36285 This may be fewer bytes than supplied in the request.
36288 The request was malformed, or @var{annex} was invalid.
36291 The offset was invalid, or there was an error encountered writing the data.
36292 The @var{nn} part is a hex-encoded @code{errno} value.
36295 An empty reply indicates the @var{object} string was not
36296 recognized by the stub, or that the object does not support writing.
36299 @item qXfer:@var{object}:@var{operation}:@dots{}
36300 Requests of this form may be added in the future. When a stub does
36301 not recognize the @var{object} keyword, or its support for
36302 @var{object} does not recognize the @var{operation} keyword, the stub
36303 must respond with an empty packet.
36305 @item qAttached:@var{pid}
36306 @cindex query attached, remote request
36307 @cindex @samp{qAttached} packet
36308 Return an indication of whether the remote server attached to an
36309 existing process or created a new process. When the multiprocess
36310 protocol extensions are supported (@pxref{multiprocess extensions}),
36311 @var{pid} is an integer in hexadecimal format identifying the target
36312 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
36313 the query packet will be simplified as @samp{qAttached}.
36315 This query is used, for example, to know whether the remote process
36316 should be detached or killed when a @value{GDBN} session is ended with
36317 the @code{quit} command.
36322 The remote server attached to an existing process.
36324 The remote server created a new process.
36326 A badly formed request or an error was encountered.
36330 Enable branch tracing for the current thread using bts tracing.
36335 Branch tracing has been enabled.
36337 A badly formed request or an error was encountered.
36341 Disable branch tracing for the current thread.
36346 Branch tracing has been disabled.
36348 A badly formed request or an error was encountered.
36353 @node Architecture-Specific Protocol Details
36354 @section Architecture-Specific Protocol Details
36356 This section describes how the remote protocol is applied to specific
36357 target architectures. Also see @ref{Standard Target Features}, for
36358 details of XML target descriptions for each architecture.
36361 * ARM-Specific Protocol Details::
36362 * MIPS-Specific Protocol Details::
36365 @node ARM-Specific Protocol Details
36366 @subsection @acronym{ARM}-specific Protocol Details
36369 * ARM Breakpoint Kinds::
36372 @node ARM Breakpoint Kinds
36373 @subsubsection @acronym{ARM} Breakpoint Kinds
36374 @cindex breakpoint kinds, @acronym{ARM}
36376 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
36381 16-bit Thumb mode breakpoint.
36384 32-bit Thumb mode (Thumb-2) breakpoint.
36387 32-bit @acronym{ARM} mode breakpoint.
36391 @node MIPS-Specific Protocol Details
36392 @subsection @acronym{MIPS}-specific Protocol Details
36395 * MIPS Register packet Format::
36396 * MIPS Breakpoint Kinds::
36399 @node MIPS Register packet Format
36400 @subsubsection @acronym{MIPS} Register Packet Format
36401 @cindex register packet format, @acronym{MIPS}
36403 The following @code{g}/@code{G} packets have previously been defined.
36404 In the below, some thirty-two bit registers are transferred as
36405 sixty-four bits. Those registers should be zero/sign extended (which?)
36406 to fill the space allocated. Register bytes are transferred in target
36407 byte order. The two nibbles within a register byte are transferred
36408 most-significant -- least-significant.
36413 All registers are transferred as thirty-two bit quantities in the order:
36414 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
36415 registers; fsr; fir; fp.
36418 All registers are transferred as sixty-four bit quantities (including
36419 thirty-two bit registers such as @code{sr}). The ordering is the same
36424 @node MIPS Breakpoint Kinds
36425 @subsubsection @acronym{MIPS} Breakpoint Kinds
36426 @cindex breakpoint kinds, @acronym{MIPS}
36428 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
36433 16-bit @acronym{MIPS16} mode breakpoint.
36436 16-bit @acronym{microMIPS} mode breakpoint.
36439 32-bit standard @acronym{MIPS} mode breakpoint.
36442 32-bit @acronym{microMIPS} mode breakpoint.
36446 @node Tracepoint Packets
36447 @section Tracepoint Packets
36448 @cindex tracepoint packets
36449 @cindex packets, tracepoint
36451 Here we describe the packets @value{GDBN} uses to implement
36452 tracepoints (@pxref{Tracepoints}).
36456 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
36457 @cindex @samp{QTDP} packet
36458 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
36459 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
36460 the tracepoint is disabled. The @var{step} gives the tracepoint's step
36461 count, and @var{pass} gives its pass count. If an @samp{F} is present,
36462 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
36463 the number of bytes that the target should copy elsewhere to make room
36464 for the tracepoint. If an @samp{X} is present, it introduces a
36465 tracepoint condition, which consists of a hexadecimal length, followed
36466 by a comma and hex-encoded bytes, in a manner similar to action
36467 encodings as described below. If the trailing @samp{-} is present,
36468 further @samp{QTDP} packets will follow to specify this tracepoint's
36474 The packet was understood and carried out.
36476 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
36478 The packet was not recognized.
36481 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
36482 Define actions to be taken when a tracepoint is hit. The @var{n} and
36483 @var{addr} must be the same as in the initial @samp{QTDP} packet for
36484 this tracepoint. This packet may only be sent immediately after
36485 another @samp{QTDP} packet that ended with a @samp{-}. If the
36486 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
36487 specifying more actions for this tracepoint.
36489 In the series of action packets for a given tracepoint, at most one
36490 can have an @samp{S} before its first @var{action}. If such a packet
36491 is sent, it and the following packets define ``while-stepping''
36492 actions. Any prior packets define ordinary actions --- that is, those
36493 taken when the tracepoint is first hit. If no action packet has an
36494 @samp{S}, then all the packets in the series specify ordinary
36495 tracepoint actions.
36497 The @samp{@var{action}@dots{}} portion of the packet is a series of
36498 actions, concatenated without separators. Each action has one of the
36504 Collect the registers whose bits are set in @var{mask},
36505 a hexadecimal number whose @var{i}'th bit is set if register number
36506 @var{i} should be collected. (The least significant bit is numbered
36507 zero.) Note that @var{mask} may be any number of digits long; it may
36508 not fit in a 32-bit word.
36510 @item M @var{basereg},@var{offset},@var{len}
36511 Collect @var{len} bytes of memory starting at the address in register
36512 number @var{basereg}, plus @var{offset}. If @var{basereg} is
36513 @samp{-1}, then the range has a fixed address: @var{offset} is the
36514 address of the lowest byte to collect. The @var{basereg},
36515 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
36516 values (the @samp{-1} value for @var{basereg} is a special case).
36518 @item X @var{len},@var{expr}
36519 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
36520 it directs. The agent expression @var{expr} is as described in
36521 @ref{Agent Expressions}. Each byte of the expression is encoded as a
36522 two-digit hex number in the packet; @var{len} is the number of bytes
36523 in the expression (and thus one-half the number of hex digits in the
36528 Any number of actions may be packed together in a single @samp{QTDP}
36529 packet, as long as the packet does not exceed the maximum packet
36530 length (400 bytes, for many stubs). There may be only one @samp{R}
36531 action per tracepoint, and it must precede any @samp{M} or @samp{X}
36532 actions. Any registers referred to by @samp{M} and @samp{X} actions
36533 must be collected by a preceding @samp{R} action. (The
36534 ``while-stepping'' actions are treated as if they were attached to a
36535 separate tracepoint, as far as these restrictions are concerned.)
36540 The packet was understood and carried out.
36542 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
36544 The packet was not recognized.
36547 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
36548 @cindex @samp{QTDPsrc} packet
36549 Specify a source string of tracepoint @var{n} at address @var{addr}.
36550 This is useful to get accurate reproduction of the tracepoints
36551 originally downloaded at the beginning of the trace run. The @var{type}
36552 is the name of the tracepoint part, such as @samp{cond} for the
36553 tracepoint's conditional expression (see below for a list of types), while
36554 @var{bytes} is the string, encoded in hexadecimal.
36556 @var{start} is the offset of the @var{bytes} within the overall source
36557 string, while @var{slen} is the total length of the source string.
36558 This is intended for handling source strings that are longer than will
36559 fit in a single packet.
36560 @c Add detailed example when this info is moved into a dedicated
36561 @c tracepoint descriptions section.
36563 The available string types are @samp{at} for the location,
36564 @samp{cond} for the conditional, and @samp{cmd} for an action command.
36565 @value{GDBN} sends a separate packet for each command in the action
36566 list, in the same order in which the commands are stored in the list.
36568 The target does not need to do anything with source strings except
36569 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
36572 Although this packet is optional, and @value{GDBN} will only send it
36573 if the target replies with @samp{TracepointSource} @xref{General
36574 Query Packets}, it makes both disconnected tracing and trace files
36575 much easier to use. Otherwise the user must be careful that the
36576 tracepoints in effect while looking at trace frames are identical to
36577 the ones in effect during the trace run; even a small discrepancy
36578 could cause @samp{tdump} not to work, or a particular trace frame not
36581 @item QTDV:@var{n}:@var{value}
36582 @cindex define trace state variable, remote request
36583 @cindex @samp{QTDV} packet
36584 Create a new trace state variable, number @var{n}, with an initial
36585 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
36586 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
36587 the option of not using this packet for initial values of zero; the
36588 target should simply create the trace state variables as they are
36589 mentioned in expressions.
36591 @item QTFrame:@var{n}
36592 @cindex @samp{QTFrame} packet
36593 Select the @var{n}'th tracepoint frame from the buffer, and use the
36594 register and memory contents recorded there to answer subsequent
36595 request packets from @value{GDBN}.
36597 A successful reply from the stub indicates that the stub has found the
36598 requested frame. The response is a series of parts, concatenated
36599 without separators, describing the frame we selected. Each part has
36600 one of the following forms:
36604 The selected frame is number @var{n} in the trace frame buffer;
36605 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
36606 was no frame matching the criteria in the request packet.
36609 The selected trace frame records a hit of tracepoint number @var{t};
36610 @var{t} is a hexadecimal number.
36614 @item QTFrame:pc:@var{addr}
36615 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
36616 currently selected frame whose PC is @var{addr};
36617 @var{addr} is a hexadecimal number.
36619 @item QTFrame:tdp:@var{t}
36620 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
36621 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
36622 is a hexadecimal number.
36624 @item QTFrame:range:@var{start}:@var{end}
36625 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
36626 currently selected frame whose PC is between @var{start} (inclusive)
36627 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
36630 @item QTFrame:outside:@var{start}:@var{end}
36631 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
36632 frame @emph{outside} the given range of addresses (exclusive).
36635 @cindex @samp{qTMinFTPILen} packet
36636 This packet requests the minimum length of instruction at which a fast
36637 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
36638 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
36639 it depends on the target system being able to create trampolines in
36640 the first 64K of memory, which might or might not be possible for that
36641 system. So the reply to this packet will be 4 if it is able to
36648 The minimum instruction length is currently unknown.
36650 The minimum instruction length is @var{length}, where @var{length}
36651 is a hexadecimal number greater or equal to 1. A reply
36652 of 1 means that a fast tracepoint may be placed on any instruction
36653 regardless of size.
36655 An error has occurred.
36657 An empty reply indicates that the request is not supported by the stub.
36661 @cindex @samp{QTStart} packet
36662 Begin the tracepoint experiment. Begin collecting data from
36663 tracepoint hits in the trace frame buffer. This packet supports the
36664 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
36665 instruction reply packet}).
36668 @cindex @samp{QTStop} packet
36669 End the tracepoint experiment. Stop collecting trace frames.
36671 @item QTEnable:@var{n}:@var{addr}
36673 @cindex @samp{QTEnable} packet
36674 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
36675 experiment. If the tracepoint was previously disabled, then collection
36676 of data from it will resume.
36678 @item QTDisable:@var{n}:@var{addr}
36680 @cindex @samp{QTDisable} packet
36681 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
36682 experiment. No more data will be collected from the tracepoint unless
36683 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
36686 @cindex @samp{QTinit} packet
36687 Clear the table of tracepoints, and empty the trace frame buffer.
36689 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
36690 @cindex @samp{QTro} packet
36691 Establish the given ranges of memory as ``transparent''. The stub
36692 will answer requests for these ranges from memory's current contents,
36693 if they were not collected as part of the tracepoint hit.
36695 @value{GDBN} uses this to mark read-only regions of memory, like those
36696 containing program code. Since these areas never change, they should
36697 still have the same contents they did when the tracepoint was hit, so
36698 there's no reason for the stub to refuse to provide their contents.
36700 @item QTDisconnected:@var{value}
36701 @cindex @samp{QTDisconnected} packet
36702 Set the choice to what to do with the tracing run when @value{GDBN}
36703 disconnects from the target. A @var{value} of 1 directs the target to
36704 continue the tracing run, while 0 tells the target to stop tracing if
36705 @value{GDBN} is no longer in the picture.
36708 @cindex @samp{qTStatus} packet
36709 Ask the stub if there is a trace experiment running right now.
36711 The reply has the form:
36715 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
36716 @var{running} is a single digit @code{1} if the trace is presently
36717 running, or @code{0} if not. It is followed by semicolon-separated
36718 optional fields that an agent may use to report additional status.
36722 If the trace is not running, the agent may report any of several
36723 explanations as one of the optional fields:
36728 No trace has been run yet.
36730 @item tstop[:@var{text}]:0
36731 The trace was stopped by a user-originated stop command. The optional
36732 @var{text} field is a user-supplied string supplied as part of the
36733 stop command (for instance, an explanation of why the trace was
36734 stopped manually). It is hex-encoded.
36737 The trace stopped because the trace buffer filled up.
36739 @item tdisconnected:0
36740 The trace stopped because @value{GDBN} disconnected from the target.
36742 @item tpasscount:@var{tpnum}
36743 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
36745 @item terror:@var{text}:@var{tpnum}
36746 The trace stopped because tracepoint @var{tpnum} had an error. The
36747 string @var{text} is available to describe the nature of the error
36748 (for instance, a divide by zero in the condition expression); it
36752 The trace stopped for some other reason.
36756 Additional optional fields supply statistical and other information.
36757 Although not required, they are extremely useful for users monitoring
36758 the progress of a trace run. If a trace has stopped, and these
36759 numbers are reported, they must reflect the state of the just-stopped
36764 @item tframes:@var{n}
36765 The number of trace frames in the buffer.
36767 @item tcreated:@var{n}
36768 The total number of trace frames created during the run. This may
36769 be larger than the trace frame count, if the buffer is circular.
