1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988-1996, 1998-2012 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.
11 @settitle Debugging with @value{GDBN}
12 @setchapternewpage odd
24 @c readline appendices use @vindex, @findex and @ftable,
25 @c annotate.texi and gdbmi use @findex.
29 @c !!set GDB manual's edition---not the same as GDB version!
30 @c This is updated by GNU Press.
33 @c !!set GDB edit command default editor
36 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
38 @c This is a dir.info fragment to support semi-automated addition of
39 @c manuals to an info tree.
40 @dircategory Software development
42 * Gdb: (gdb). The GNU debugger.
46 Copyright @copyright{} 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996,
47 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
49 Free Software Foundation, Inc.
51 Permission is granted to copy, distribute and/or modify this document
52 under the terms of the GNU Free Documentation License, Version 1.3 or
53 any later version published by the Free Software Foundation; with the
54 Invariant Sections being ``Free Software'' and ``Free Software Needs
55 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
56 and with the Back-Cover Texts as in (a) below.
58 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
59 this GNU Manual. Buying copies from GNU Press supports the FSF in
60 developing GNU and promoting software freedom.''
64 This file documents the @sc{gnu} debugger @value{GDBN}.
66 This is the @value{EDITION} Edition, of @cite{Debugging with
67 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
68 @ifset VERSION_PACKAGE
69 @value{VERSION_PACKAGE}
71 Version @value{GDBVN}.
77 @title Debugging with @value{GDBN}
78 @subtitle The @sc{gnu} Source-Level Debugger
80 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
81 @ifset VERSION_PACKAGE
83 @subtitle @value{VERSION_PACKAGE}
85 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
89 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
90 \hfill {\it Debugging with @value{GDBN}}\par
91 \hfill \TeX{}info \texinfoversion\par
95 @vskip 0pt plus 1filll
96 Published by the Free Software Foundation @*
97 51 Franklin Street, Fifth Floor,
98 Boston, MA 02110-1301, USA@*
99 ISBN 978-0-9831592-3-0 @*
106 @node Top, Summary, (dir), (dir)
108 @top Debugging with @value{GDBN}
110 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
112 This is the @value{EDITION} Edition, for @value{GDBN}
113 @ifset VERSION_PACKAGE
114 @value{VERSION_PACKAGE}
116 Version @value{GDBVN}.
118 Copyright (C) 1988-2012 Free Software Foundation, Inc.
120 This edition of the GDB manual is dedicated to the memory of Fred
121 Fish. Fred was a long-standing contributor to GDB and to Free
122 software in general. We will miss him.
125 * Summary:: Summary of @value{GDBN}
126 * Sample Session:: A sample @value{GDBN} session
128 * Invocation:: Getting in and out of @value{GDBN}
129 * Commands:: @value{GDBN} commands
130 * Running:: Running programs under @value{GDBN}
131 * Stopping:: Stopping and continuing
132 * Reverse Execution:: Running programs backward
133 * Process Record and Replay:: Recording inferior's execution and replaying it
134 * Stack:: Examining the stack
135 * Source:: Examining source files
136 * Data:: Examining data
137 * Optimized Code:: Debugging optimized code
138 * Macros:: Preprocessor Macros
139 * Tracepoints:: Debugging remote targets non-intrusively
140 * Overlays:: Debugging programs that use overlays
142 * Languages:: Using @value{GDBN} with different languages
144 * Symbols:: Examining the symbol table
145 * Altering:: Altering execution
146 * GDB Files:: @value{GDBN} files
147 * Targets:: Specifying a debugging target
148 * Remote Debugging:: Debugging remote programs
149 * Configurations:: Configuration-specific information
150 * Controlling GDB:: Controlling @value{GDBN}
151 * Extending GDB:: Extending @value{GDBN}
152 * Interpreters:: Command Interpreters
153 * TUI:: @value{GDBN} Text User Interface
154 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
155 * GDB/MI:: @value{GDBN}'s Machine Interface.
156 * Annotations:: @value{GDBN}'s annotation interface.
157 * JIT Interface:: Using the JIT debugging interface.
158 * In-Process Agent:: In-Process Agent
160 * GDB Bugs:: Reporting bugs in @value{GDBN}
162 @ifset SYSTEM_READLINE
163 * Command Line Editing: (rluserman). Command Line Editing
164 * Using History Interactively: (history). Using History Interactively
166 @ifclear SYSTEM_READLINE
167 * Command Line Editing:: Command Line Editing
168 * Using History Interactively:: Using History Interactively
170 * In Memoriam:: In Memoriam
171 * Formatting Documentation:: How to format and print @value{GDBN} documentation
172 * Installing GDB:: Installing GDB
173 * Maintenance Commands:: Maintenance Commands
174 * Remote Protocol:: GDB Remote Serial Protocol
175 * Agent Expressions:: The GDB Agent Expression Mechanism
176 * Target Descriptions:: How targets can describe themselves to
178 * Operating System Information:: Getting additional information from
180 * Trace File Format:: GDB trace file format
181 * Index Section Format:: .gdb_index section format
182 * Copying:: GNU General Public License says
183 how you can copy and share GDB
184 * GNU Free Documentation License:: The license for this documentation
193 @unnumbered Summary of @value{GDBN}
195 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
196 going on ``inside'' another program while it executes---or what another
197 program was doing at the moment it crashed.
199 @value{GDBN} can do four main kinds of things (plus other things in support of
200 these) to help you catch bugs in the act:
204 Start your program, specifying anything that might affect its behavior.
207 Make your program stop on specified conditions.
210 Examine what has happened, when your program has stopped.
213 Change things in your program, so you can experiment with correcting the
214 effects of one bug and go on to learn about another.
217 You can use @value{GDBN} to debug programs written in C and C@t{++}.
218 For more information, see @ref{Supported Languages,,Supported Languages}.
219 For more information, see @ref{C,,C and C++}.
221 Support for D is partial. For information on D, see
225 Support for Modula-2 is partial. For information on Modula-2, see
226 @ref{Modula-2,,Modula-2}.
228 Support for OpenCL C is partial. For information on OpenCL C, see
229 @ref{OpenCL C,,OpenCL C}.
232 Debugging Pascal programs which use sets, subranges, file variables, or
233 nested functions does not currently work. @value{GDBN} does not support
234 entering expressions, printing values, or similar features using Pascal
238 @value{GDBN} can be used to debug programs written in Fortran, although
239 it may be necessary to refer to some variables with a trailing
242 @value{GDBN} can be used to debug programs written in Objective-C,
243 using either the Apple/NeXT or the GNU Objective-C runtime.
246 * Free Software:: Freely redistributable software
247 * Free Documentation:: Free Software Needs Free Documentation
248 * Contributors:: Contributors to GDB
252 @unnumberedsec Free Software
254 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
255 General Public License
256 (GPL). The GPL gives you the freedom to copy or adapt a licensed
257 program---but every person getting a copy also gets with it the
258 freedom to modify that copy (which means that they must get access to
259 the source code), and the freedom to distribute further copies.