36771 @item tsize:@var{n}
36772 The total size of the trace buffer, in bytes.
36774 @item tfree:@var{n}
36775 The number of bytes still unused in the buffer.
36777 @item circular:@var{n}
36778 The value of the circular trace buffer flag. @code{1} means that the
36779 trace buffer is circular and old trace frames will be discarded if
36780 necessary to make room, @code{0} means that the trace buffer is linear
36783 @item disconn:@var{n}
36784 The value of the disconnected tracing flag. @code{1} means that
36785 tracing will continue after @value{GDBN} disconnects, @code{0} means
36786 that the trace run will stop.
36790 @item qTP:@var{tp}:@var{addr}
36791 @cindex tracepoint status, remote request
36792 @cindex @samp{qTP} packet
36793 Ask the stub for the current state of tracepoint number @var{tp} at
36794 address @var{addr}.
36798 @item V@var{hits}:@var{usage}
36799 The tracepoint has been hit @var{hits} times so far during the trace
36800 run, and accounts for @var{usage} in the trace buffer. Note that
36801 @code{while-stepping} steps are not counted as separate hits, but the
36802 steps' space consumption is added into the usage number.
36806 @item qTV:@var{var}
36807 @cindex trace state variable value, remote request
36808 @cindex @samp{qTV} packet
36809 Ask the stub for the value of the trace state variable number @var{var}.
36814 The value of the variable is @var{value}. This will be the current
36815 value of the variable if the user is examining a running target, or a
36816 saved value if the variable was collected in the trace frame that the
36817 user is looking at. Note that multiple requests may result in
36818 different reply values, such as when requesting values while the
36819 program is running.
36822 The value of the variable is unknown. This would occur, for example,
36823 if the user is examining a trace frame in which the requested variable
36828 @cindex @samp{qTfP} packet
36830 @cindex @samp{qTsP} packet
36831 These packets request data about tracepoints that are being used by
36832 the target. @value{GDBN} sends @code{qTfP} to get the first piece
36833 of data, and multiple @code{qTsP} to get additional pieces. Replies
36834 to these packets generally take the form of the @code{QTDP} packets
36835 that define tracepoints. (FIXME add detailed syntax)
36838 @cindex @samp{qTfV} packet
36840 @cindex @samp{qTsV} packet
36841 These packets request data about trace state variables that are on the
36842 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
36843 and multiple @code{qTsV} to get additional variables. Replies to
36844 these packets follow the syntax of the @code{QTDV} packets that define
36845 trace state variables.
36851 @cindex @samp{qTfSTM} packet
36852 @cindex @samp{qTsSTM} packet
36853 These packets request data about static tracepoint markers that exist
36854 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
36855 first piece of data, and multiple @code{qTsSTM} to get additional
36856 pieces. Replies to these packets take the following form:
36860 @item m @var{address}:@var{id}:@var{extra}
36862 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
36863 a comma-separated list of markers
36865 (lower case letter @samp{L}) denotes end of list.
36867 An error occurred. The error number @var{nn} is given as hex digits.
36869 An empty reply indicates that the request is not supported by the
36873 The @var{address} is encoded in hex;
36874 @var{id} and @var{extra} are strings encoded in hex.
36876 In response to each query, the target will reply with a list of one or
36877 more markers, separated by commas. @value{GDBN} will respond to each
36878 reply with a request for more markers (using the @samp{qs} form of the
36879 query), until the target responds with @samp{l} (lower-case ell, for
36882 @item qTSTMat:@var{address}
36884 @cindex @samp{qTSTMat} packet
36885 This packets requests data about static tracepoint markers in the
36886 target program at @var{address}. Replies to this packet follow the
36887 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
36888 tracepoint markers.
36890 @item QTSave:@var{filename}
36891 @cindex @samp{QTSave} packet
36892 This packet directs the target to save trace data to the file name
36893 @var{filename} in the target's filesystem. The @var{filename} is encoded
36894 as a hex string; the interpretation of the file name (relative vs
36895 absolute, wild cards, etc) is up to the target.
36897 @item qTBuffer:@var{offset},@var{len}
36898 @cindex @samp{qTBuffer} packet
36899 Return up to @var{len} bytes of the current contents of trace buffer,
36900 starting at @var{offset}. The trace buffer is treated as if it were
36901 a contiguous collection of traceframes, as per the trace file format.
36902 The reply consists as many hex-encoded bytes as the target can deliver
36903 in a packet; it is not an error to return fewer than were asked for.
36904 A reply consisting of just @code{l} indicates that no bytes are
36907 @item QTBuffer:circular:@var{value}
36908 This packet directs the target to use a circular trace buffer if
36909 @var{value} is 1, or a linear buffer if the value is 0.
36911 @item QTBuffer:size:@var{size}
36912 @anchor{QTBuffer-size}
36913 @cindex @samp{QTBuffer size} packet
36914 This packet directs the target to make the trace buffer be of size
36915 @var{size} if possible. A value of @code{-1} tells the target to
36916 use whatever size it prefers.
36918 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
36919 @cindex @samp{QTNotes} packet
36920 This packet adds optional textual notes to the trace run. Allowable
36921 types include @code{user}, @code{notes}, and @code{tstop}, the
36922 @var{text} fields are arbitrary strings, hex-encoded.
36926 @subsection Relocate instruction reply packet
36927 When installing fast tracepoints in memory, the target may need to
36928 relocate the instruction currently at the tracepoint address to a
36929 different address in memory. For most instructions, a simple copy is
36930 enough, but, for example, call instructions that implicitly push the
36931 return address on the stack, and relative branches or other
36932 PC-relative instructions require offset adjustment, so that the effect
36933 of executing the instruction at a different address is the same as if
36934 it had executed in the original location.
36936 In response to several of the tracepoint packets, the target may also
36937 respond with a number of intermediate @samp{qRelocInsn} request
36938 packets before the final result packet, to have @value{GDBN} handle
36939 this relocation operation. If a packet supports this mechanism, its
36940 documentation will explicitly say so. See for example the above
36941 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
36942 format of the request is:
36945 @item qRelocInsn:@var{from};@var{to}
36947 This requests @value{GDBN} to copy instruction at address @var{from}
36948 to address @var{to}, possibly adjusted so that executing the
36949 instruction at @var{to} has the same effect as executing it at
36950 @var{from}. @value{GDBN} writes the adjusted instruction to target
36951 memory starting at @var{to}.
36956 @item qRelocInsn:@var{adjusted_size}
36957 Informs the stub the relocation is complete. The @var{adjusted_size} is
36958 the length in bytes of resulting relocated instruction sequence.
36960 A badly formed request was detected, or an error was encountered while
36961 relocating the instruction.
36964 @node Host I/O Packets
36965 @section Host I/O Packets
36966 @cindex Host I/O, remote protocol
36967 @cindex file transfer, remote protocol
36969 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
36970 operations on the far side of a remote link. For example, Host I/O is
36971 used to upload and download files to a remote target with its own
36972 filesystem. Host I/O uses the same constant values and data structure
36973 layout as the target-initiated File-I/O protocol. However, the
36974 Host I/O packets are structured differently. The target-initiated
36975 protocol relies on target memory to store parameters and buffers.
36976 Host I/O requests are initiated by @value{GDBN}, and the
36977 target's memory is not involved. @xref{File-I/O Remote Protocol
36978 Extension}, for more details on the target-initiated protocol.
36980 The Host I/O request packets all encode a single operation along with
36981 its arguments. They have this format:
36985 @item vFile:@var{operation}: @var{parameter}@dots{}
36986 @var{operation} is the name of the particular request; the target
36987 should compare the entire packet name up to the second colon when checking
36988 for a supported operation. The format of @var{parameter} depends on
36989 the operation. Numbers are always passed in hexadecimal. Negative
36990 numbers have an explicit minus sign (i.e.@: two's complement is not
36991 used). Strings (e.g.@: filenames) are encoded as a series of
36992 hexadecimal bytes. The last argument to a system call may be a
36993 buffer of escaped binary data (@pxref{Binary Data}).
36997 The valid responses to Host I/O packets are:
37001 @item F @var{result} [, @var{errno}] [; @var{attachment}]
37002 @var{result} is the integer value returned by this operation, usually
37003 non-negative for success and -1 for errors. If an error has occured,
37004 @var{errno} will be included in the result specifying a
37005 value defined by the File-I/O protocol (@pxref{Errno Values}). For
37006 operations which return data, @var{attachment} supplies the data as a
37007 binary buffer. Binary buffers in response packets are escaped in the
37008 normal way (@pxref{Binary Data}). See the individual packet
37009 documentation for the interpretation of @var{result} and
37013 An empty response indicates that this operation is not recognized.
37017 These are the supported Host I/O operations:
37020 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
37021 Open a file at @var{filename} and return a file descriptor for it, or
37022 return -1 if an error occurs. The @var{filename} is a string,
37023 @var{flags} is an integer indicating a mask of open flags
37024 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
37025 of mode bits to use if the file is created (@pxref{mode_t Values}).
37026 @xref{open}, for details of the open flags and mode values.
37028 @item vFile:close: @var{fd}
37029 Close the open file corresponding to @var{fd} and return 0, or
37030 -1 if an error occurs.
37032 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
37033 Read data from the open file corresponding to @var{fd}. Up to
37034 @var{count} bytes will be read from the file, starting at @var{offset}
37035 relative to the start of the file. The target may read fewer bytes;
37036 common reasons include packet size limits and an end-of-file
37037 condition. The number of bytes read is returned. Zero should only be
37038 returned for a successful read at the end of the file, or if
37039 @var{count} was zero.
37041 The data read should be returned as a binary attachment on success.
37042 If zero bytes were read, the response should include an empty binary
37043 attachment (i.e.@: a trailing semicolon). The return value is the
37044 number of target bytes read; the binary attachment may be longer if
37045 some characters were escaped.
37047 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
37048 Write @var{data} (a binary buffer) to the open file corresponding
37049 to @var{fd}. Start the write at @var{offset} from the start of the
37050 file. Unlike many @code{write} system calls, there is no
37051 separate @var{count} argument; the length of @var{data} in the
37052 packet is used. @samp{vFile:write} returns the number of bytes written,
37053 which may be shorter than the length of @var{data}, or -1 if an
37056 @item vFile:unlink: @var{filename}
37057 Delete the file at @var{filename} on the target. Return 0,
37058 or -1 if an error occurs. The @var{filename} is a string.
37060 @item vFile:readlink: @var{filename}
37061 Read value of symbolic link @var{filename} on the target. Return
37062 the number of bytes read, or -1 if an error occurs.
37064 The data read should be returned as a binary attachment on success.
37065 If zero bytes were read, the response should include an empty binary
37066 attachment (i.e.@: a trailing semicolon). The return value is the
37067 number of target bytes read; the binary attachment may be longer if
37068 some characters were escaped.
37073 @section Interrupts
37074 @cindex interrupts (remote protocol)
37076 When a program on the remote target is running, @value{GDBN} may
37077 attempt to interrupt it by sending a @samp{Ctrl-C}, @code{BREAK} or
37078 a @code{BREAK} followed by @code{g},
37079 control of which is specified via @value{GDBN}'s @samp{interrupt-sequence}.
37081 The precise meaning of @code{BREAK} is defined by the transport
37082 mechanism and may, in fact, be undefined. @value{GDBN} does not
37083 currently define a @code{BREAK} mechanism for any of the network
37084 interfaces except for TCP, in which case @value{GDBN} sends the
37085 @code{telnet} BREAK sequence.
37087 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
37088 transport mechanisms. It is represented by sending the single byte
37089 @code{0x03} without any of the usual packet overhead described in
37090 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
37091 transmitted as part of a packet, it is considered to be packet data
37092 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
37093 (@pxref{X packet}), used for binary downloads, may include an unescaped
37094 @code{0x03} as part of its packet.
37096 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
37097 When Linux kernel receives this sequence from serial port,
37098 it stops execution and connects to gdb.
37100 Stubs are not required to recognize these interrupt mechanisms and the
37101 precise meaning associated with receipt of the interrupt is
37102 implementation defined. If the target supports debugging of multiple
37103 threads and/or processes, it should attempt to interrupt all
37104 currently-executing threads and processes.
37105 If the stub is successful at interrupting the
37106 running program, it should send one of the stop
37107 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
37108 of successfully stopping the program in all-stop mode, and a stop reply
37109 for each stopped thread in non-stop mode.
37110 Interrupts received while the
37111 program is stopped are discarded.
37113 @node Notification Packets
37114 @section Notification Packets
37115 @cindex notification packets
37116 @cindex packets, notification
37118 The @value{GDBN} remote serial protocol includes @dfn{notifications},
37119 packets that require no acknowledgment. Both the GDB and the stub
37120 may send notifications (although the only notifications defined at
37121 present are sent by the stub). Notifications carry information
37122 without incurring the round-trip latency of an acknowledgment, and so
37123 are useful for low-impact communications where occasional packet loss
37126 A notification packet has the form @samp{% @var{data} #
37127 @var{checksum}}, where @var{data} is the content of the notification,
37128 and @var{checksum} is a checksum of @var{data}, computed and formatted
37129 as for ordinary @value{GDBN} packets. A notification's @var{data}
37130 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
37131 receiving a notification, the recipient sends no @samp{+} or @samp{-}
37132 to acknowledge the notification's receipt or to report its corruption.
37134 Every notification's @var{data} begins with a name, which contains no
37135 colon characters, followed by a colon character.
37137 Recipients should silently ignore corrupted notifications and
37138 notifications they do not understand. Recipients should restart
37139 timeout periods on receipt of a well-formed notification, whether or
37140 not they understand it.
37142 Senders should only send the notifications described here when this
37143 protocol description specifies that they are permitted. In the
37144 future, we may extend the protocol to permit existing notifications in
37145 new contexts; this rule helps older senders avoid confusing newer
37148 (Older versions of @value{GDBN} ignore bytes received until they see
37149 the @samp{$} byte that begins an ordinary packet, so new stubs may
37150 transmit notifications without fear of confusing older clients. There
37151 are no notifications defined for @value{GDBN} to send at the moment, but we
37152 assume that most older stubs would ignore them, as well.)
37154 Each notification is comprised of three parts:
37156 @item @var{name}:@var{event}
37157 The notification packet is sent by the side that initiates the
37158 exchange (currently, only the stub does that), with @var{event}
37159 carrying the specific information about the notification, and
37160 @var{name} specifying the name of the notification.
37162 The acknowledge sent by the other side, usually @value{GDBN}, to
37163 acknowledge the exchange and request the event.
37166 The purpose of an asynchronous notification mechanism is to report to
37167 @value{GDBN} that something interesting happened in the remote stub.
37169 The remote stub may send notification @var{name}:@var{event}
37170 at any time, but @value{GDBN} acknowledges the notification when
37171 appropriate. The notification event is pending before @value{GDBN}
37172 acknowledges. Only one notification at a time may be pending; if
37173 additional events occur before @value{GDBN} has acknowledged the
37174 previous notification, they must be queued by the stub for later
37175 synchronous transmission in response to @var{ack} packets from
37176 @value{GDBN}. Because the notification mechanism is unreliable,
37177 the stub is permitted to resend a notification if it believes
37178 @value{GDBN} may not have received it.
37180 Specifically, notifications may appear when @value{GDBN} is not
37181 otherwise reading input from the stub, or when @value{GDBN} is
37182 expecting to read a normal synchronous response or a
37183 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
37184 Notification packets are distinct from any other communication from
37185 the stub so there is no ambiguity.
37187 After receiving a notification, @value{GDBN} shall acknowledge it by
37188 sending a @var{ack} packet as a regular, synchronous request to the
37189 stub. Such acknowledgment is not required to happen immediately, as
37190 @value{GDBN} is permitted to send other, unrelated packets to the
37191 stub first, which the stub should process normally.
37193 Upon receiving a @var{ack} packet, if the stub has other queued
37194 events to report to @value{GDBN}, it shall respond by sending a
37195 normal @var{event}. @value{GDBN} shall then send another @var{ack}
37196 packet to solicit further responses; again, it is permitted to send
37197 other, unrelated packets as well which the stub should process
37200 If the stub receives a @var{ack} packet and there are no additional
37201 @var{event} to report, the stub shall return an @samp{OK} response.
37202 At this point, @value{GDBN} has finished processing a notification
37203 and the stub has completed sending any queued events. @value{GDBN}
37204 won't accept any new notifications until the final @samp{OK} is
37205 received . If further notification events occur, the stub shall send
37206 a new notification, @value{GDBN} shall accept the notification, and
37207 the process shall be repeated.
37209 The process of asynchronous notification can be illustrated by the
37212 <- @code{%%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
37215 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
37217 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
37222 The following notifications are defined:
37223 @multitable @columnfractions 0.12 0.12 0.38 0.38
37232 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
37233 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
37234 for information on how these notifications are acknowledged by
37236 @tab Report an asynchronous stop event in non-stop mode.
37240 @node Remote Non-Stop
37241 @section Remote Protocol Support for Non-Stop Mode
37243 @value{GDBN}'s remote protocol supports non-stop debugging of
37244 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
37245 supports non-stop mode, it should report that to @value{GDBN} by including
37246 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
37248 @value{GDBN} typically sends a @samp{QNonStop} packet only when
37249 establishing a new connection with the stub. Entering non-stop mode
37250 does not alter the state of any currently-running threads, but targets
37251 must stop all threads in any already-attached processes when entering
37252 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
37253 probe the target state after a mode change.
37255 In non-stop mode, when an attached process encounters an event that
37256 would otherwise be reported with a stop reply, it uses the
37257 asynchronous notification mechanism (@pxref{Notification Packets}) to
37258 inform @value{GDBN}. In contrast to all-stop mode, where all threads
37259 in all processes are stopped when a stop reply is sent, in non-stop
37260 mode only the thread reporting the stop event is stopped. That is,
37261 when reporting a @samp{S} or @samp{T} response to indicate completion
37262 of a step operation, hitting a breakpoint, or a fault, only the
37263 affected thread is stopped; any other still-running threads continue
37264 to run. When reporting a @samp{W} or @samp{X} response, all running
37265 threads belonging to other attached processes continue to run.
37267 In non-stop mode, the target shall respond to the @samp{?} packet as
37268 follows. First, any incomplete stop reply notification/@samp{vStopped}
37269 sequence in progress is abandoned. The target must begin a new
37270 sequence reporting stop events for all stopped threads, whether or not
37271 it has previously reported those events to @value{GDBN}. The first
37272 stop reply is sent as a synchronous reply to the @samp{?} packet, and
37273 subsequent stop replies are sent as responses to @samp{vStopped} packets
37274 using the mechanism described above. The target must not send
37275 asynchronous stop reply notifications until the sequence is complete.
37276 If all threads are running when the target receives the @samp{?} packet,
37277 or if the target is not attached to any process, it shall respond
37280 @node Packet Acknowledgment
37281 @section Packet Acknowledgment
37283 @cindex acknowledgment, for @value{GDBN} remote
37284 @cindex packet acknowledgment, for @value{GDBN} remote
37285 By default, when either the host or the target machine receives a packet,
37286 the first response expected is an acknowledgment: either @samp{+} (to indicate
37287 the package was received correctly) or @samp{-} (to request retransmission).
37288 This mechanism allows the @value{GDBN} remote protocol to operate over
37289 unreliable transport mechanisms, such as a serial line.
37291 In cases where the transport mechanism is itself reliable (such as a pipe or
37292 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
37293 It may be desirable to disable them in that case to reduce communication
37294 overhead, or for other reasons. This can be accomplished by means of the
37295 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
37297 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
37298 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
37299 and response format still includes the normal checksum, as described in
37300 @ref{Overview}, but the checksum may be ignored by the receiver.
37302 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
37303 no-acknowledgment mode, it should report that to @value{GDBN}
37304 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
37305 @pxref{qSupported}.
37306 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
37307 disabled via the @code{set remote noack-packet off} command
37308 (@pxref{Remote Configuration}),
37309 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
37310 Only then may the stub actually turn off packet acknowledgments.
37311 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
37312 response, which can be safely ignored by the stub.
37314 Note that @code{set remote noack-packet} command only affects negotiation
37315 between @value{GDBN} and the stub when subsequent connections are made;
37316 it does not affect the protocol acknowledgment state for any current
37318 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
37319 new connection is established,
37320 there is also no protocol request to re-enable the acknowledgments
37321 for the current connection, once disabled.
37326 Example sequence of a target being re-started. Notice how the restart
37327 does not get any direct output:
37332 @emph{target restarts}
37335 <- @code{T001:1234123412341234}
37339 Example sequence of a target being stepped by a single instruction:
37342 -> @code{G1445@dots{}}
37347 <- @code{T001:1234123412341234}
37351 <- @code{1455@dots{}}
37355 @node File-I/O Remote Protocol Extension
37356 @section File-I/O Remote Protocol Extension
37357 @cindex File-I/O remote protocol extension
37360 * File-I/O Overview::
37361 * Protocol Basics::
37362 * The F Request Packet::
37363 * The F Reply Packet::
37364 * The Ctrl-C Message::
37366 * List of Supported Calls::
37367 * Protocol-specific Representation of Datatypes::
37369 * File-I/O Examples::
37372 @node File-I/O Overview
37373 @subsection File-I/O Overview
37374 @cindex file-i/o overview
37376 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
37377 target to use the host's file system and console I/O to perform various
37378 system calls. System calls on the target system are translated into a
37379 remote protocol packet to the host system, which then performs the needed
37380 actions and returns a response packet to the target system.
37381 This simulates file system operations even on targets that lack file systems.
37383 The protocol is defined to be independent of both the host and target systems.
37384 It uses its own internal representation of datatypes and values. Both
37385 @value{GDBN} and the target's @value{GDBN} stub are responsible for
37386 translating the system-dependent value representations into the internal
37387 protocol representations when data is transmitted.
37389 The communication is synchronous. A system call is possible only when
37390 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
37391 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
37392 the target is stopped to allow deterministic access to the target's
37393 memory. Therefore File-I/O is not interruptible by target signals. On
37394 the other hand, it is possible to interrupt File-I/O by a user interrupt
37395 (@samp{Ctrl-C}) within @value{GDBN}.
37397 The target's request to perform a host system call does not finish
37398 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
37399 after finishing the system call, the target returns to continuing the
37400 previous activity (continue, step). No additional continue or step
37401 request from @value{GDBN} is required.
37404 (@value{GDBP}) continue
37405 <- target requests 'system call X'
37406 target is stopped, @value{GDBN} executes system call
37407 -> @value{GDBN} returns result
37408 ... target continues, @value{GDBN} returns to wait for the target
37409 <- target hits breakpoint and sends a Txx packet
37412 The protocol only supports I/O on the console and to regular files on
37413 the host file system. Character or block special devices, pipes,
37414 named pipes, sockets or any other communication method on the host
37415 system are not supported by this protocol.
37417 File I/O is not supported in non-stop mode.
37419 @node Protocol Basics
37420 @subsection Protocol Basics
37421 @cindex protocol basics, file-i/o
37423 The File-I/O protocol uses the @code{F} packet as the request as well
37424 as reply packet. Since a File-I/O system call can only occur when
37425 @value{GDBN} is waiting for a response from the continuing or stepping target,
37426 the File-I/O request is a reply that @value{GDBN} has to expect as a result
37427 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
37428 This @code{F} packet contains all information needed to allow @value{GDBN}
37429 to call the appropriate host system call:
37433 A unique identifier for the requested system call.
37436 All parameters to the system call. Pointers are given as addresses
37437 in the target memory address space. Pointers to strings are given as
37438 pointer/length pair. Numerical values are given as they are.
37439 Numerical control flags are given in a protocol-specific representation.
37443 At this point, @value{GDBN} has to perform the following actions.
37447 If the parameters include pointer values to data needed as input to a
37448 system call, @value{GDBN} requests this data from the target with a
37449 standard @code{m} packet request. This additional communication has to be
37450 expected by the target implementation and is handled as any other @code{m}
37454 @value{GDBN} translates all value from protocol representation to host
37455 representation as needed. Datatypes are coerced into the host types.
37458 @value{GDBN} calls the system call.
37461 It then coerces datatypes back to protocol representation.
37464 If the system call is expected to return data in buffer space specified
37465 by pointer parameters to the call, the data is transmitted to the
37466 target using a @code{M} or @code{X} packet. This packet has to be expected
37467 by the target implementation and is handled as any other @code{M} or @code{X}
37472 Eventually @value{GDBN} replies with another @code{F} packet which contains all
37473 necessary information for the target to continue. This at least contains
37480 @code{errno}, if has been changed by the system call.
37487 After having done the needed type and value coercion, the target continues
37488 the latest continue or step action.
37490 @node The F Request Packet
37491 @subsection The @code{F} Request Packet
37492 @cindex file-i/o request packet
37493 @cindex @code{F} request packet
37495 The @code{F} request packet has the following format:
37498 @item F@var{call-id},@var{parameter@dots{}}
37500 @var{call-id} is the identifier to indicate the host system call to be called.
37501 This is just the name of the function.
37503 @var{parameter@dots{}} are the parameters to the system call.
37504 Parameters are hexadecimal integer values, either the actual values in case
37505 of scalar datatypes, pointers to target buffer space in case of compound
37506 datatypes and unspecified memory areas, or pointer/length pairs in case
37507 of string parameters. These are appended to the @var{call-id} as a
37508 comma-delimited list. All values are transmitted in ASCII
37509 string representation, pointer/length pairs separated by a slash.
37515 @node The F Reply Packet
37516 @subsection The @code{F} Reply Packet
37517 @cindex file-i/o reply packet
37518 @cindex @code{F} reply packet
37520 The @code{F} reply packet has the following format:
37524 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
37526 @var{retcode} is the return code of the system call as hexadecimal value.
37528 @var{errno} is the @code{errno} set by the call, in protocol-specific
37530 This parameter can be omitted if the call was successful.
37532 @var{Ctrl-C flag} is only sent if the user requested a break. In this
37533 case, @var{errno} must be sent as well, even if the call was successful.
37534 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
37541 or, if the call was interrupted before the host call has been performed:
37548 assuming 4 is the protocol-specific representation of @code{EINTR}.
37553 @node The Ctrl-C Message
37554 @subsection The @samp{Ctrl-C} Message
37555 @cindex ctrl-c message, in file-i/o protocol
37557 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
37558 reply packet (@pxref{The F Reply Packet}),
37559 the target should behave as if it had
37560 gotten a break message. The meaning for the target is ``system call
37561 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
37562 (as with a break message) and return to @value{GDBN} with a @code{T02}
37565 It's important for the target to know in which
37566 state the system call was interrupted. There are two possible cases:
37570 The system call hasn't been performed on the host yet.
37573 The system call on the host has been finished.
37577 These two states can be distinguished by the target by the value of the
37578 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
37579 call hasn't been performed. This is equivalent to the @code{EINTR} handling
37580 on POSIX systems. In any other case, the target may presume that the
37581 system call has been finished --- successfully or not --- and should behave
37582 as if the break message arrived right after the system call.
37584 @value{GDBN} must behave reliably. If the system call has not been called
37585 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
37586 @code{errno} in the packet. If the system call on the host has been finished
37587 before the user requests a break, the full action must be finished by
37588 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
37589 The @code{F} packet may only be sent when either nothing has happened
37590 or the full action has been completed.
37593 @subsection Console I/O
37594 @cindex console i/o as part of file-i/o
37596 By default and if not explicitly closed by the target system, the file
37597 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
37598 on the @value{GDBN} console is handled as any other file output operation
37599 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
37600 by @value{GDBN} so that after the target read request from file descriptor
37601 0 all following typing is buffered until either one of the following
37606 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
37608 system call is treated as finished.
37611 The user presses @key{RET}. This is treated as end of input with a trailing
37615 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
37616 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
37620 If the user has typed more characters than fit in the buffer given to
37621 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
37622 either another @code{read(0, @dots{})} is requested by the target, or debugging
37623 is stopped at the user's request.