260 Typical software companies use copyrights to limit your freedoms; the
261 Free Software Foundation uses the GPL to preserve these freedoms.
263 Fundamentally, the General Public License is a license which says that
264 you have these freedoms and that you cannot take these freedoms away
267 @node Free Documentation
268 @unnumberedsec Free Software Needs Free Documentation
270 The biggest deficiency in the free software community today is not in
271 the software---it is the lack of good free documentation that we can
272 include with the free software. Many of our most important
273 programs do not come with free reference manuals and free introductory
274 texts. Documentation is an essential part of any software package;
275 when an important free software package does not come with a free
276 manual and a free tutorial, that is a major gap. We have many such
279 Consider Perl, for instance. The tutorial manuals that people
280 normally use are non-free. How did this come about? Because the
281 authors of those manuals published them with restrictive terms---no
282 copying, no modification, source files not available---which exclude
283 them from the free software world.
285 That wasn't the first time this sort of thing happened, and it was far
286 from the last. Many times we have heard a GNU user eagerly describe a
287 manual that he is writing, his intended contribution to the community,
288 only to learn that he had ruined everything by signing a publication
289 contract to make it non-free.
291 Free documentation, like free software, is a matter of freedom, not
292 price. The problem with the non-free manual is not that publishers
293 charge a price for printed copies---that in itself is fine. (The Free
294 Software Foundation sells printed copies of manuals, too.) The
295 problem is the restrictions on the use of the manual. Free manuals
296 are available in source code form, and give you permission to copy and
297 modify. Non-free manuals do not allow this.
299 The criteria of freedom for a free manual are roughly the same as for
300 free software. Redistribution (including the normal kinds of
301 commercial redistribution) must be permitted, so that the manual can
302 accompany every copy of the program, both on-line and on paper.
304 Permission for modification of the technical content is crucial too.
305 When people modify the software, adding or changing features, if they
306 are conscientious they will change the manual too---so they can
307 provide accurate and clear documentation for the modified program. A
308 manual that leaves you no choice but to write a new manual to document
309 a changed version of the program is not really available to our
312 Some kinds of limits on the way modification is handled are
313 acceptable. For example, requirements to preserve the original
314 author's copyright notice, the distribution terms, or the list of
315 authors, are ok. It is also no problem to require modified versions
316 to include notice that they were modified. Even entire sections that
317 may not be deleted or changed are acceptable, as long as they deal
318 with nontechnical topics (like this one). These kinds of restrictions
319 are acceptable because they don't obstruct the community's normal use
322 However, it must be possible to modify all the @emph{technical}
323 content of the manual, and then distribute the result in all the usual
324 media, through all the usual channels. Otherwise, the restrictions
325 obstruct the use of the manual, it is not free, and we need another
326 manual to replace it.
328 Please spread the word about this issue. Our community continues to
329 lose manuals to proprietary publishing. If we spread the word that
330 free software needs free reference manuals and free tutorials, perhaps
331 the next person who wants to contribute by writing documentation will
332 realize, before it is too late, that only free manuals contribute to
333 the free software community.
335 If you are writing documentation, please insist on publishing it under
336 the GNU Free Documentation License or another free documentation
337 license. Remember that this decision requires your approval---you
338 don't have to let the publisher decide. Some commercial publishers
339 will use a free license if you insist, but they will not propose the
340 option; it is up to you to raise the issue and say firmly that this is
341 what you want. If the publisher you are dealing with refuses, please
342 try other publishers. If you're not sure whether a proposed license
343 is free, write to @email{licensing@@gnu.org}.
345 You can encourage commercial publishers to sell more free, copylefted
346 manuals and tutorials by buying them, and particularly by buying
347 copies from the publishers that paid for their writing or for major
348 improvements. Meanwhile, try to avoid buying non-free documentation
349 at all. Check the distribution terms of a manual before you buy it,
350 and insist that whoever seeks your business must respect your freedom.
351 Check the history of the book, and try to reward the publishers that
352 have paid or pay the authors to work on it.
354 The Free Software Foundation maintains a list of free documentation
355 published by other publishers, at
356 @url{http://www.fsf.org/doc/other-free-books.html}.
359 @unnumberedsec Contributors to @value{GDBN}
361 Richard Stallman was the original author of @value{GDBN}, and of many
362 other @sc{gnu} programs. Many others have contributed to its
363 development. This section attempts to credit major contributors. One
364 of the virtues of free software is that everyone is free to contribute
365 to it; with regret, we cannot actually acknowledge everyone here. The
366 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
367 blow-by-blow account.
369 Changes much prior to version 2.0 are lost in the mists of time.
372 @emph{Plea:} Additions to this section are particularly welcome. If you
373 or your friends (or enemies, to be evenhanded) have been unfairly
374 omitted from this list, we would like to add your names!
377 So that they may not regard their many labors as thankless, we
378 particularly thank those who shepherded @value{GDBN} through major
380 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
381 Jim Blandy (release 4.18);
382 Jason Molenda (release 4.17);
383 Stan Shebs (release 4.14);
384 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
385 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
386 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
387 Jim Kingdon (releases 3.5, 3.4, and 3.3);
388 and Randy Smith (releases 3.2, 3.1, and 3.0).
390 Richard Stallman, assisted at various times by Peter TerMaat, Chris
391 Hanson, and Richard Mlynarik, handled releases through 2.8.
393 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
394 in @value{GDBN}, with significant additional contributions from Per
395 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
396 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
397 much general update work leading to release 3.0).
399 @value{GDBN} uses the BFD subroutine library to examine multiple
400 object-file formats; BFD was a joint project of David V.
401 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
403 David Johnson wrote the original COFF support; Pace Willison did
404 the original support for encapsulated COFF.
406 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
408 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
409 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
411 Jean-Daniel Fekete contributed Sun 386i support.
412 Chris Hanson improved the HP9000 support.
413 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
414 David Johnson contributed Encore Umax support.
415 Jyrki Kuoppala contributed Altos 3068 support.
416 Jeff Law contributed HP PA and SOM support.
417 Keith Packard contributed NS32K support.
418 Doug Rabson contributed Acorn Risc Machine support.
419 Bob Rusk contributed Harris Nighthawk CX-UX support.
420 Chris Smith contributed Convex support (and Fortran debugging).
421 Jonathan Stone contributed Pyramid support.
422 Michael Tiemann contributed SPARC support.
423 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
424 Pace Willison contributed Intel 386 support.
425 Jay Vosburgh contributed Symmetry support.
426 Marko Mlinar contributed OpenRISC 1000 support.
428 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
430 Rich Schaefer and Peter Schauer helped with support of SunOS shared
433 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
434 about several machine instruction sets.
436 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
437 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
438 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
439 and RDI targets, respectively.
441 Brian Fox is the author of the readline libraries providing
442 command-line editing and command history.
444 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
445 Modula-2 support, and contributed the Languages chapter of this manual.
447 Fred Fish wrote most of the support for Unix System Vr4.
448 He also enhanced the command-completion support to cover C@t{++} overloaded
451 Hitachi America (now Renesas America), Ltd. sponsored the support for
452 H8/300, H8/500, and Super-H processors.
454 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
456 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
459 Toshiba sponsored the support for the TX39 Mips processor.
461 Matsushita sponsored the support for the MN10200 and MN10300 processors.
463 Fujitsu sponsored the support for SPARClite and FR30 processors.
465 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
468 Michael Snyder added support for tracepoints.
470 Stu Grossman wrote gdbserver.
472 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
473 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
475 The following people at the Hewlett-Packard Company contributed
476 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
477 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
478 compiler, and the Text User Interface (nee Terminal User Interface):
479 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
480 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
481 provided HP-specific information in this manual.