37626 @node List of Supported Calls
37627 @subsection List of Supported Calls
37628 @cindex list of supported file-i/o calls
37645 @unnumberedsubsubsec open
37646 @cindex open, file-i/o system call
37651 int open(const char *pathname, int flags);
37652 int open(const char *pathname, int flags, mode_t mode);
37656 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
37659 @var{flags} is the bitwise @code{OR} of the following values:
37663 If the file does not exist it will be created. The host
37664 rules apply as far as file ownership and time stamps
37668 When used with @code{O_CREAT}, if the file already exists it is
37669 an error and open() fails.
37672 If the file already exists and the open mode allows
37673 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
37674 truncated to zero length.
37677 The file is opened in append mode.
37680 The file is opened for reading only.
37683 The file is opened for writing only.
37686 The file is opened for reading and writing.
37690 Other bits are silently ignored.
37694 @var{mode} is the bitwise @code{OR} of the following values:
37698 User has read permission.
37701 User has write permission.
37704 Group has read permission.
37707 Group has write permission.
37710 Others have read permission.
37713 Others have write permission.
37717 Other bits are silently ignored.
37720 @item Return value:
37721 @code{open} returns the new file descriptor or -1 if an error
37728 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
37731 @var{pathname} refers to a directory.
37734 The requested access is not allowed.
37737 @var{pathname} was too long.
37740 A directory component in @var{pathname} does not exist.
37743 @var{pathname} refers to a device, pipe, named pipe or socket.
37746 @var{pathname} refers to a file on a read-only filesystem and
37747 write access was requested.
37750 @var{pathname} is an invalid pointer value.
37753 No space on device to create the file.
37756 The process already has the maximum number of files open.
37759 The limit on the total number of files open on the system
37763 The call was interrupted by the user.
37769 @unnumberedsubsubsec close
37770 @cindex close, file-i/o system call
37779 @samp{Fclose,@var{fd}}
37781 @item Return value:
37782 @code{close} returns zero on success, or -1 if an error occurred.
37788 @var{fd} isn't a valid open file descriptor.
37791 The call was interrupted by the user.
37797 @unnumberedsubsubsec read
37798 @cindex read, file-i/o system call
37803 int read(int fd, void *buf, unsigned int count);
37807 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
37809 @item Return value:
37810 On success, the number of bytes read is returned.
37811 Zero indicates end of file. If count is zero, read
37812 returns zero as well. On error, -1 is returned.
37818 @var{fd} is not a valid file descriptor or is not open for
37822 @var{bufptr} is an invalid pointer value.
37825 The call was interrupted by the user.
37831 @unnumberedsubsubsec write
37832 @cindex write, file-i/o system call
37837 int write(int fd, const void *buf, unsigned int count);
37841 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
37843 @item Return value:
37844 On success, the number of bytes written are returned.
37845 Zero indicates nothing was written. On error, -1
37852 @var{fd} is not a valid file descriptor or is not open for
37856 @var{bufptr} is an invalid pointer value.
37859 An attempt was made to write a file that exceeds the
37860 host-specific maximum file size allowed.
37863 No space on device to write the data.
37866 The call was interrupted by the user.
37872 @unnumberedsubsubsec lseek
37873 @cindex lseek, file-i/o system call
37878 long lseek (int fd, long offset, int flag);
37882 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
37884 @var{flag} is one of:
37888 The offset is set to @var{offset} bytes.
37891 The offset is set to its current location plus @var{offset}
37895 The offset is set to the size of the file plus @var{offset}
37899 @item Return value:
37900 On success, the resulting unsigned offset in bytes from
37901 the beginning of the file is returned. Otherwise, a
37902 value of -1 is returned.
37908 @var{fd} is not a valid open file descriptor.
37911 @var{fd} is associated with the @value{GDBN} console.
37914 @var{flag} is not a proper value.
37917 The call was interrupted by the user.
37923 @unnumberedsubsubsec rename
37924 @cindex rename, file-i/o system call
37929 int rename(const char *oldpath, const char *newpath);
37933 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
37935 @item Return value:
37936 On success, zero is returned. On error, -1 is returned.
37942 @var{newpath} is an existing directory, but @var{oldpath} is not a
37946 @var{newpath} is a non-empty directory.
37949 @var{oldpath} or @var{newpath} is a directory that is in use by some
37953 An attempt was made to make a directory a subdirectory
37957 A component used as a directory in @var{oldpath} or new
37958 path is not a directory. Or @var{oldpath} is a directory
37959 and @var{newpath} exists but is not a directory.
37962 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
37965 No access to the file or the path of the file.
37969 @var{oldpath} or @var{newpath} was too long.
37972 A directory component in @var{oldpath} or @var{newpath} does not exist.
37975 The file is on a read-only filesystem.
37978 The device containing the file has no room for the new
37982 The call was interrupted by the user.
37988 @unnumberedsubsubsec unlink
37989 @cindex unlink, file-i/o system call
37994 int unlink(const char *pathname);
37998 @samp{Funlink,@var{pathnameptr}/@var{len}}
38000 @item Return value:
38001 On success, zero is returned. On error, -1 is returned.
38007 No access to the file or the path of the file.
38010 The system does not allow unlinking of directories.
38013 The file @var{pathname} cannot be unlinked because it's
38014 being used by another process.
38017 @var{pathnameptr} is an invalid pointer value.
38020 @var{pathname} was too long.
38023 A directory component in @var{pathname} does not exist.
38026 A component of the path is not a directory.
38029 The file is on a read-only filesystem.
38032 The call was interrupted by the user.
38038 @unnumberedsubsubsec stat/fstat
38039 @cindex fstat, file-i/o system call
38040 @cindex stat, file-i/o system call
38045 int stat(const char *pathname, struct stat *buf);
38046 int fstat(int fd, struct stat *buf);
38050 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
38051 @samp{Ffstat,@var{fd},@var{bufptr}}
38053 @item Return value:
38054 On success, zero is returned. On error, -1 is returned.
38060 @var{fd} is not a valid open file.
38063 A directory component in @var{pathname} does not exist or the
38064 path is an empty string.
38067 A component of the path is not a directory.
38070 @var{pathnameptr} is an invalid pointer value.
38073 No access to the file or the path of the file.
38076 @var{pathname} was too long.
38079 The call was interrupted by the user.
38085 @unnumberedsubsubsec gettimeofday
38086 @cindex gettimeofday, file-i/o system call
38091 int gettimeofday(struct timeval *tv, void *tz);
38095 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
38097 @item Return value:
38098 On success, 0 is returned, -1 otherwise.
38104 @var{tz} is a non-NULL pointer.
38107 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
38113 @unnumberedsubsubsec isatty
38114 @cindex isatty, file-i/o system call
38119 int isatty(int fd);
38123 @samp{Fisatty,@var{fd}}
38125 @item Return value:
38126 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
38132 The call was interrupted by the user.
38137 Note that the @code{isatty} call is treated as a special case: it returns
38138 1 to the target if the file descriptor is attached
38139 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
38140 would require implementing @code{ioctl} and would be more complex than
38145 @unnumberedsubsubsec system
38146 @cindex system, file-i/o system call
38151 int system(const char *command);
38155 @samp{Fsystem,@var{commandptr}/@var{len}}
38157 @item Return value:
38158 If @var{len} is zero, the return value indicates whether a shell is
38159 available. A zero return value indicates a shell is not available.
38160 For non-zero @var{len}, the value returned is -1 on error and the
38161 return status of the command otherwise. Only the exit status of the
38162 command is returned, which is extracted from the host's @code{system}
38163 return value by calling @code{WEXITSTATUS(retval)}. In case
38164 @file{/bin/sh} could not be executed, 127 is returned.
38170 The call was interrupted by the user.
38175 @value{GDBN} takes over the full task of calling the necessary host calls
38176 to perform the @code{system} call. The return value of @code{system} on
38177 the host is simplified before it's returned
38178 to the target. Any termination signal information from the child process
38179 is discarded, and the return value consists
38180 entirely of the exit status of the called command.
38182 Due to security concerns, the @code{system} call is by default refused
38183 by @value{GDBN}. The user has to allow this call explicitly with the
38184 @code{set remote system-call-allowed 1} command.
38187 @item set remote system-call-allowed
38188 @kindex set remote system-call-allowed
38189 Control whether to allow the @code{system} calls in the File I/O
38190 protocol for the remote target. The default is zero (disabled).
38192 @item show remote system-call-allowed
38193 @kindex show remote system-call-allowed
38194 Show whether the @code{system} calls are allowed in the File I/O
38198 @node Protocol-specific Representation of Datatypes
38199 @subsection Protocol-specific Representation of Datatypes
38200 @cindex protocol-specific representation of datatypes, in file-i/o protocol
38203 * Integral Datatypes::
38205 * Memory Transfer::
38210 @node Integral Datatypes
38211 @unnumberedsubsubsec Integral Datatypes
38212 @cindex integral datatypes, in file-i/o protocol
38214 The integral datatypes used in the system calls are @code{int},
38215 @code{unsigned int}, @code{long}, @code{unsigned long},
38216 @code{mode_t}, and @code{time_t}.
38218 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
38219 implemented as 32 bit values in this protocol.
38221 @code{long} and @code{unsigned long} are implemented as 64 bit types.
38223 @xref{Limits}, for corresponding MIN and MAX values (similar to those
38224 in @file{limits.h}) to allow range checking on host and target.
38226 @code{time_t} datatypes are defined as seconds since the Epoch.
38228 All integral datatypes transferred as part of a memory read or write of a
38229 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
38232 @node Pointer Values
38233 @unnumberedsubsubsec Pointer Values
38234 @cindex pointer values, in file-i/o protocol
38236 Pointers to target data are transmitted as they are. An exception
38237 is made for pointers to buffers for which the length isn't
38238 transmitted as part of the function call, namely strings. Strings
38239 are transmitted as a pointer/length pair, both as hex values, e.g.@:
38246 which is a pointer to data of length 18 bytes at position 0x1aaf.
38247 The length is defined as the full string length in bytes, including
38248 the trailing null byte. For example, the string @code{"hello world"}
38249 at address 0x123456 is transmitted as
38255 @node Memory Transfer
38256 @unnumberedsubsubsec Memory Transfer
38257 @cindex memory transfer, in file-i/o protocol
38259 Structured data which is transferred using a memory read or write (for
38260 example, a @code{struct stat}) is expected to be in a protocol-specific format
38261 with all scalar multibyte datatypes being big endian. Translation to
38262 this representation needs to be done both by the target before the @code{F}
38263 packet is sent, and by @value{GDBN} before
38264 it transfers memory to the target. Transferred pointers to structured
38265 data should point to the already-coerced data at any time.
38269 @unnumberedsubsubsec struct stat
38270 @cindex struct stat, in file-i/o protocol
38272 The buffer of type @code{struct stat} used by the target and @value{GDBN}
38273 is defined as follows:
38277 unsigned int st_dev; /* device */
38278 unsigned int st_ino; /* inode */
38279 mode_t st_mode; /* protection */
38280 unsigned int st_nlink; /* number of hard links */
38281 unsigned int st_uid; /* user ID of owner */
38282 unsigned int st_gid; /* group ID of owner */
38283 unsigned int st_rdev; /* device type (if inode device) */
38284 unsigned long st_size; /* total size, in bytes */
38285 unsigned long st_blksize; /* blocksize for filesystem I/O */
38286 unsigned long st_blocks; /* number of blocks allocated */
38287 time_t st_atime; /* time of last access */
38288 time_t st_mtime; /* time of last modification */
38289 time_t st_ctime; /* time of last change */
38293 The integral datatypes conform to the definitions given in the
38294 appropriate section (see @ref{Integral Datatypes}, for details) so this
38295 structure is of size 64 bytes.
38297 The values of several fields have a restricted meaning and/or
38303 A value of 0 represents a file, 1 the console.
38306 No valid meaning for the target. Transmitted unchanged.
38309 Valid mode bits are described in @ref{Constants}. Any other
38310 bits have currently no meaning for the target.
38315 No valid meaning for the target. Transmitted unchanged.
38320 These values have a host and file system dependent
38321 accuracy. Especially on Windows hosts, the file system may not
38322 support exact timing values.
38325 The target gets a @code{struct stat} of the above representation and is
38326 responsible for coercing it to the target representation before
38329 Note that due to size differences between the host, target, and protocol
38330 representations of @code{struct stat} members, these members could eventually
38331 get truncated on the target.
38333 @node struct timeval
38334 @unnumberedsubsubsec struct timeval
38335 @cindex struct timeval, in file-i/o protocol
38337 The buffer of type @code{struct timeval} used by the File-I/O protocol
38338 is defined as follows:
38342 time_t tv_sec; /* second */
38343 long tv_usec; /* microsecond */
38347 The integral datatypes conform to the definitions given in the
38348 appropriate section (see @ref{Integral Datatypes}, for details) so this
38349 structure is of size 8 bytes.
38352 @subsection Constants
38353 @cindex constants, in file-i/o protocol
38355 The following values are used for the constants inside of the
38356 protocol. @value{GDBN} and target are responsible for translating these
38357 values before and after the call as needed.
38368 @unnumberedsubsubsec Open Flags
38369 @cindex open flags, in file-i/o protocol
38371 All values are given in hexadecimal representation.
38383 @node mode_t Values
38384 @unnumberedsubsubsec mode_t Values
38385 @cindex mode_t values, in file-i/o protocol
38387 All values are given in octal representation.
38404 @unnumberedsubsubsec Errno Values
38405 @cindex errno values, in file-i/o protocol
38407 All values are given in decimal representation.
38432 @code{EUNKNOWN} is used as a fallback error value if a host system returns
38433 any error value not in the list of supported error numbers.