483 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
484 Robert Hoehne made significant contributions to the DJGPP port.
486 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
487 development since 1991. Cygnus engineers who have worked on @value{GDBN}
488 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
489 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
490 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
491 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
492 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
493 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
494 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
495 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
496 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
497 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
498 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
499 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
500 Zuhn have made contributions both large and small.
502 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
503 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
505 Jim Blandy added support for preprocessor macros, while working for Red
508 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
509 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
510 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
511 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
512 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
513 with the migration of old architectures to this new framework.
515 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
516 unwinder framework, this consisting of a fresh new design featuring
517 frame IDs, independent frame sniffers, and the sentinel frame. Mark
518 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
519 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
520 trad unwinders. The architecture-specific changes, each involving a
521 complete rewrite of the architecture's frame code, were carried out by
522 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
523 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
524 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
525 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
528 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
529 Tensilica, Inc.@: contributed support for Xtensa processors. Others
530 who have worked on the Xtensa port of @value{GDBN} in the past include
531 Steve Tjiang, John Newlin, and Scott Foehner.
533 Michael Eager and staff of Xilinx, Inc., contributed support for the
534 Xilinx MicroBlaze architecture.
537 @chapter A Sample @value{GDBN} Session
539 You can use this manual at your leisure to read all about @value{GDBN}.
540 However, a handful of commands are enough to get started using the
541 debugger. This chapter illustrates those commands.
544 In this sample session, we emphasize user input like this: @b{input},
545 to make it easier to pick out from the surrounding output.
548 @c FIXME: this example may not be appropriate for some configs, where
549 @c FIXME...primary interest is in remote use.
551 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
552 processor) exhibits the following bug: sometimes, when we change its
553 quote strings from the default, the commands used to capture one macro
554 definition within another stop working. In the following short @code{m4}
555 session, we define a macro @code{foo} which expands to @code{0000}; we
556 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
557 same thing. However, when we change the open quote string to
558 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
559 procedure fails to define a new synonym @code{baz}:
568 @b{define(bar,defn(`foo'))}
572 @b{changequote(<QUOTE>,<UNQUOTE>)}
574 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
577 m4: End of input: 0: fatal error: EOF in string
581 Let us use @value{GDBN} to try to see what is going on.
584 $ @b{@value{GDBP} m4}
585 @c FIXME: this falsifies the exact text played out, to permit smallbook
586 @c FIXME... format to come out better.
587 @value{GDBN} is free software and you are welcome to distribute copies
588 of it under certain conditions; type "show copying" to see
590 There is absolutely no warranty for @value{GDBN}; type "show warranty"
593 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
598 @value{GDBN} reads only enough symbol data to know where to find the
599 rest when needed; as a result, the first prompt comes up very quickly.
600 We now tell @value{GDBN} to use a narrower display width than usual, so
601 that examples fit in this manual.
604 (@value{GDBP}) @b{set width 70}
608 We need to see how the @code{m4} built-in @code{changequote} works.
609 Having looked at the source, we know the relevant subroutine is
610 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
611 @code{break} command.
614 (@value{GDBP}) @b{break m4_changequote}
615 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
619 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
620 control; as long as control does not reach the @code{m4_changequote}
621 subroutine, the program runs as usual:
624 (@value{GDBP}) @b{run}
625 Starting program: /work/Editorial/gdb/gnu/m4/m4
633 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
634 suspends execution of @code{m4}, displaying information about the
635 context where it stops.
638 @b{changequote(<QUOTE>,<UNQUOTE>)}
640 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
642 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
646 Now we use the command @code{n} (@code{next}) to advance execution to
647 the next line of the current function.
651 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
656 @code{set_quotes} looks like a promising subroutine. We can go into it
657 by using the command @code{s} (@code{step}) instead of @code{next}.
658 @code{step} goes to the next line to be executed in @emph{any}
659 subroutine, so it steps into @code{set_quotes}.
663 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
665 530 if (lquote != def_lquote)
669 The display that shows the subroutine where @code{m4} is now
670 suspended (and its arguments) is called a stack frame display. It
671 shows a summary of the stack. We can use the @code{backtrace}
672 command (which can also be spelled @code{bt}), to see where we are
673 in the stack as a whole: the @code{backtrace} command displays a
674 stack frame for each active subroutine.
677 (@value{GDBP}) @b{bt}
678 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
680 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
682 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
683 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
685 #4 0x79dc in expand_input () at macro.c:40
686 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
690 We step through a few more lines to see what happens. The first two
691 times, we can use @samp{s}; the next two times we use @code{n} to avoid
692 falling into the @code{xstrdup} subroutine.
696 0x3b5c 532 if (rquote != def_rquote)
698 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
699 def_lquote : xstrdup(lq);
701 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
704 538 len_lquote = strlen(rquote);
708 The last line displayed looks a little odd; we can examine the variables
709 @code{lquote} and @code{rquote} to see if they are in fact the new left
710 and right quotes we specified. We use the command @code{p}
711 (@code{print}) to see their values.
714 (@value{GDBP}) @b{p lquote}
715 $1 = 0x35d40 "<QUOTE>"
716 (@value{GDBP}) @b{p rquote}
717 $2 = 0x35d50 "<UNQUOTE>"
721 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
722 To look at some context, we can display ten lines of source
723 surrounding the current line with the @code{l} (@code{list}) command.
729 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
731 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
734 538 len_lquote = strlen(rquote);
735 539 len_rquote = strlen(lquote);
742 Let us step past the two lines that set @code{len_lquote} and
743 @code{len_rquote}, and then examine the values of those variables.
747 539 len_rquote = strlen(lquote);
750 (@value{GDBP}) @b{p len_lquote}
752 (@value{GDBP}) @b{p len_rquote}
757 That certainly looks wrong, assuming @code{len_lquote} and
758 @code{len_rquote} are meant to be the lengths of @code{lquote} and
759 @code{rquote} respectively. We can set them to better values using
760 the @code{p} command, since it can print the value of
761 any expression---and that expression can include subroutine calls and
765 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
767 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
772 Is that enough to fix the problem of using the new quotes with the
773 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
774 executing with the @code{c} (@code{continue}) command, and then try the
775 example that caused trouble initially:
781 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
788 Success! The new quotes now work just as well as the default ones. The
789 problem seems to have been just the two typos defining the wrong
790 lengths. We allow @code{m4} exit by giving it an EOF as input:
794 Program exited normally.
798 The message @samp{Program exited normally.} is from @value{GDBN}; it
799 indicates @code{m4} has finished executing. We can end our @value{GDBN}
800 session with the @value{GDBN} @code{quit} command.
803 (@value{GDBP}) @b{quit}
807 @chapter Getting In and Out of @value{GDBN}
809 This chapter discusses how to start @value{GDBN}, and how to get out of it.
813 type @samp{@value{GDBP}} to start @value{GDBN}.
815 type @kbd{quit} or @kbd{Ctrl-d} to exit.
819 * Invoking GDB:: How to start @value{GDBN}
820 * Quitting GDB:: How to quit @value{GDBN}
821 * Shell Commands:: How to use shell commands inside @value{GDBN}
822 * Logging Output:: How to log @value{GDBN}'s output to a file
826 @section Invoking @value{GDBN}
828 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
829 @value{GDBN} reads commands from the terminal until you tell it to exit.
831 You can also run @code{@value{GDBP}} with a variety of arguments and options,
832 to specify more of your debugging environment at the outset.
834 The command-line options described here are designed
835 to cover a variety of situations; in some environments, some of these
836 options may effectively be unavailable.
838 The most usual way to start @value{GDBN} is with one argument,
839 specifying an executable program:
842 @value{GDBP} @var{program}
846 You can also start with both an executable program and a core file
850 @value{GDBP} @var{program} @var{core}
853 You can, instead, specify a process ID as a second argument, if you want
854 to debug a running process:
857 @value{GDBP} @var{program} 1234
861 would attach @value{GDBN} to process @code{1234} (unless you also have a file
862 named @file{1234}; @value{GDBN} does check for a core file first).
864 Taking advantage of the second command-line argument requires a fairly
865 complete operating system; when you use @value{GDBN} as a remote
866 debugger attached to a bare board, there may not be any notion of
867 ``process'', and there is often no way to get a core dump. @value{GDBN}
868 will warn you if it is unable to attach or to read core dumps.
870 You can optionally have @code{@value{GDBP}} pass any arguments after the
871 executable file to the inferior using @code{--args}. This option stops
874 @value{GDBP} --args gcc -O2 -c foo.c
876 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
877 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
879 You can run @code{@value{GDBP}} without printing the front material, which describes
880 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
887 You can further control how @value{GDBN} starts up by using command-line
888 options. @value{GDBN} itself can remind you of the options available.
898 to display all available options and briefly describe their use
899 (@samp{@value{GDBP} -h} is a shorter equivalent).
901 All options and command line arguments you give are processed
902 in sequential order. The order makes a difference when the
903 @samp{-x} option is used.
907 * File Options:: Choosing files
908 * Mode Options:: Choosing modes
909 * Startup:: What @value{GDBN} does during startup
913 @subsection Choosing Files
915 When @value{GDBN} starts, it reads any arguments other than options as
916 specifying an executable file and core file (or process ID). This is
917 the same as if the arguments were specified by the @samp{-se} and
918 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
919 first argument that does not have an associated option flag as
920 equivalent to the @samp{-se} option followed by that argument; and the
921 second argument that does not have an associated option flag, if any, as
922 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
923 If the second argument begins with a decimal digit, @value{GDBN} will
924 first attempt to attach to it as a process, and if that fails, attempt
925 to open it as a corefile. If you have a corefile whose name begins with
926 a digit, you can prevent @value{GDBN} from treating it as a pid by
927 prefixing it with @file{./}, e.g.@: @file{./12345}.
929 If @value{GDBN} has not been configured to included core file support,
930 such as for most embedded targets, then it will complain about a second
931 argument and ignore it.
933 Many options have both long and short forms; both are shown in the
934 following list. @value{GDBN} also recognizes the long forms if you truncate
935 them, so long as enough of the option is present to be unambiguous.
936 (If you prefer, you can flag option arguments with @samp{--} rather
937 than @samp{-}, though we illustrate the more usual convention.)
939 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
940 @c way, both those who look for -foo and --foo in the index, will find
944 @item -symbols @var{file}
946 @cindex @code{--symbols}
948 Read symbol table from file @var{file}.
950 @item -exec @var{file}
952 @cindex @code{--exec}
954 Use file @var{file} as the executable file to execute when appropriate,
955 and for examining pure data in conjunction with a core dump.
959 Read symbol table from file @var{file} and use it as the executable
962 @item -core @var{file}
964 @cindex @code{--core}
966 Use file @var{file} as a core dump to examine.
968 @item -pid @var{number}
969 @itemx -p @var{number}
972 Connect to process ID @var{number}, as with the @code{attach} command.
974 @item -command @var{file}
976 @cindex @code{--command}
978 Execute commands from file @var{file}. The contents of this file is
979 evaluated exactly as the @code{source} command would.
980 @xref{Command Files,, Command files}.
982 @item -eval-command @var{command}
983 @itemx -ex @var{command}
984 @cindex @code{--eval-command}
986 Execute a single @value{GDBN} command.
988 This option may be used multiple times to call multiple commands. It may
989 also be interleaved with @samp{-command} as required.
992 @value{GDBP} -ex 'target sim' -ex 'load' \
993 -x setbreakpoints -ex 'run' a.out
996 @item -init-command @var{file}
997 @itemx -ix @var{file}
998 @cindex @code{--init-command}
1000 Execute commands from file @var{file} before loading the inferior (but
1001 after loading gdbinit files).
1004 @item -init-eval-command @var{command}
1005 @itemx -iex @var{command}
1006 @cindex @code{--init-eval-command}
1008 Execute a single @value{GDBN} command before loading the inferior (but
1009 after loading gdbinit files).
1012 @item -directory @var{directory}
1013 @itemx -d @var{directory}
1014 @cindex @code{--directory}
1016 Add @var{directory} to the path to search for source and script files.
1020 @cindex @code{--readnow}
1022 Read each symbol file's entire symbol table immediately, rather than
1023 the default, which is to read it incrementally as it is needed.
1024 This makes startup slower, but makes future operations faster.
1029 @subsection Choosing Modes
1031 You can run @value{GDBN} in various alternative modes---for example, in
1032 batch mode or quiet mode.
1040 Do not execute commands found in any initialization files. Normally,
1041 @value{GDBN} executes the commands in these files after all the command
1042 options and arguments have been processed. @xref{Command Files,,Command
1048 @cindex @code{--quiet}
1049 @cindex @code{--silent}
1051 ``Quiet''. Do not print the introductory and copyright messages. These
1052 messages are also suppressed in batch mode.
1055 @cindex @code{--batch}
1056 Run in batch mode. Exit with status @code{0} after processing all the
1057 command files specified with @samp{-x} (and all commands from
1058 initialization files, if not inhibited with @samp{-n}). Exit with
1059 nonzero status if an error occurs in executing the @value{GDBN} commands
1060 in the command files. Batch mode also disables pagination, sets unlimited
1061 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1062 off} were in effect (@pxref{Messages/Warnings}).
1064 Batch mode may be useful for running @value{GDBN} as a filter, for
1065 example to download and run a program on another computer; in order to
1066 make this more useful, the message
1069 Program exited normally.
1073 (which is ordinarily issued whenever a program running under
1074 @value{GDBN} control terminates) is not issued when running in batch
1078 @cindex @code{--batch-silent}
1079 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1080 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1081 unaffected). This is much quieter than @samp{-silent} and would be useless
1082 for an interactive session.
1084 This is particularly useful when using targets that give @samp{Loading section}
1085 messages, for example.
1087 Note that targets that give their output via @value{GDBN}, as opposed to
1088 writing directly to @code{stdout}, will also be made silent.
1090 @item -return-child-result
1091 @cindex @code{--return-child-result}
1092 The return code from @value{GDBN} will be the return code from the child
1093 process (the process being debugged), with the following exceptions:
1097 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1098 internal error. In this case the exit code is the same as it would have been
1099 without @samp{-return-child-result}.
1101 The user quits with an explicit value. E.g., @samp{quit 1}.
1103 The child process never runs, or is not allowed to terminate, in which case
1104 the exit code will be -1.
1107 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1108 when @value{GDBN} is being used as a remote program loader or simulator
1113 @cindex @code{--nowindows}
1115 ``No windows''. If @value{GDBN} comes with a graphical user interface
1116 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1117 interface. If no GUI is available, this option has no effect.