38436 @unnumberedsubsubsec Lseek Flags
38437 @cindex lseek flags, in file-i/o protocol
38446 @unnumberedsubsubsec Limits
38447 @cindex limits, in file-i/o protocol
38449 All values are given in decimal representation.
38452 INT_MIN -2147483648
38454 UINT_MAX 4294967295
38455 LONG_MIN -9223372036854775808
38456 LONG_MAX 9223372036854775807
38457 ULONG_MAX 18446744073709551615
38460 @node File-I/O Examples
38461 @subsection File-I/O Examples
38462 @cindex file-i/o examples
38464 Example sequence of a write call, file descriptor 3, buffer is at target
38465 address 0x1234, 6 bytes should be written:
38468 <- @code{Fwrite,3,1234,6}
38469 @emph{request memory read from target}
38472 @emph{return "6 bytes written"}
38476 Example sequence of a read call, file descriptor 3, buffer is at target
38477 address 0x1234, 6 bytes should be read:
38480 <- @code{Fread,3,1234,6}
38481 @emph{request memory write to target}
38482 -> @code{X1234,6:XXXXXX}
38483 @emph{return "6 bytes read"}
38487 Example sequence of a read call, call fails on the host due to invalid
38488 file descriptor (@code{EBADF}):
38491 <- @code{Fread,3,1234,6}
38495 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
38499 <- @code{Fread,3,1234,6}
38504 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
38508 <- @code{Fread,3,1234,6}
38509 -> @code{X1234,6:XXXXXX}
38513 @node Library List Format
38514 @section Library List Format
38515 @cindex library list format, remote protocol
38517 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
38518 same process as your application to manage libraries. In this case,
38519 @value{GDBN} can use the loader's symbol table and normal memory
38520 operations to maintain a list of shared libraries. On other
38521 platforms, the operating system manages loaded libraries.
38522 @value{GDBN} can not retrieve the list of currently loaded libraries
38523 through memory operations, so it uses the @samp{qXfer:libraries:read}
38524 packet (@pxref{qXfer library list read}) instead. The remote stub
38525 queries the target's operating system and reports which libraries
38528 The @samp{qXfer:libraries:read} packet returns an XML document which
38529 lists loaded libraries and their offsets. Each library has an
38530 associated name and one or more segment or section base addresses,
38531 which report where the library was loaded in memory.
38533 For the common case of libraries that are fully linked binaries, the
38534 library should have a list of segments. If the target supports
38535 dynamic linking of a relocatable object file, its library XML element
38536 should instead include a list of allocated sections. The segment or
38537 section bases are start addresses, not relocation offsets; they do not
38538 depend on the library's link-time base addresses.
38540 @value{GDBN} must be linked with the Expat library to support XML
38541 library lists. @xref{Expat}.
38543 A simple memory map, with one loaded library relocated by a single
38544 offset, looks like this:
38548 <library name="/lib/libc.so.6">
38549 <segment address="0x10000000"/>
38554 Another simple memory map, with one loaded library with three
38555 allocated sections (.text, .data, .bss), looks like this:
38559 <library name="sharedlib.o">
38560 <section address="0x10000000"/>
38561 <section address="0x20000000"/>
38562 <section address="0x30000000"/>
38567 The format of a library list is described by this DTD:
38570 <!-- library-list: Root element with versioning -->
38571 <!ELEMENT library-list (library)*>
38572 <!ATTLIST library-list version CDATA #FIXED "1.0">
38573 <!ELEMENT library (segment*, section*)>
38574 <!ATTLIST library name CDATA #REQUIRED>
38575 <!ELEMENT segment EMPTY>
38576 <!ATTLIST segment address CDATA #REQUIRED>
38577 <!ELEMENT section EMPTY>
38578 <!ATTLIST section address CDATA #REQUIRED>
38581 In addition, segments and section descriptors cannot be mixed within a
38582 single library element, and you must supply at least one segment or
38583 section for each library.
38585 @node Library List Format for SVR4 Targets
38586 @section Library List Format for SVR4 Targets
38587 @cindex library list format, remote protocol
38589 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
38590 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
38591 shared libraries. Still a special library list provided by this packet is
38592 more efficient for the @value{GDBN} remote protocol.
38594 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
38595 loaded libraries and their SVR4 linker parameters. For each library on SVR4
38596 target, the following parameters are reported:
38600 @code{name}, the absolute file name from the @code{l_name} field of
38601 @code{struct link_map}.
38603 @code{lm} with address of @code{struct link_map} used for TLS
38604 (Thread Local Storage) access.
38606 @code{l_addr}, the displacement as read from the field @code{l_addr} of
38607 @code{struct link_map}. For prelinked libraries this is not an absolute
38608 memory address. It is a displacement of absolute memory address against
38609 address the file was prelinked to during the library load.
38611 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
38614 Additionally the single @code{main-lm} attribute specifies address of
38615 @code{struct link_map} used for the main executable. This parameter is used
38616 for TLS access and its presence is optional.
38618 @value{GDBN} must be linked with the Expat library to support XML
38619 SVR4 library lists. @xref{Expat}.
38621 A simple memory map, with two loaded libraries (which do not use prelink),
38625 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
38626 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
38628 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
38630 </library-list-svr>
38633 The format of an SVR4 library list is described by this DTD:
38636 <!-- library-list-svr4: Root element with versioning -->
38637 <!ELEMENT library-list-svr4 (library)*>
38638 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
38639 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
38640 <!ELEMENT library EMPTY>
38641 <!ATTLIST library name CDATA #REQUIRED>
38642 <!ATTLIST library lm CDATA #REQUIRED>
38643 <!ATTLIST library l_addr CDATA #REQUIRED>
38644 <!ATTLIST library l_ld CDATA #REQUIRED>
38647 @node Memory Map Format
38648 @section Memory Map Format
38649 @cindex memory map format
38651 To be able to write into flash memory, @value{GDBN} needs to obtain a
38652 memory map from the target. This section describes the format of the
38655 The memory map is obtained using the @samp{qXfer:memory-map:read}
38656 (@pxref{qXfer memory map read}) packet and is an XML document that
38657 lists memory regions.
38659 @value{GDBN} must be linked with the Expat library to support XML
38660 memory maps. @xref{Expat}.
38662 The top-level structure of the document is shown below:
38665 <?xml version="1.0"?>
38666 <!DOCTYPE memory-map
38667 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
38668 "http://sourceware.org/gdb/gdb-memory-map.dtd">
38674 Each region can be either:
38679 A region of RAM starting at @var{addr} and extending for @var{length}
38683 <memory type="ram" start="@var{addr}" length="@var{length}"/>
38688 A region of read-only memory:
38691 <memory type="rom" start="@var{addr}" length="@var{length}"/>
38696 A region of flash memory, with erasure blocks @var{blocksize}
38700 <memory type="flash" start="@var{addr}" length="@var{length}">
38701 <property name="blocksize">@var{blocksize}</property>
38707 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
38708 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
38709 packets to write to addresses in such ranges.
38711 The formal DTD for memory map format is given below:
38714 <!-- ................................................... -->
38715 <!-- Memory Map XML DTD ................................ -->
38716 <!-- File: memory-map.dtd .............................. -->
38717 <!-- .................................... .............. -->
38718 <!-- memory-map.dtd -->
38719 <!-- memory-map: Root element with versioning -->
38720 <!ELEMENT memory-map (memory | property)>
38721 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
38722 <!ELEMENT memory (property)>
38723 <!-- memory: Specifies a memory region,
38724 and its type, or device. -->
38725 <!ATTLIST memory type CDATA #REQUIRED
38726 start CDATA #REQUIRED
38727 length CDATA #REQUIRED
38728 device CDATA #IMPLIED>
38729 <!-- property: Generic attribute tag -->
38730 <!ELEMENT property (#PCDATA | property)*>
38731 <!ATTLIST property name CDATA #REQUIRED>
38734 @node Thread List Format
38735 @section Thread List Format
38736 @cindex thread list format
38738 To efficiently update the list of threads and their attributes,
38739 @value{GDBN} issues the @samp{qXfer:threads:read} packet
38740 (@pxref{qXfer threads read}) and obtains the XML document with
38741 the following structure:
38744 <?xml version="1.0"?>
38746 <thread id="id" core="0">
38747 ... description ...
38752 Each @samp{thread} element must have the @samp{id} attribute that
38753 identifies the thread (@pxref{thread-id syntax}). The
38754 @samp{core} attribute, if present, specifies which processor core
38755 the thread was last executing on. The content of the of @samp{thread}
38756 element is interpreted as human-readable auxilliary information.
38758 @node Traceframe Info Format
38759 @section Traceframe Info Format
38760 @cindex traceframe info format
38762 To be able to know which objects in the inferior can be examined when
38763 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
38764 memory ranges, registers and trace state variables that have been
38765 collected in a traceframe.
38767 This list is obtained using the @samp{qXfer:traceframe-info:read}
38768 (@pxref{qXfer traceframe info read}) packet and is an XML document.
38770 @value{GDBN} must be linked with the Expat library to support XML
38771 traceframe info discovery. @xref{Expat}.
38773 The top-level structure of the document is shown below:
38776 <?xml version="1.0"?>
38777 <!DOCTYPE traceframe-info
38778 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
38779 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
38785 Each traceframe block can be either:
38790 A region of collected memory starting at @var{addr} and extending for
38791 @var{length} bytes from there:
38794 <memory start="@var{addr}" length="@var{length}"/>
38798 A block indicating trace state variable numbered @var{number} has been
38802 <tvar id="@var{number}"/>
38807 The formal DTD for the traceframe info format is given below:
38810 <!ELEMENT traceframe-info (memory | tvar)* >
38811 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
38813 <!ELEMENT memory EMPTY>
38814 <!ATTLIST memory start CDATA #REQUIRED
38815 length CDATA #REQUIRED>
38817 <!ATTLIST tvar id CDATA #REQUIRED>
38820 @node Branch Trace Format
38821 @section Branch Trace Format
38822 @cindex branch trace format
38824 In order to display the branch trace of an inferior thread,
38825 @value{GDBN} needs to obtain the list of branches. This list is
38826 represented as list of sequential code blocks that are connected via
38827 branches. The code in each block has been executed sequentially.
38829 This list is obtained using the @samp{qXfer:btrace:read}
38830 (@pxref{qXfer btrace read}) packet and is an XML document.
38832 @value{GDBN} must be linked with the Expat library to support XML
38833 traceframe info discovery. @xref{Expat}.
38835 The top-level structure of the document is shown below:
38838 <?xml version="1.0"?>
38840 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
38841 "http://sourceware.org/gdb/gdb-btrace.dtd">
38850 A block of sequentially executed instructions starting at @var{begin}
38851 and ending at @var{end}:
38854 <block begin="@var{begin}" end="@var{end}"/>
38859 The formal DTD for the branch trace format is given below:
38862 <!ELEMENT btrace (block)* >
38863 <!ATTLIST btrace version CDATA #FIXED "1.0">
38865 <!ELEMENT block EMPTY>
38866 <!ATTLIST block begin CDATA #REQUIRED
38867 end CDATA #REQUIRED>
38870 @include agentexpr.texi
38872 @node Target Descriptions
38873 @appendix Target Descriptions
38874 @cindex target descriptions
38876 One of the challenges of using @value{GDBN} to debug embedded systems
38877 is that there are so many minor variants of each processor
38878 architecture in use. It is common practice for vendors to start with
38879 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
38880 and then make changes to adapt it to a particular market niche. Some
38881 architectures have hundreds of variants, available from dozens of
38882 vendors. This leads to a number of problems:
38886 With so many different customized processors, it is difficult for
38887 the @value{GDBN} maintainers to keep up with the changes.
38889 Since individual variants may have short lifetimes or limited
38890 audiences, it may not be worthwhile to carry information about every
38891 variant in the @value{GDBN} source tree.
38893 When @value{GDBN} does support the architecture of the embedded system
38894 at hand, the task of finding the correct architecture name to give the
38895 @command{set architecture} command can be error-prone.
38898 To address these problems, the @value{GDBN} remote protocol allows a
38899 target system to not only identify itself to @value{GDBN}, but to
38900 actually describe its own features. This lets @value{GDBN} support
38901 processor variants it has never seen before --- to the extent that the
38902 descriptions are accurate, and that @value{GDBN} understands them.
38904 @value{GDBN} must be linked with the Expat library to support XML
38905 target descriptions. @xref{Expat}.
38908 * Retrieving Descriptions:: How descriptions are fetched from a target.
38909 * Target Description Format:: The contents of a target description.
38910 * Predefined Target Types:: Standard types available for target
38912 * Standard Target Features:: Features @value{GDBN} knows about.
38915 @node Retrieving Descriptions
38916 @section Retrieving Descriptions
38918 Target descriptions can be read from the target automatically, or
38919 specified by the user manually. The default behavior is to read the
38920 description from the target. @value{GDBN} retrieves it via the remote
38921 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
38922 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
38923 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
38924 XML document, of the form described in @ref{Target Description
38927 Alternatively, you can specify a file to read for the target description.
38928 If a file is set, the target will not be queried. The commands to
38929 specify a file are:
38932 @cindex set tdesc filename
38933 @item set tdesc filename @var{path}
38934 Read the target description from @var{path}.
38936 @cindex unset tdesc filename
38937 @item unset tdesc filename
38938 Do not read the XML target description from a file. @value{GDBN}
38939 will use the description supplied by the current target.
38941 @cindex show tdesc filename
38942 @item show tdesc filename
38943 Show the filename to read for a target description, if any.
38947 @node Target Description Format
38948 @section Target Description Format
38949 @cindex target descriptions, XML format
38951 A target description annex is an @uref{http://www.w3.org/XML/, XML}
38952 document which complies with the Document Type Definition provided in
38953 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
38954 means you can use generally available tools like @command{xmllint} to
38955 check that your feature descriptions are well-formed and valid.
38956 However, to help people unfamiliar with XML write descriptions for
38957 their targets, we also describe the grammar here.
38959 Target descriptions can identify the architecture of the remote target
38960 and (for some architectures) provide information about custom register
38961 sets. They can also identify the OS ABI of the remote target.
38962 @value{GDBN} can use this information to autoconfigure for your
38963 target, or to warn you if you connect to an unsupported target.
38965 Here is a simple target description:
38968 <target version="1.0">
38969 <architecture>i386:x86-64</architecture>
38974 This minimal description only says that the target uses
38975 the x86-64 architecture.
38977 A target description has the following overall form, with [ ] marking
38978 optional elements and @dots{} marking repeatable elements. The elements
38979 are explained further below.