1121 @cindex @code{--windows}
1123 If @value{GDBN} includes a GUI, then this option requires it to be
1126 @item -cd @var{directory}
1128 Run @value{GDBN} using @var{directory} as its working directory,
1129 instead of the current directory.
1131 @item -data-directory @var{directory}
1132 @cindex @code{--data-directory}
1133 Run @value{GDBN} using @var{directory} as its data directory.
1134 The data directory is where @value{GDBN} searches for its
1135 auxiliary files. @xref{Data Files}.
1139 @cindex @code{--fullname}
1141 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1142 subprocess. It tells @value{GDBN} to output the full file name and line
1143 number in a standard, recognizable fashion each time a stack frame is
1144 displayed (which includes each time your program stops). This
1145 recognizable format looks like two @samp{\032} characters, followed by
1146 the file name, line number and character position separated by colons,
1147 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1148 @samp{\032} characters as a signal to display the source code for the
1152 @cindex @code{--epoch}
1153 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1154 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1155 routines so as to allow Epoch to display values of expressions in a
1158 @item -annotate @var{level}
1159 @cindex @code{--annotate}
1160 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1161 effect is identical to using @samp{set annotate @var{level}}
1162 (@pxref{Annotations}). The annotation @var{level} controls how much
1163 information @value{GDBN} prints together with its prompt, values of
1164 expressions, source lines, and other types of output. Level 0 is the
1165 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1166 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1167 that control @value{GDBN}, and level 2 has been deprecated.
1169 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1173 @cindex @code{--args}
1174 Change interpretation of command line so that arguments following the
1175 executable file are passed as command line arguments to the inferior.
1176 This option stops option processing.
1178 @item -baud @var{bps}
1180 @cindex @code{--baud}
1182 Set the line speed (baud rate or bits per second) of any serial
1183 interface used by @value{GDBN} for remote debugging.
1185 @item -l @var{timeout}
1187 Set the timeout (in seconds) of any communication used by @value{GDBN}
1188 for remote debugging.
1190 @item -tty @var{device}
1191 @itemx -t @var{device}
1192 @cindex @code{--tty}
1194 Run using @var{device} for your program's standard input and output.
1195 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1197 @c resolve the situation of these eventually
1199 @cindex @code{--tui}
1200 Activate the @dfn{Text User Interface} when starting. The Text User
1201 Interface manages several text windows on the terminal, showing
1202 source, assembly, registers and @value{GDBN} command outputs
1203 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1204 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1205 Using @value{GDBN} under @sc{gnu} Emacs}).
1208 @c @cindex @code{--xdb}
1209 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1210 @c For information, see the file @file{xdb_trans.html}, which is usually
1211 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1214 @item -interpreter @var{interp}
1215 @cindex @code{--interpreter}
1216 Use the interpreter @var{interp} for interface with the controlling
1217 program or device. This option is meant to be set by programs which
1218 communicate with @value{GDBN} using it as a back end.
1219 @xref{Interpreters, , Command Interpreters}.
1221 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1222 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1223 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1224 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1225 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1226 @sc{gdb/mi} interfaces are no longer supported.
1229 @cindex @code{--write}
1230 Open the executable and core files for both reading and writing. This
1231 is equivalent to the @samp{set write on} command inside @value{GDBN}
1235 @cindex @code{--statistics}
1236 This option causes @value{GDBN} to print statistics about time and
1237 memory usage after it completes each command and returns to the prompt.
1240 @cindex @code{--version}
1241 This option causes @value{GDBN} to print its version number and
1242 no-warranty blurb, and exit.
1244 @item -use-deprecated-index-sections
1245 @cindex @code{--use-deprecated-index-sections}
1246 This option causes @value{GDBN} to read and use deprecated
1247 @samp{.gdb_index} sections from symbol files. This can speed up
1248 startup, but may result in some functionality being lost.
1249 @xref{Index Section Format}.
1254 @subsection What @value{GDBN} Does During Startup
1255 @cindex @value{GDBN} startup
1257 Here's the description of what @value{GDBN} does during session startup:
1261 Sets up the command interpreter as specified by the command line
1262 (@pxref{Mode Options, interpreter}).
1266 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1267 used when building @value{GDBN}; @pxref{System-wide configuration,
1268 ,System-wide configuration and settings}) and executes all the commands in
1271 @anchor{Home Directory Init File}
1273 Reads the init file (if any) in your home directory@footnote{On
1274 DOS/Windows systems, the home directory is the one pointed to by the
1275 @code{HOME} environment variable.} and executes all the commands in
1278 @anchor{Option -init-eval-command}
1280 Executes commands and command files specified by the @samp{-iex} and
1281 @samp{-ix} options in their specified order. Usually you should use the
1282 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1283 settings before @value{GDBN} init files get executed and before inferior
1287 Processes command line options and operands.
1289 @anchor{Init File in the Current Directory during Startup}
1291 Reads and executes the commands from init file (if any) in the current
1292 working directory as long as @samp{set auto-load local-gdbinit} is set to
1293 @samp{on} (@pxref{Init File in the Current Directory}).
1294 This is only done if the current directory is
1295 different from your home directory. Thus, you can have more than one
1296 init file, one generic in your home directory, and another, specific
1297 to the program you are debugging, in the directory where you invoke
1301 If the command line specified a program to debug, or a process to
1302 attach to, or a core file, @value{GDBN} loads any auto-loaded
1303 scripts provided for the program or for its loaded shared libraries.
1304 @xref{Auto-loading}.
1306 If you wish to disable the auto-loading during startup,
1307 you must do something like the following:
1310 $ gdb -iex "set auto-load python-scripts off" myprogram
1313 Option @samp{-ex} does not work because the auto-loading is then turned
1317 Executes commands and command files specified by the @samp{-ex} and
1318 @samp{-x} options in their specified order. @xref{Command Files}, for
1319 more details about @value{GDBN} command files.
1322 Reads the command history recorded in the @dfn{history file}.
1323 @xref{Command History}, for more details about the command history and the
1324 files where @value{GDBN} records it.
1327 Init files use the same syntax as @dfn{command files} (@pxref{Command
1328 Files}) and are processed by @value{GDBN} in the same way. The init
1329 file in your home directory can set options (such as @samp{set
1330 complaints}) that affect subsequent processing of command line options
1331 and operands. Init files are not executed if you use the @samp{-nx}
1332 option (@pxref{Mode Options, ,Choosing Modes}).
1334 To display the list of init files loaded by gdb at startup, you
1335 can use @kbd{gdb --help}.
1337 @cindex init file name
1338 @cindex @file{.gdbinit}
1339 @cindex @file{gdb.ini}
1340 The @value{GDBN} init files are normally called @file{.gdbinit}.
1341 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1342 the limitations of file names imposed by DOS filesystems. The Windows
1343 ports of @value{GDBN} use the standard name, but if they find a
1344 @file{gdb.ini} file, they warn you about that and suggest to rename
1345 the file to the standard name.
1349 @section Quitting @value{GDBN}
1350 @cindex exiting @value{GDBN}
1351 @cindex leaving @value{GDBN}
1354 @kindex quit @r{[}@var{expression}@r{]}
1355 @kindex q @r{(@code{quit})}
1356 @item quit @r{[}@var{expression}@r{]}
1358 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1359 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1360 do not supply @var{expression}, @value{GDBN} will terminate normally;
1361 otherwise it will terminate using the result of @var{expression} as the
1366 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1367 terminates the action of any @value{GDBN} command that is in progress and
1368 returns to @value{GDBN} command level. It is safe to type the interrupt
1369 character at any time because @value{GDBN} does not allow it to take effect
1370 until a time when it is safe.
1372 If you have been using @value{GDBN} to control an attached process or
1373 device, you can release it with the @code{detach} command
1374 (@pxref{Attach, ,Debugging an Already-running Process}).
1376 @node Shell Commands
1377 @section Shell Commands
1379 If you need to execute occasional shell commands during your
1380 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1381 just use the @code{shell} command.
1386 @cindex shell escape
1387 @item shell @var{command-string}
1388 @itemx !@var{command-string}
1389 Invoke a standard shell to execute @var{command-string}.
1390 Note that no space is needed between @code{!} and @var{command-string}.
1391 If it exists, the environment variable @code{SHELL} determines which
1392 shell to run. Otherwise @value{GDBN} uses the default shell
1393 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1396 The utility @code{make} is often needed in development environments.
1397 You do not have to use the @code{shell} command for this purpose in
1402 @cindex calling make
1403 @item make @var{make-args}
1404 Execute the @code{make} program with the specified
1405 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1408 @node Logging Output
1409 @section Logging Output
1410 @cindex logging @value{GDBN} output
1411 @cindex save @value{GDBN} output to a file
1413 You may want to save the output of @value{GDBN} commands to a file.
1414 There are several commands to control @value{GDBN}'s logging.
1418 @item set logging on
1420 @item set logging off
1422 @cindex logging file name
1423 @item set logging file @var{file}
1424 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1425 @item set logging overwrite [on|off]
1426 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1427 you want @code{set logging on} to overwrite the logfile instead.
1428 @item set logging redirect [on|off]
1429 By default, @value{GDBN} output will go to both the terminal and the logfile.
1430 Set @code{redirect} if you want output to go only to the log file.
1431 @kindex show logging
1433 Show the current values of the logging settings.
1437 @chapter @value{GDBN} Commands
1439 You can abbreviate a @value{GDBN} command to the first few letters of the command
1440 name, if that abbreviation is unambiguous; and you can repeat certain
1441 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1442 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1443 show you the alternatives available, if there is more than one possibility).
1446 * Command Syntax:: How to give commands to @value{GDBN}
1447 * Completion:: Command completion
1448 * Help:: How to ask @value{GDBN} for help
1451 @node Command Syntax
1452 @section Command Syntax
1454 A @value{GDBN} command is a single line of input. There is no limit on
1455 how long it can be. It starts with a command name, which is followed by
1456 arguments whose meaning depends on the command name. For example, the
1457 command @code{step} accepts an argument which is the number of times to
1458 step, as in @samp{step 5}. You can also use the @code{step} command
1459 with no arguments. Some commands do not allow any arguments.
1461 @cindex abbreviation
1462 @value{GDBN} command names may always be truncated if that abbreviation is
1463 unambiguous. Other possible command abbreviations are listed in the
1464 documentation for individual commands. In some cases, even ambiguous
1465 abbreviations are allowed; for example, @code{s} is specially defined as
1466 equivalent to @code{step} even though there are other commands whose
1467 names start with @code{s}. You can test abbreviations by using them as
1468 arguments to the @code{help} command.
1470 @cindex repeating commands
1471 @kindex RET @r{(repeat last command)}
1472 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1473 repeat the previous command. Certain commands (for example, @code{run})
1474 will not repeat this way; these are commands whose unintentional
1475 repetition might cause trouble and which you are unlikely to want to
1476 repeat. User-defined commands can disable this feature; see
1477 @ref{Define, dont-repeat}.
1479 The @code{list} and @code{x} commands, when you repeat them with
1480 @key{RET}, construct new arguments rather than repeating
1481 exactly as typed. This permits easy scanning of source or memory.
1483 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1484 output, in a way similar to the common utility @code{more}
1485 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1486 @key{RET} too many in this situation, @value{GDBN} disables command
1487 repetition after any command that generates this sort of display.
1489 @kindex # @r{(a comment)}
1491 Any text from a @kbd{#} to the end of the line is a comment; it does
1492 nothing. This is useful mainly in command files (@pxref{Command
1493 Files,,Command Files}).
1495 @cindex repeating command sequences
1496 @kindex Ctrl-o @r{(operate-and-get-next)}
1497 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1498 commands. This command accepts the current line, like @key{RET}, and
1499 then fetches the next line relative to the current line from the history
1503 @section Command Completion
1506 @cindex word completion
1507 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1508 only one possibility; it can also show you what the valid possibilities
1509 are for the next word in a command, at any time. This works for @value{GDBN}
1510 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1512 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1513 of a word. If there is only one possibility, @value{GDBN} fills in the
1514 word, and waits for you to finish the command (or press @key{RET} to
1515 enter it). For example, if you type
1517 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1518 @c complete accuracy in these examples; space introduced for clarity.
1519 @c If texinfo enhancements make it unnecessary, it would be nice to
1520 @c replace " @key" by "@key" in the following...
1522 (@value{GDBP}) info bre @key{TAB}
1526 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1527 the only @code{info} subcommand beginning with @samp{bre}:
1530 (@value{GDBP}) info breakpoints
1534 You can either press @key{RET} at this point, to run the @code{info
1535 breakpoints} command, or backspace and enter something else, if
1536 @samp{breakpoints} does not look like the command you expected. (If you
1537 were sure you wanted @code{info breakpoints} in the first place, you
1538 might as well just type @key{RET} immediately after @samp{info bre},
1539 to exploit command abbreviations rather than command completion).
1541 If there is more than one possibility for the next word when you press
1542 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1543 characters and try again, or just press @key{TAB} a second time;
1544 @value{GDBN} displays all the possible completions for that word. For
1545 example, you might want to set a breakpoint on a subroutine whose name
1546 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1547 just sounds the bell. Typing @key{TAB} again displays all the
1548 function names in your program that begin with those characters, for
1552 (@value{GDBP}) b make_ @key{TAB}
1553 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1554 make_a_section_from_file make_environ
1555 make_abs_section make_function_type
1556 make_blockvector make_pointer_type
1557 make_cleanup make_reference_type
1558 make_command make_symbol_completion_list
1559 (@value{GDBP}) b make_
1563 After displaying the available possibilities, @value{GDBN} copies your
1564 partial input (@samp{b make_} in the example) so you can finish the
1567 If you just want to see the list of alternatives in the first place, you
1568 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1569 means @kbd{@key{META} ?}. You can type this either by holding down a
1570 key designated as the @key{META} shift on your keyboard (if there is
1571 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1573 @cindex quotes in commands
1574 @cindex completion of quoted strings
1575 Sometimes the string you need, while logically a ``word'', may contain
1576 parentheses or other characters that @value{GDBN} normally excludes from
1577 its notion of a word. To permit word completion to work in this
1578 situation, you may enclose words in @code{'} (single quote marks) in
1579 @value{GDBN} commands.