38982 <?xml version="1.0"?>
38983 <!DOCTYPE target SYSTEM "gdb-target.dtd">
38984 <target version="1.0">
38985 @r{[}@var{architecture}@r{]}
38986 @r{[}@var{osabi}@r{]}
38987 @r{[}@var{compatible}@r{]}
38988 @r{[}@var{feature}@dots{}@r{]}
38993 The description is generally insensitive to whitespace and line
38994 breaks, under the usual common-sense rules. The XML version
38995 declaration and document type declaration can generally be omitted
38996 (@value{GDBN} does not require them), but specifying them may be
38997 useful for XML validation tools. The @samp{version} attribute for
38998 @samp{<target>} may also be omitted, but we recommend
38999 including it; if future versions of @value{GDBN} use an incompatible
39000 revision of @file{gdb-target.dtd}, they will detect and report
39001 the version mismatch.
39003 @subsection Inclusion
39004 @cindex target descriptions, inclusion
39007 @cindex <xi:include>
39010 It can sometimes be valuable to split a target description up into
39011 several different annexes, either for organizational purposes, or to
39012 share files between different possible target descriptions. You can
39013 divide a description into multiple files by replacing any element of
39014 the target description with an inclusion directive of the form:
39017 <xi:include href="@var{document}"/>
39021 When @value{GDBN} encounters an element of this form, it will retrieve
39022 the named XML @var{document}, and replace the inclusion directive with
39023 the contents of that document. If the current description was read
39024 using @samp{qXfer}, then so will be the included document;
39025 @var{document} will be interpreted as the name of an annex. If the
39026 current description was read from a file, @value{GDBN} will look for
39027 @var{document} as a file in the same directory where it found the
39028 original description.
39030 @subsection Architecture
39031 @cindex <architecture>
39033 An @samp{<architecture>} element has this form:
39036 <architecture>@var{arch}</architecture>
39039 @var{arch} is one of the architectures from the set accepted by
39040 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
39043 @cindex @code{<osabi>}
39045 This optional field was introduced in @value{GDBN} version 7.0.
39046 Previous versions of @value{GDBN} ignore it.
39048 An @samp{<osabi>} element has this form:
39051 <osabi>@var{abi-name}</osabi>
39054 @var{abi-name} is an OS ABI name from the same selection accepted by
39055 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
39057 @subsection Compatible Architecture
39058 @cindex @code{<compatible>}
39060 This optional field was introduced in @value{GDBN} version 7.0.
39061 Previous versions of @value{GDBN} ignore it.
39063 A @samp{<compatible>} element has this form:
39066 <compatible>@var{arch}</compatible>
39069 @var{arch} is one of the architectures from the set accepted by
39070 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
39072 A @samp{<compatible>} element is used to specify that the target
39073 is able to run binaries in some other than the main target architecture
39074 given by the @samp{<architecture>} element. For example, on the
39075 Cell Broadband Engine, the main architecture is @code{powerpc:common}
39076 or @code{powerpc:common64}, but the system is able to run binaries
39077 in the @code{spu} architecture as well. The way to describe this
39078 capability with @samp{<compatible>} is as follows:
39081 <architecture>powerpc:common</architecture>
39082 <compatible>spu</compatible>
39085 @subsection Features
39088 Each @samp{<feature>} describes some logical portion of the target
39089 system. Features are currently used to describe available CPU
39090 registers and the types of their contents. A @samp{<feature>} element
39094 <feature name="@var{name}">
39095 @r{[}@var{type}@dots{}@r{]}
39101 Each feature's name should be unique within the description. The name
39102 of a feature does not matter unless @value{GDBN} has some special
39103 knowledge of the contents of that feature; if it does, the feature
39104 should have its standard name. @xref{Standard Target Features}.
39108 Any register's value is a collection of bits which @value{GDBN} must
39109 interpret. The default interpretation is a two's complement integer,
39110 but other types can be requested by name in the register description.
39111 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
39112 Target Types}), and the description can define additional composite types.
39114 Each type element must have an @samp{id} attribute, which gives
39115 a unique (within the containing @samp{<feature>}) name to the type.
39116 Types must be defined before they are used.
39119 Some targets offer vector registers, which can be treated as arrays
39120 of scalar elements. These types are written as @samp{<vector>} elements,
39121 specifying the array element type, @var{type}, and the number of elements,
39125 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
39129 If a register's value is usefully viewed in multiple ways, define it
39130 with a union type containing the useful representations. The
39131 @samp{<union>} element contains one or more @samp{<field>} elements,
39132 each of which has a @var{name} and a @var{type}:
39135 <union id="@var{id}">
39136 <field name="@var{name}" type="@var{type}"/>
39142 If a register's value is composed from several separate values, define
39143 it with a structure type. There are two forms of the @samp{<struct>}
39144 element; a @samp{<struct>} element must either contain only bitfields
39145 or contain no bitfields. If the structure contains only bitfields,
39146 its total size in bytes must be specified, each bitfield must have an
39147 explicit start and end, and bitfields are automatically assigned an
39148 integer type. The field's @var{start} should be less than or
39149 equal to its @var{end}, and zero represents the least significant bit.
39152 <struct id="@var{id}" size="@var{size}">
39153 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
39158 If the structure contains no bitfields, then each field has an
39159 explicit type, and no implicit padding is added.
39162 <struct id="@var{id}">
39163 <field name="@var{name}" type="@var{type}"/>
39169 If a register's value is a series of single-bit flags, define it with
39170 a flags type. The @samp{<flags>} element has an explicit @var{size}
39171 and contains one or more @samp{<field>} elements. Each field has a
39172 @var{name}, a @var{start}, and an @var{end}. Only single-bit flags
39176 <flags id="@var{id}" size="@var{size}">
39177 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
39182 @subsection Registers
39185 Each register is represented as an element with this form:
39188 <reg name="@var{name}"
39189 bitsize="@var{size}"
39190 @r{[}regnum="@var{num}"@r{]}
39191 @r{[}save-restore="@var{save-restore}"@r{]}
39192 @r{[}type="@var{type}"@r{]}
39193 @r{[}group="@var{group}"@r{]}/>
39197 The components are as follows:
39202 The register's name; it must be unique within the target description.
39205 The register's size, in bits.
39208 The register's number. If omitted, a register's number is one greater
39209 than that of the previous register (either in the current feature or in
39210 a preceding feature); the first register in the target description
39211 defaults to zero. This register number is used to read or write
39212 the register; e.g.@: it is used in the remote @code{p} and @code{P}
39213 packets, and registers appear in the @code{g} and @code{G} packets
39214 in order of increasing register number.
39217 Whether the register should be preserved across inferior function
39218 calls; this must be either @code{yes} or @code{no}. The default is
39219 @code{yes}, which is appropriate for most registers except for
39220 some system control registers; this is not related to the target's
39224 The type of the register. It may be a predefined type, a type
39225 defined in the current feature, or one of the special types @code{int}
39226 and @code{float}. @code{int} is an integer type of the correct size
39227 for @var{bitsize}, and @code{float} is a floating point type (in the
39228 architecture's normal floating point format) of the correct size for
39229 @var{bitsize}. The default is @code{int}.
39232 The register group to which this register belongs. It must
39233 be either @code{general}, @code{float}, or @code{vector}. If no
39234 @var{group} is specified, @value{GDBN} will not display the register
39235 in @code{info registers}.
39239 @node Predefined Target Types
39240 @section Predefined Target Types
39241 @cindex target descriptions, predefined types
39243 Type definitions in the self-description can build up composite types
39244 from basic building blocks, but can not define fundamental types. Instead,
39245 standard identifiers are provided by @value{GDBN} for the fundamental
39246 types. The currently supported types are:
39255 Signed integer types holding the specified number of bits.
39262 Unsigned integer types holding the specified number of bits.
39266 Pointers to unspecified code and data. The program counter and
39267 any dedicated return address register may be marked as code
39268 pointers; printing a code pointer converts it into a symbolic
39269 address. The stack pointer and any dedicated address registers
39270 may be marked as data pointers.
39273 Single precision IEEE floating point.
39276 Double precision IEEE floating point.
39279 The 12-byte extended precision format used by ARM FPA registers.
39282 The 10-byte extended precision format used by x87 registers.
39285 32bit @sc{eflags} register used by x86.
39288 32bit @sc{mxcsr} register used by x86.
39292 @node Standard Target Features
39293 @section Standard Target Features
39294 @cindex target descriptions, standard features
39296 A target description must contain either no registers or all the
39297 target's registers. If the description contains no registers, then
39298 @value{GDBN} will assume a default register layout, selected based on
39299 the architecture. If the description contains any registers, the
39300 default layout will not be used; the standard registers must be
39301 described in the target description, in such a way that @value{GDBN}
39302 can recognize them.
39304 This is accomplished by giving specific names to feature elements
39305 which contain standard registers. @value{GDBN} will look for features
39306 with those names and verify that they contain the expected registers;
39307 if any known feature is missing required registers, or if any required
39308 feature is missing, @value{GDBN} will reject the target
39309 description. You can add additional registers to any of the
39310 standard features --- @value{GDBN} will display them just as if
39311 they were added to an unrecognized feature.
39313 This section lists the known features and their expected contents.
39314 Sample XML documents for these features are included in the
39315 @value{GDBN} source tree, in the directory @file{gdb/features}.
39317 Names recognized by @value{GDBN} should include the name of the
39318 company or organization which selected the name, and the overall
39319 architecture to which the feature applies; so e.g.@: the feature
39320 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
39322 The names of registers are not case sensitive for the purpose
39323 of recognizing standard features, but @value{GDBN} will only display
39324 registers using the capitalization used in the description.
39327 * AArch64 Features::
39330 * MicroBlaze Features::
39333 * Nios II Features::
39334 * PowerPC Features::
39335 * S/390 and System z Features::
39340 @node AArch64 Features
39341 @subsection AArch64 Features
39342 @cindex target descriptions, AArch64 features
39344 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
39345 targets. It should contain registers @samp{x0} through @samp{x30},
39346 @samp{sp}, @samp{pc}, and @samp{cpsr}.
39348 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
39349 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
39353 @subsection ARM Features
39354 @cindex target descriptions, ARM features
39356 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
39358 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
39359 @samp{lr}, @samp{pc}, and @samp{cpsr}.
39361 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
39362 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
39363 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
39366 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
39367 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
39369 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
39370 it should contain at least registers @samp{wR0} through @samp{wR15} and
39371 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
39372 @samp{wCSSF}, and @samp{wCASF} registers are optional.
39374 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
39375 should contain at least registers @samp{d0} through @samp{d15}. If
39376 they are present, @samp{d16} through @samp{d31} should also be included.
39377 @value{GDBN} will synthesize the single-precision registers from
39378 halves of the double-precision registers.
39380 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
39381 need to contain registers; it instructs @value{GDBN} to display the
39382 VFP double-precision registers as vectors and to synthesize the
39383 quad-precision registers from pairs of double-precision registers.
39384 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
39385 be present and include 32 double-precision registers.
39387 @node i386 Features
39388 @subsection i386 Features
39389 @cindex target descriptions, i386 features
39391 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
39392 targets. It should describe the following registers:
39396 @samp{eax} through @samp{edi} plus @samp{eip} for i386
39398 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
39400 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
39401 @samp{fs}, @samp{gs}
39403 @samp{st0} through @samp{st7}
39405 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
39406 @samp{foseg}, @samp{fooff} and @samp{fop}
39409 The register sets may be different, depending on the target.
39411 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
39412 describe registers:
39416 @samp{xmm0} through @samp{xmm7} for i386
39418 @samp{xmm0} through @samp{xmm15} for amd64
39423 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
39424 @samp{org.gnu.gdb.i386.sse} feature. It should
39425 describe the upper 128 bits of @sc{ymm} registers:
39429 @samp{ymm0h} through @samp{ymm7h} for i386
39431 @samp{ymm0h} through @samp{ymm15h} for amd64
39434 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel(R)
39435 Memory Protection Extension (MPX). It should describe the following registers:
39439 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
39441 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
39444 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
39445 describe a single register, @samp{orig_eax}.
39447 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
39448 @samp{org.gnu.gdb.i386.avx} feature. It should
39449 describe additional @sc{xmm} registers:
39453 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
39456 It should describe the upper 128 bits of additional @sc{ymm} registers:
39460 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
39464 describe the upper 256 bits of @sc{zmm} registers:
39468 @samp{zmm0h} through @samp{zmm7h} for i386.
39470 @samp{zmm0h} through @samp{zmm15h} for amd64.
39474 describe the additional @sc{zmm} registers:
39478 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
39481 @node MicroBlaze Features
39482 @subsection MicroBlaze Features
39483 @cindex target descriptions, MicroBlaze features
39485 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
39486 targets. It should contain registers @samp{r0} through @samp{r31},
39487 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
39488 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
39489 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
39491 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
39492 If present, it should contain registers @samp{rshr} and @samp{rslr}
39494 @node MIPS Features
39495 @subsection @acronym{MIPS} Features
39496 @cindex target descriptions, @acronym{MIPS} features
39498 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
39499 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
39500 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
39503 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
39504 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
39505 registers. They may be 32-bit or 64-bit depending on the target.
39507 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
39508 it may be optional in a future version of @value{GDBN}. It should
39509 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
39510 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
39512 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
39513 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
39514 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
39515 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
39517 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
39518 contain a single register, @samp{restart}, which is used by the
39519 Linux kernel to control restartable syscalls.
39521 @node M68K Features
39522 @subsection M68K Features
39523 @cindex target descriptions, M68K features
39526 @item @samp{org.gnu.gdb.m68k.core}
39527 @itemx @samp{org.gnu.gdb.coldfire.core}
39528 @itemx @samp{org.gnu.gdb.fido.core}
39529 One of those features must be always present.
39530 The feature that is present determines which flavor of m68k is
39531 used. The feature that is present should contain registers
39532 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
39533 @samp{sp}, @samp{ps} and @samp{pc}.
39535 @item @samp{org.gnu.gdb.coldfire.fp}
39536 This feature is optional. If present, it should contain registers
39537 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
39541 @node Nios II Features
39542 @subsection Nios II Features
39543 @cindex target descriptions, Nios II features
39545 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
39546 targets. It should contain the 32 core registers (@samp{zero},
39547 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
39548 @samp{pc}, and the 16 control registers (@samp{status} through
39551 @node PowerPC Features
39552 @subsection PowerPC Features
39553 @cindex target descriptions, PowerPC features
39555 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
39556 targets. It should contain registers @samp{r0} through @samp{r31},
39557 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
39558 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
39560 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
39561 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
39563 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
39564 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
39567 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
39568 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
39569 will combine these registers with the floating point registers
39570 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
39571 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
39572 through @samp{vs63}, the set of vector registers for POWER7.