1581 The most likely situation where you might need this is in typing the
1582 name of a C@t{++} function. This is because C@t{++} allows function
1583 overloading (multiple definitions of the same function, distinguished
1584 by argument type). For example, when you want to set a breakpoint you
1585 may need to distinguish whether you mean the version of @code{name}
1586 that takes an @code{int} parameter, @code{name(int)}, or the version
1587 that takes a @code{float} parameter, @code{name(float)}. To use the
1588 word-completion facilities in this situation, type a single quote
1589 @code{'} at the beginning of the function name. This alerts
1590 @value{GDBN} that it may need to consider more information than usual
1591 when you press @key{TAB} or @kbd{M-?} to request word completion:
1594 (@value{GDBP}) b 'bubble( @kbd{M-?}
1595 bubble(double,double) bubble(int,int)
1596 (@value{GDBP}) b 'bubble(
1599 In some cases, @value{GDBN} can tell that completing a name requires using
1600 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1601 completing as much as it can) if you do not type the quote in the first
1605 (@value{GDBP}) b bub @key{TAB}
1606 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1607 (@value{GDBP}) b 'bubble(
1611 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1612 you have not yet started typing the argument list when you ask for
1613 completion on an overloaded symbol.
1615 For more information about overloaded functions, see @ref{C Plus Plus
1616 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1617 overload-resolution off} to disable overload resolution;
1618 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1620 @cindex completion of structure field names
1621 @cindex structure field name completion
1622 @cindex completion of union field names
1623 @cindex union field name completion
1624 When completing in an expression which looks up a field in a
1625 structure, @value{GDBN} also tries@footnote{The completer can be
1626 confused by certain kinds of invalid expressions. Also, it only
1627 examines the static type of the expression, not the dynamic type.} to
1628 limit completions to the field names available in the type of the
1632 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1633 magic to_fputs to_rewind
1634 to_data to_isatty to_write
1635 to_delete to_put to_write_async_safe
1640 This is because the @code{gdb_stdout} is a variable of the type
1641 @code{struct ui_file} that is defined in @value{GDBN} sources as
1648 ui_file_flush_ftype *to_flush;
1649 ui_file_write_ftype *to_write;
1650 ui_file_write_async_safe_ftype *to_write_async_safe;
1651 ui_file_fputs_ftype *to_fputs;
1652 ui_file_read_ftype *to_read;
1653 ui_file_delete_ftype *to_delete;
1654 ui_file_isatty_ftype *to_isatty;
1655 ui_file_rewind_ftype *to_rewind;
1656 ui_file_put_ftype *to_put;
1663 @section Getting Help
1664 @cindex online documentation
1667 You can always ask @value{GDBN} itself for information on its commands,
1668 using the command @code{help}.
1671 @kindex h @r{(@code{help})}
1674 You can use @code{help} (abbreviated @code{h}) with no arguments to
1675 display a short list of named classes of commands:
1679 List of classes of commands:
1681 aliases -- Aliases of other commands
1682 breakpoints -- Making program stop at certain points
1683 data -- Examining data
1684 files -- Specifying and examining files
1685 internals -- Maintenance commands
1686 obscure -- Obscure features
1687 running -- Running the program
1688 stack -- Examining the stack
1689 status -- Status inquiries
1690 support -- Support facilities
1691 tracepoints -- Tracing of program execution without
1692 stopping the program
1693 user-defined -- User-defined commands
1695 Type "help" followed by a class name for a list of
1696 commands in that class.
1697 Type "help" followed by command name for full
1699 Command name abbreviations are allowed if unambiguous.
1702 @c the above line break eliminates huge line overfull...
1704 @item help @var{class}
1705 Using one of the general help classes as an argument, you can get a
1706 list of the individual commands in that class. For example, here is the
1707 help display for the class @code{status}:
1710 (@value{GDBP}) help status
1715 @c Line break in "show" line falsifies real output, but needed
1716 @c to fit in smallbook page size.
1717 info -- Generic command for showing things
1718 about the program being debugged
1719 show -- Generic command for showing things
1722 Type "help" followed by command name for full
1724 Command name abbreviations are allowed if unambiguous.
1728 @item help @var{command}
1729 With a command name as @code{help} argument, @value{GDBN} displays a
1730 short paragraph on how to use that command.
1733 @item apropos @var{args}
1734 The @code{apropos} command searches through all of the @value{GDBN}
1735 commands, and their documentation, for the regular expression specified in
1736 @var{args}. It prints out all matches found. For example:
1747 alias -- Define a new command that is an alias of an existing command
1748 aliases -- Aliases of other commands
1749 d -- Delete some breakpoints or auto-display expressions
1750 del -- Delete some breakpoints or auto-display expressions
1751 delete -- Delete some breakpoints or auto-display expressions
1756 @item complete @var{args}
1757 The @code{complete @var{args}} command lists all the possible completions
1758 for the beginning of a command. Use @var{args} to specify the beginning of the
1759 command you want completed. For example:
1765 @noindent results in:
1776 @noindent This is intended for use by @sc{gnu} Emacs.
1779 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1780 and @code{show} to inquire about the state of your program, or the state
1781 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1782 manual introduces each of them in the appropriate context. The listings
1783 under @code{info} and under @code{show} in the Index point to
1784 all the sub-commands. @xref{Index}.
1789 @kindex i @r{(@code{info})}
1791 This command (abbreviated @code{i}) is for describing the state of your
1792 program. For example, you can show the arguments passed to a function
1793 with @code{info args}, list the registers currently in use with @code{info
1794 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1795 You can get a complete list of the @code{info} sub-commands with
1796 @w{@code{help info}}.
1800 You can assign the result of an expression to an environment variable with
1801 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1802 @code{set prompt $}.
1806 In contrast to @code{info}, @code{show} is for describing the state of
1807 @value{GDBN} itself.
1808 You can change most of the things you can @code{show}, by using the
1809 related command @code{set}; for example, you can control what number
1810 system is used for displays with @code{set radix}, or simply inquire
1811 which is currently in use with @code{show radix}.
1814 To display all the settable parameters and their current
1815 values, you can use @code{show} with no arguments; you may also use
1816 @code{info set}. Both commands produce the same display.
1817 @c FIXME: "info set" violates the rule that "info" is for state of
1818 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1819 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1823 Here are three miscellaneous @code{show} subcommands, all of which are
1824 exceptional in lacking corresponding @code{set} commands:
1827 @kindex show version
1828 @cindex @value{GDBN} version number
1830 Show what version of @value{GDBN} is running. You should include this
1831 information in @value{GDBN} bug-reports. If multiple versions of
1832 @value{GDBN} are in use at your site, you may need to determine which
1833 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1834 commands are introduced, and old ones may wither away. Also, many
1835 system vendors ship variant versions of @value{GDBN}, and there are
1836 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1837 The version number is the same as the one announced when you start
1840 @kindex show copying
1841 @kindex info copying
1842 @cindex display @value{GDBN} copyright
1845 Display information about permission for copying @value{GDBN}.
1847 @kindex show warranty
1848 @kindex info warranty
1850 @itemx info warranty
1851 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1852 if your version of @value{GDBN} comes with one.
1857 @chapter Running Programs Under @value{GDBN}
1859 When you run a program under @value{GDBN}, you must first generate
1860 debugging information when you compile it.
1862 You may start @value{GDBN} with its arguments, if any, in an environment
1863 of your choice. If you are doing native debugging, you may redirect
1864 your program's input and output, debug an already running process, or
1865 kill a child process.