39574 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
39575 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
39576 @samp{spefscr}. SPE targets should provide 32-bit registers in
39577 @samp{org.gnu.gdb.power.core} and provide the upper halves in
39578 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
39579 these to present registers @samp{ev0} through @samp{ev31} to the
39582 @node S/390 and System z Features
39583 @subsection S/390 and System z Features
39584 @cindex target descriptions, S/390 features
39585 @cindex target descriptions, System z features
39587 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
39588 System z targets. It should contain the PSW and the 16 general
39589 registers. In particular, System z targets should provide the 64-bit
39590 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
39591 S/390 targets should provide the 32-bit versions of these registers.
39592 A System z target that runs in 31-bit addressing mode should provide
39593 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
39594 register's upper halves @samp{r0h} through @samp{r15h}, and their
39595 lower halves @samp{r0l} through @samp{r15l}.
39597 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
39598 contain the 64-bit registers @samp{f0} through @samp{f15}, and
39601 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
39602 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
39604 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
39605 contain the register @samp{orig_r2}, which is 64-bit wide on System z
39606 targets and 32-bit otherwise. In addition, the feature may contain
39607 the @samp{last_break} register, whose width depends on the addressing
39608 mode, as well as the @samp{system_call} register, which is always
39611 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
39612 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
39613 @samp{atia}, and @samp{tr0} through @samp{tr15}.
39615 @node TIC6x Features
39616 @subsection TMS320C6x Features
39617 @cindex target descriptions, TIC6x features
39618 @cindex target descriptions, TMS320C6x features
39619 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
39620 targets. It should contain registers @samp{A0} through @samp{A15},
39621 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
39623 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
39624 contain registers @samp{A16} through @samp{A31} and @samp{B16}
39625 through @samp{B31}.
39627 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
39628 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
39630 @node Operating System Information
39631 @appendix Operating System Information
39632 @cindex operating system information
39638 Users of @value{GDBN} often wish to obtain information about the state of
39639 the operating system running on the target---for example the list of
39640 processes, or the list of open files. This section describes the
39641 mechanism that makes it possible. This mechanism is similar to the
39642 target features mechanism (@pxref{Target Descriptions}), but focuses
39643 on a different aspect of target.
39645 Operating system information is retrived from the target via the
39646 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
39647 read}). The object name in the request should be @samp{osdata}, and
39648 the @var{annex} identifies the data to be fetched.
39651 @appendixsection Process list
39652 @cindex operating system information, process list
39654 When requesting the process list, the @var{annex} field in the
39655 @samp{qXfer} request should be @samp{processes}. The returned data is
39656 an XML document. The formal syntax of this document is defined in
39657 @file{gdb/features/osdata.dtd}.
39659 An example document is:
39662 <?xml version="1.0"?>
39663 <!DOCTYPE target SYSTEM "osdata.dtd">
39664 <osdata type="processes">
39666 <column name="pid">1</column>
39667 <column name="user">root</column>
39668 <column name="command">/sbin/init</column>
39669 <column name="cores">1,2,3</column>
39674 Each item should include a column whose name is @samp{pid}. The value
39675 of that column should identify the process on the target. The
39676 @samp{user} and @samp{command} columns are optional, and will be
39677 displayed by @value{GDBN}. The @samp{cores} column, if present,
39678 should contain a comma-separated list of cores that this process
39679 is running on. Target may provide additional columns,
39680 which @value{GDBN} currently ignores.
39682 @node Trace File Format
39683 @appendix Trace File Format
39684 @cindex trace file format
39686 The trace file comes in three parts: a header, a textual description
39687 section, and a trace frame section with binary data.
39689 The header has the form @code{\x7fTRACE0\n}. The first byte is
39690 @code{0x7f} so as to indicate that the file contains binary data,
39691 while the @code{0} is a version number that may have different values
39694 The description section consists of multiple lines of @sc{ascii} text
39695 separated by newline characters (@code{0xa}). The lines may include a
39696 variety of optional descriptive or context-setting information, such
39697 as tracepoint definitions or register set size. @value{GDBN} will
39698 ignore any line that it does not recognize. An empty line marks the end
39701 @c FIXME add some specific types of data
39703 The trace frame section consists of a number of consecutive frames.
39704 Each frame begins with a two-byte tracepoint number, followed by a
39705 four-byte size giving the amount of data in the frame. The data in
39706 the frame consists of a number of blocks, each introduced by a
39707 character indicating its type (at least register, memory, and trace
39708 state variable). The data in this section is raw binary, not a
39709 hexadecimal or other encoding; its endianness matches the target's
39712 @c FIXME bi-arch may require endianness/arch info in description section
39715 @item R @var{bytes}
39716 Register block. The number and ordering of bytes matches that of a
39717 @code{g} packet in the remote protocol. Note that these are the
39718 actual bytes, in target order and @value{GDBN} register order, not a
39719 hexadecimal encoding.
39721 @item M @var{address} @var{length} @var{bytes}...
39722 Memory block. This is a contiguous block of memory, at the 8-byte
39723 address @var{address}, with a 2-byte length @var{length}, followed by
39724 @var{length} bytes.
39726 @item V @var{number} @var{value}
39727 Trace state variable block. This records the 8-byte signed value
39728 @var{value} of trace state variable numbered @var{number}.
39732 Future enhancements of the trace file format may include additional types
39735 @node Index Section Format
39736 @appendix @code{.gdb_index} section format
39737 @cindex .gdb_index section format
39738 @cindex index section format
39740 This section documents the index section that is created by @code{save
39741 gdb-index} (@pxref{Index Files}). The index section is
39742 DWARF-specific; some knowledge of DWARF is assumed in this
39745 The mapped index file format is designed to be directly
39746 @code{mmap}able on any architecture. In most cases, a datum is
39747 represented using a little-endian 32-bit integer value, called an
39748 @code{offset_type}. Big endian machines must byte-swap the values
39749 before using them. Exceptions to this rule are noted. The data is
39750 laid out such that alignment is always respected.
39752 A mapped index consists of several areas, laid out in order.
39756 The file header. This is a sequence of values, of @code{offset_type}
39757 unless otherwise noted:
39761 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
39762 Version 4 uses a different hashing function from versions 5 and 6.
39763 Version 6 includes symbols for inlined functions, whereas versions 4
39764 and 5 do not. Version 7 adds attributes to the CU indices in the
39765 symbol table. Version 8 specifies that symbols from DWARF type units
39766 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
39767 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
39769 @value{GDBN} will only read version 4, 5, or 6 indices
39770 by specifying @code{set use-deprecated-index-sections on}.
39771 GDB has a workaround for potentially broken version 7 indices so it is
39772 currently not flagged as deprecated.
39775 The offset, from the start of the file, of the CU list.
39778 The offset, from the start of the file, of the types CU list. Note
39779 that this area can be empty, in which case this offset will be equal
39780 to the next offset.
39783 The offset, from the start of the file, of the address area.
39786 The offset, from the start of the file, of the symbol table.
39789 The offset, from the start of the file, of the constant pool.
39793 The CU list. This is a sequence of pairs of 64-bit little-endian
39794 values, sorted by the CU offset. The first element in each pair is
39795 the offset of a CU in the @code{.debug_info} section. The second
39796 element in each pair is the length of that CU. References to a CU
39797 elsewhere in the map are done using a CU index, which is just the
39798 0-based index into this table. Note that if there are type CUs, then
39799 conceptually CUs and type CUs form a single list for the purposes of
39803 The types CU list. This is a sequence of triplets of 64-bit
39804 little-endian values. In a triplet, the first value is the CU offset,
39805 the second value is the type offset in the CU, and the third value is
39806 the type signature. The types CU list is not sorted.
39809 The address area. The address area consists of a sequence of address
39810 entries. Each address entry has three elements:
39814 The low address. This is a 64-bit little-endian value.
39817 The high address. This is a 64-bit little-endian value. Like
39818 @code{DW_AT_high_pc}, the value is one byte beyond the end.
39821 The CU index. This is an @code{offset_type} value.
39825 The symbol table. This is an open-addressed hash table. The size of
39826 the hash table is always a power of 2.
39828 Each slot in the hash table consists of a pair of @code{offset_type}
39829 values. The first value is the offset of the symbol's name in the
39830 constant pool. The second value is the offset of the CU vector in the
39833 If both values are 0, then this slot in the hash table is empty. This
39834 is ok because while 0 is a valid constant pool index, it cannot be a
39835 valid index for both a string and a CU vector.
39837 The hash value for a table entry is computed by applying an
39838 iterative hash function to the symbol's name. Starting with an
39839 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
39840 the string is incorporated into the hash using the formula depending on the
39845 The formula is @code{r = r * 67 + c - 113}.
39847 @item Versions 5 to 7
39848 The formula is @code{r = r * 67 + tolower (c) - 113}.
39851 The terminating @samp{\0} is not incorporated into the hash.
39853 The step size used in the hash table is computed via
39854 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
39855 value, and @samp{size} is the size of the hash table. The step size
39856 is used to find the next candidate slot when handling a hash
39859 The names of C@t{++} symbols in the hash table are canonicalized. We
39860 don't currently have a simple description of the canonicalization
39861 algorithm; if you intend to create new index sections, you must read
39865 The constant pool. This is simply a bunch of bytes. It is organized
39866 so that alignment is correct: CU vectors are stored first, followed by
39869 A CU vector in the constant pool is a sequence of @code{offset_type}
39870 values. The first value is the number of CU indices in the vector.
39871 Each subsequent value is the index and symbol attributes of a CU in
39872 the CU list. This element in the hash table is used to indicate which
39873 CUs define the symbol and how the symbol is used.
39874 See below for the format of each CU index+attributes entry.
39876 A string in the constant pool is zero-terminated.
39879 Attributes were added to CU index values in @code{.gdb_index} version 7.
39880 If a symbol has multiple uses within a CU then there is one
39881 CU index+attributes value for each use.
39883 The format of each CU index+attributes entry is as follows
39889 This is the index of the CU in the CU list.
39891 These bits are reserved for future purposes and must be zero.
39893 The kind of the symbol in the CU.
39897 This value is reserved and should not be used.
39898 By reserving zero the full @code{offset_type} value is backwards compatible
39899 with previous versions of the index.
39901 The symbol is a type.
39903 The symbol is a variable or an enum value.
39905 The symbol is a function.
39907 Any other kind of symbol.
39909 These values are reserved.
39913 This bit is zero if the value is global and one if it is static.
39915 The determination of whether a symbol is global or static is complicated.
39916 The authorative reference is the file @file{dwarf2read.c} in
39917 @value{GDBN} sources.
39921 This pseudo-code describes the computation of a symbol's kind and
39922 global/static attributes in the index.
39925 is_external = get_attribute (die, DW_AT_external);
39926 language = get_attribute (cu_die, DW_AT_language);
39929 case DW_TAG_typedef:
39930 case DW_TAG_base_type:
39931 case DW_TAG_subrange_type:
39935 case DW_TAG_enumerator:
39937 is_static = (language != CPLUS && language != JAVA);
39939 case DW_TAG_subprogram:
39941 is_static = ! (is_external || language == ADA);
39943 case DW_TAG_constant:
39945 is_static = ! is_external;
39947 case DW_TAG_variable:
39949 is_static = ! is_external;
39951 case DW_TAG_namespace:
39955 case DW_TAG_class_type:
39956 case DW_TAG_interface_type:
39957 case DW_TAG_structure_type:
39958 case DW_TAG_union_type:
39959 case DW_TAG_enumeration_type:
39961 is_static = (language != CPLUS && language != JAVA);
39969 @appendix Manual pages
39973 * gdb man:: The GNU Debugger man page
39974 * gdbserver man:: Remote Server for the GNU Debugger man page
39975 * gcore man:: Generate a core file of a running program
39976 * gdbinit man:: gdbinit scripts
39982 @c man title gdb The GNU Debugger
39984 @c man begin SYNOPSIS gdb
39985 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
39986 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
39987 [@option{-b}@w{ }@var{bps}]
39988 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
39989 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
39990 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
39991 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
39992 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
39995 @c man begin DESCRIPTION gdb
39996 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
39997 going on ``inside'' another program while it executes -- or what another
39998 program was doing at the moment it crashed.
40000 @value{GDBN} can do four main kinds of things (plus other things in support of
40001 these) to help you catch bugs in the act:
40005 Start your program, specifying anything that might affect its behavior.
40008 Make your program stop on specified conditions.
40011 Examine what has happened, when your program has stopped.
40014 Change things in your program, so you can experiment with correcting the
40015 effects of one bug and go on to learn about another.
40018 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
40021 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
40022 commands from the terminal until you tell it to exit with the @value{GDBN}
40023 command @code{quit}. You can get online help from @value{GDBN} itself
40024 by using the command @code{help}.
40026 You can run @code{gdb} with no arguments or options; but the most
40027 usual way to start @value{GDBN} is with one argument or two, specifying an
40028 executable program as the argument:
40034 You can also start with both an executable program and a core file specified:
40040 You can, instead, specify a process ID as a second argument, if you want
40041 to debug a running process:
40049 would attach @value{GDBN} to process @code{1234} (unless you also have a file
40050 named @file{1234}; @value{GDBN} does check for a core file first).
40051 With option @option{-p} you can omit the @var{program} filename.
40053 Here are some of the most frequently needed @value{GDBN} commands:
40055 @c pod2man highlights the right hand side of the @item lines.
40057 @item break [@var{file}:]@var{functiop}
40058 Set a breakpoint at @var{function} (in @var{file}).
40060 @item run [@var{arglist}]
40061 Start your program (with @var{arglist}, if specified).
40064 Backtrace: display the program stack.
40066 @item print @var{expr}
40067 Display the value of an expression.
40070 Continue running your program (after stopping, e.g. at a breakpoint).
40073 Execute next program line (after stopping); step @emph{over} any
40074 function calls in the line.
40076 @item edit [@var{file}:]@var{function}
40077 look at the program line where it is presently stopped.
40079 @item list [@var{file}:]@var{function}
40080 type the text of the program in the vicinity of where it is presently stopped.
40083 Execute next program line (after stopping); step @emph{into} any
40084 function calls in the line.
40086 @item help [@var{name}]
40087 Show information about @value{GDBN} command @var{name}, or general information
40088 about using @value{GDBN}.
40091 Exit from @value{GDBN}.
40095 For full details on @value{GDBN},
40096 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
40097 by Richard M. Stallman and Roland H. Pesch. The same text is available online
40098 as the @code{gdb} entry in the @code{info} program.
40102 @c man begin OPTIONS gdb
40103 Any arguments other than options specify an executable
40104 file and core file (or process ID); that is, the first argument
40105 encountered with no
40106 associated option flag is equivalent to a @option{-se} option, and the second,
40107 if any, is equivalent to a @option{-c} option if it's the name of a file.
40109 both long and short forms; both are shown here. The long forms are also
40110 recognized if you truncate them, so long as enough of the option is
40111 present to be unambiguous. (If you prefer, you can flag option
40112 arguments with @option{+} rather than @option{-}, though we illustrate the
40113 more usual convention.)
40115 All the options and command line arguments you give are processed
40116 in sequential order. The order makes a difference when the @option{-x}
40122 List all options, with brief explanations.
40124 @item -symbols=@var{file}
40125 @itemx -s @var{file}
40126 Read symbol table from file @var{file}.
40129 Enable writing into executable and core files.
40131 @item -exec=@var{file}
40132 @itemx -e @var{file}
40133 Use file @var{file} as the executable file to execute when
40134 appropriate, and for examining pure data in conjunction with a core
40137 @item -se=@var{file}
40138 Read symbol table from file @var{file} and use it as the executable
40141 @item -core=@var{file}
40142 @itemx -c @var{file}
40143 Use file @var{file} as a core dump to examine.
40145 @item -command=@var{file}
40146 @itemx -x @var{file}
40147 Execute @value{GDBN} commands from file @var{file}.
40149 @item -ex @var{command}
40150 Execute given @value{GDBN} @var{command}.
40152 @item -directory=@var{directory}
40153 @itemx -d @var{directory}
40154 Add @var{directory} to the path to search for source files.
40157 Do not execute commands from @file{~/.gdbinit}.
40161 Do not execute commands from any @file{.gdbinit} initialization files.
40165 ``Quiet''. Do not print the introductory and copyright messages. These
40166 messages are also suppressed in batch mode.
40169 Run in batch mode. Exit with status @code{0} after processing all the command
40170 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
40171 Exit with nonzero status if an error occurs in executing the @value{GDBN}
40172 commands in the command files.
40174 Batch mode may be useful for running @value{GDBN} as a filter, for example to
40175 download and run a program on another computer; in order to make this
40176 more useful, the message
40179 Program exited normally.
40183 (which is ordinarily issued whenever a program running under @value{GDBN} control
40184 terminates) is not issued when running in batch mode.
40186 @item -cd=@var{directory}
40187 Run @value{GDBN} using @var{directory} as its working directory,
40188 instead of the current directory.
40192 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
40193 @value{GDBN} to output the full file name and line number in a standard,
40194 recognizable fashion each time a stack frame is displayed (which
40195 includes each time the program stops). This recognizable format looks
40196 like two @samp{\032} characters, followed by the file name, line number
40197 and character position separated by colons, and a newline. The
40198 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
40199 characters as a signal to display the source code for the frame.
40202 Set the line speed (baud rate or bits per second) of any serial
40203 interface used by @value{GDBN} for remote debugging.
40205 @item -tty=@var{device}
40206 Run using @var{device} for your program's standard input and output.
40210 @c man begin SEEALSO gdb
40212 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
40213 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
40214 documentation are properly installed at your site, the command
40221 should give you access to the complete manual.
40223 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
40224 Richard M. Stallman and Roland H. Pesch, July 1991.
40228 @node gdbserver man
40229 @heading gdbserver man
40231 @c man title gdbserver Remote Server for the GNU Debugger
40233 @c man begin SYNOPSIS gdbserver
40234 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
40236 gdbserver --attach @var{comm} @var{pid}
40238 gdbserver --multi @var{comm}
40242 @c man begin DESCRIPTION gdbserver
40243 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
40244 than the one which is running the program being debugged.
40247 @subheading Usage (server (target) side)
40250 Usage (server (target) side):
40253 First, you need to have a copy of the program you want to debug put onto
40254 the target system. The program can be stripped to save space if needed, as
40255 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
40256 the @value{GDBN} running on the host system.
40258 To use the server, you log on to the target system, and run the @command{gdbserver}
40259 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
40260 your program, and (c) its arguments. The general syntax is:
40263 target> gdbserver @var{comm} @var{program} [@var{args} ...]
40266 For example, using a serial port, you might say:
40270 @c @file would wrap it as F</dev/com1>.
40271 target> gdbserver /dev/com1 emacs foo.txt
40274 target> gdbserver @file{/dev/com1} emacs foo.txt
40278 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
40279 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
40280 waits patiently for the host @value{GDBN} to communicate with it.
40282 To use a TCP connection, you could say:
40285 target> gdbserver host:2345 emacs foo.txt
40288 This says pretty much the same thing as the last example, except that we are
40289 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
40290 that we are expecting to see a TCP connection from @code{host} to local TCP port
40291 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
40292 want for the port number as long as it does not conflict with any existing TCP
40293 ports on the target system. This same port number must be used in the host
40294 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
40295 you chose a port number that conflicts with another service, @command{gdbserver} will
40296 print an error message and exit.
40298 @command{gdbserver} can also attach to running programs.
40299 This is accomplished via the @option{--attach} argument. The syntax is:
40302 target> gdbserver --attach @var{comm} @var{pid}
40305 @var{pid} is the process ID of a currently running process. It isn't
40306 necessary to point @command{gdbserver} at a binary for the running process.
40308 To start @code{gdbserver} without supplying an initial command to run
40309 or process ID to attach, use the @option{--multi} command line option.
40310 In such case you should connect using @kbd{target extended-remote} to start
40311 the program you want to debug.
40314 target> gdbserver --multi @var{comm}
40318 @subheading Usage (host side)
40324 You need an unstripped copy of the target program on your host system, since
40325 @value{GDBN} needs to examine it's symbol tables and such. Start up @value{GDBN} as you normally
40326 would, with the target program as the first argument. (You may need to use the
40327 @option{--baud} option if the serial line is running at anything except 9600 baud.)
40328 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
40329 new command you need to know about is @code{target remote}
40330 (or @code{target extended-remote}). Its argument is either
40331 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
40332 descriptor. For example:
40336 @c @file would wrap it as F</dev/ttyb>.
40337 (gdb) target remote /dev/ttyb
40340 (gdb) target remote @file{/dev/ttyb}
40345 communicates with the server via serial line @file{/dev/ttyb}, and:
40348 (gdb) target remote the-target:2345
40352 communicates via a TCP connection to port 2345 on host `the-target', where
40353 you previously started up @command{gdbserver} with the same port number. Note that for
40354 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
40355 command, otherwise you may get an error that looks something like
40356 `Connection refused'.
40358 @command{gdbserver} can also debug multiple inferiors at once,
40361 the @value{GDBN} manual in node @code{Inferiors and Programs}
40362 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
40365 @ref{Inferiors and Programs}.
40367 In such case use the @code{extended-remote} @value{GDBN} command variant:
40370 (gdb) target extended-remote the-target:2345
40373 The @command{gdbserver} option @option{--multi} may or may not be used in such
40377 @c man begin OPTIONS gdbserver
40378 There are three different modes for invoking @command{gdbserver}:
40383 Debug a specific program specified by its program name:
40386 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
40389 The @var{comm} parameter specifies how should the server communicate
40390 with @value{GDBN}; it is either a device name (to use a serial line),
40391 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
40392 stdin/stdout of @code{gdbserver}. Specify the name of the program to
40393 debug in @var{prog}. Any remaining arguments will be passed to the
40394 program verbatim. When the program exits, @value{GDBN} will close the
40395 connection, and @code{gdbserver} will exit.
40398 Debug a specific program by specifying the process ID of a running
40402 gdbserver --attach @var{comm} @var{pid}
40405 The @var{comm} parameter is as described above. Supply the process ID
40406 of a running program in @var{pid}; @value{GDBN} will do everything
40407 else. Like with the previous mode, when the process @var{pid} exits,
40408 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
40411 Multi-process mode -- debug more than one program/process:
40414 gdbserver --multi @var{comm}
40417 In this mode, @value{GDBN} can instruct @command{gdbserver} which
40418 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
40419 close the connection when a process being debugged exits, so you can
40420 debug several processes in the same session.
40423 In each of the modes you may specify these options:
40428 List all options, with brief explanations.
40431 This option causes @command{gdbserver} to print its version number and exit.
40434 @command{gdbserver} will attach to a running program. The syntax is:
40437 target> gdbserver --attach @var{comm} @var{pid}
40440 @var{pid} is the process ID of a currently running process. It isn't
40441 necessary to point @command{gdbserver} at a binary for the running process.
40444 To start @code{gdbserver} without supplying an initial command to run
40445 or process ID to attach, use this command line option.
40446 Then you can connect using @kbd{target extended-remote} and start
40447 the program you want to debug. The syntax is:
40450 target> gdbserver --multi @var{comm}
40454 Instruct @code{gdbserver} to display extra status information about the debugging
40456 This option is intended for @code{gdbserver} development and for bug reports to
40459 @item --remote-debug
40460 Instruct @code{gdbserver} to display remote protocol debug output.
40461 This option is intended for @code{gdbserver} development and for bug reports to
40464 @item --debug-format=option1@r{[},option2,...@r{]}
40465 Instruct @code{gdbserver} to include extra information in each line
40466 of debugging output.
40467 @xref{Other Command-Line Arguments for gdbserver}.
40470 Specify a wrapper to launch programs
40471 for debugging. The option should be followed by the name of the
40472 wrapper, then any command-line arguments to pass to the wrapper, then
40473 @kbd{--} indicating the end of the wrapper arguments.
40476 By default, @command{gdbserver} keeps the listening TCP port open, so that
40477 additional connections are possible. However, if you start @code{gdbserver}
40478 with the @option{--once} option, it will stop listening for any further
40479 connection attempts after connecting to the first @value{GDBN} session.
40481 @c --disable-packet is not documented for users.
40483 @c --disable-randomization and --no-disable-randomization are superseded by
40484 @c QDisableRandomization.
40489 @c man begin SEEALSO gdbserver
40491 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
40492 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
40493 documentation are properly installed at your site, the command
40499 should give you access to the complete manual.
40501 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
40502 Richard M. Stallman and Roland H. Pesch, July 1991.
40509 @c man title gcore Generate a core file of a running program
40512 @c man begin SYNOPSIS gcore
40513 gcore [-o @var{filename}] @var{pid}
40517 @c man begin DESCRIPTION gcore
40518 Generate a core dump of a running program with process ID @var{pid}.
40519 Produced file is equivalent to a kernel produced core file as if the process
40520 crashed (and if @kbd{ulimit -c} were used to set up an appropriate core dump
40521 limit). Unlike after a crash, after @command{gcore} the program remains
40522 running without any change.
40525 @c man begin OPTIONS gcore
40527 @item -o @var{filename}
40528 The optional argument
40529 @var{filename} specifies the file name where to put the core dump.
40530 If not specified, the file name defaults to @file{core.@var{pid}},
40531 where @var{pid} is the running program process ID.
40535 @c man begin SEEALSO gcore
40537 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
40538 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
40539 documentation are properly installed at your site, the command
40546 should give you access to the complete manual.
40548 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
40549 Richard M. Stallman and Roland H. Pesch, July 1991.
40556 @c man title gdbinit GDB initialization scripts
40559 @c man begin SYNOPSIS gdbinit
40560 @ifset SYSTEM_GDBINIT
40561 @value{SYSTEM_GDBINIT}
40570 @c man begin DESCRIPTION gdbinit
40571 These files contain @value{GDBN} commands to automatically execute during
40572 @value{GDBN} startup. The lines of contents are canned sequences of commands,
40575 the @value{GDBN} manual in node @code{Sequences}
40576 -- shell command @code{info -f gdb -n Sequences}.
40582 Please read more in
40584 the @value{GDBN} manual in node @code{Startup}
40585 -- shell command @code{info -f gdb -n Startup}.
40592 @ifset SYSTEM_GDBINIT
40593 @item @value{SYSTEM_GDBINIT}
40595 @ifclear SYSTEM_GDBINIT
40596 @item (not enabled with @code{--with-system-gdbinit} during compilation)
40598 System-wide initialization file. It is executed unless user specified
40599 @value{GDBN} option @code{-nx} or @code{-n}.
40602 the @value{GDBN} manual in node @code{System-wide configuration}
40603 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
40606 @ref{System-wide configuration}.
40610 User initialization file. It is executed unless user specified
40611 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
40614 Initialization file for current directory. It may need to be enabled with
40615 @value{GDBN} security command @code{set auto-load local-gdbinit}.
40618 the @value{GDBN} manual in node @code{Init File in the Current Directory}
40619 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
40622 @ref{Init File in the Current Directory}.
40627 @c man begin SEEALSO gdbinit
40629 gdb(1), @code{info -f gdb -n Startup}
40631 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
40632 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
40633 documentation are properly installed at your site, the command
40639 should give you access to the complete manual.
40641 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
40642 Richard M. Stallman and Roland H. Pesch, July 1991.
40648 @node GNU Free Documentation License
40649 @appendix GNU Free Documentation License
40652 @node Concept Index
40653 @unnumbered Concept Index
40657 @node Command and Variable Index
40658 @unnumbered Command, Variable, and Function Index
40663 % I think something like @@colophon should be in texinfo. In the
40665 \long\def\colophon{\hbox to0pt{}\vfill
40666 \centerline{The body of this manual is set in}
40667 \centerline{\fontname\tenrm,}
40668 \centerline{with headings in {\bf\fontname\tenbf}}
40669 \centerline{and examples in {\tt\fontname\tentt}.}
40670 \centerline{{\it\fontname\tenit\/},}
40671 \centerline{{\bf\fontname\tenbf}, and}
40672 \centerline{{\sl\fontname\tensl\/}}
40673 \centerline{are used for emphasis.}\vfill}
40675 % Blame: doc@@cygnus.com, 1991.