1868 * Compilation:: Compiling for debugging
1869 * Starting:: Starting your program
1870 * Arguments:: Your program's arguments
1871 * Environment:: Your program's environment
1873 * Working Directory:: Your program's working directory
1874 * Input/Output:: Your program's input and output
1875 * Attach:: Debugging an already-running process
1876 * Kill Process:: Killing the child process
1878 * Inferiors and Programs:: Debugging multiple inferiors and programs
1879 * Threads:: Debugging programs with multiple threads
1880 * Forks:: Debugging forks
1881 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1885 @section Compiling for Debugging
1887 In order to debug a program effectively, you need to generate
1888 debugging information when you compile it. This debugging information
1889 is stored in the object file; it describes the data type of each
1890 variable or function and the correspondence between source line numbers
1891 and addresses in the executable code.
1893 To request debugging information, specify the @samp{-g} option when you run
1896 Programs that are to be shipped to your customers are compiled with
1897 optimizations, using the @samp{-O} compiler option. However, some
1898 compilers are unable to handle the @samp{-g} and @samp{-O} options
1899 together. Using those compilers, you cannot generate optimized
1900 executables containing debugging information.
1902 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1903 without @samp{-O}, making it possible to debug optimized code. We
1904 recommend that you @emph{always} use @samp{-g} whenever you compile a
1905 program. You may think your program is correct, but there is no sense
1906 in pushing your luck. For more information, see @ref{Optimized Code}.
1908 Older versions of the @sc{gnu} C compiler permitted a variant option
1909 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1910 format; if your @sc{gnu} C compiler has this option, do not use it.
1912 @value{GDBN} knows about preprocessor macros and can show you their
1913 expansion (@pxref{Macros}). Most compilers do not include information
1914 about preprocessor macros in the debugging information if you specify
1915 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
1916 the @sc{gnu} C compiler, provides macro information if you are using
1917 the DWARF debugging format, and specify the option @option{-g3}.
1919 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
1920 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}, for more
1921 information on @value{NGCC} options affecting debug information.
1923 You will have the best debugging experience if you use the latest
1924 version of the DWARF debugging format that your compiler supports.
1925 DWARF is currently the most expressive and best supported debugging
1926 format in @value{GDBN}.
1930 @section Starting your Program
1936 @kindex r @r{(@code{run})}
1939 Use the @code{run} command to start your program under @value{GDBN}.
1940 You must first specify the program name (except on VxWorks) with an
1941 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1942 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1943 (@pxref{Files, ,Commands to Specify Files}).
1947 If you are running your program in an execution environment that
1948 supports processes, @code{run} creates an inferior process and makes
1949 that process run your program. In some environments without processes,
1950 @code{run} jumps to the start of your program. Other targets,
1951 like @samp{remote}, are always running. If you get an error
1952 message like this one:
1955 The "remote" target does not support "run".
1956 Try "help target" or "continue".
1960 then use @code{continue} to run your program. You may need @code{load}
1961 first (@pxref{load}).
1963 The execution of a program is affected by certain information it
1964 receives from its superior. @value{GDBN} provides ways to specify this
1965 information, which you must do @emph{before} starting your program. (You
1966 can change it after starting your program, but such changes only affect
1967 your program the next time you start it.) This information may be
1968 divided into four categories:
1971 @item The @emph{arguments.}
1972 Specify the arguments to give your program as the arguments of the
1973 @code{run} command. If a shell is available on your target, the shell
1974 is used to pass the arguments, so that you may use normal conventions
1975 (such as wildcard expansion or variable substitution) in describing
1977 In Unix systems, you can control which shell is used with the
1978 @code{SHELL} environment variable.
1979 @xref{Arguments, ,Your Program's Arguments}.
1981 @item The @emph{environment.}
1982 Your program normally inherits its environment from @value{GDBN}, but you can
1983 use the @value{GDBN} commands @code{set environment} and @code{unset
1984 environment} to change parts of the environment that affect
1985 your program. @xref{Environment, ,Your Program's Environment}.
1987 @item The @emph{working directory.}
1988 Your program inherits its working directory from @value{GDBN}. You can set
1989 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1990 @xref{Working Directory, ,Your Program's Working Directory}.
1992 @item The @emph{standard input and output.}
1993 Your program normally uses the same device for standard input and
1994 standard output as @value{GDBN} is using. You can redirect input and output
1995 in the @code{run} command line, or you can use the @code{tty} command to
1996 set a different device for your program.
1997 @xref{Input/Output, ,Your Program's Input and Output}.
2000 @emph{Warning:} While input and output redirection work, you cannot use
2001 pipes to pass the output of the program you are debugging to another
2002 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2006 When you issue the @code{run} command, your program begins to execute
2007 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2008 of how to arrange for your program to stop. Once your program has
2009 stopped, you may call functions in your program, using the @code{print}
2010 or @code{call} commands. @xref{Data, ,Examining Data}.
2012 If the modification time of your symbol file has changed since the last
2013 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2014 table, and reads it again. When it does this, @value{GDBN} tries to retain
2015 your current breakpoints.
2020 @cindex run to main procedure
2021 The name of the main procedure can vary from language to language.
2022 With C or C@t{++}, the main procedure name is always @code{main}, but
2023 other languages such as Ada do not require a specific name for their
2024 main procedure. The debugger provides a convenient way to start the
2025 execution of the program and to stop at the beginning of the main
2026 procedure, depending on the language used.
2028 The @samp{start} command does the equivalent of setting a temporary
2029 breakpoint at the beginning of the main procedure and then invoking
2030 the @samp{run} command.
2032 @cindex elaboration phase
2033 Some programs contain an @dfn{elaboration} phase where some startup code is
2034 executed before the main procedure is called. This depends on the
2035 languages used to write your program. In C@t{++}, for instance,
2036 constructors for static and global objects are executed before
2037 @code{main} is called. It is therefore possible that the debugger stops
2038 before reaching the main procedure. However, the temporary breakpoint
2039 will remain to halt execution.
2041 Specify the arguments to give to your program as arguments to the
2042 @samp{start} command. These arguments will be given verbatim to the
2043 underlying @samp{run} command. Note that the same arguments will be
2044 reused if no argument is provided during subsequent calls to
2045 @samp{start} or @samp{run}.
2047 It is sometimes necessary to debug the program during elaboration. In
2048 these cases, using the @code{start} command would stop the execution of
2049 your program too late, as the program would have already completed the
2050 elaboration phase. Under these circumstances, insert breakpoints in your
2051 elaboration code before running your program.
2053 @kindex set exec-wrapper
2054 @item set exec-wrapper @var{wrapper}
2055 @itemx show exec-wrapper
2056 @itemx unset exec-wrapper
2057 When @samp{exec-wrapper} is set, the specified wrapper is used to
2058 launch programs for debugging. @value{GDBN} starts your program
2059 with a shell command of the form @kbd{exec @var{wrapper}
2060 @var{program}}. Quoting is added to @var{program} and its
2061 arguments, but not to @var{wrapper}, so you should add quotes if
2062 appropriate for your shell. The wrapper runs until it executes
2063 your program, and then @value{GDBN} takes control.
2065 You can use any program that eventually calls @code{execve} with
2066 its arguments as a wrapper. Several standard Unix utilities do
2067 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2068 with @code{exec "$@@"} will also work.
2070 For example, you can use @code{env} to pass an environment variable to
2071 the debugged program, without setting the variable in your shell's
2075 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2079 This command is available when debugging locally on most targets, excluding
2080 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2082 @kindex set disable-randomization
2083 @item set disable-randomization
2084 @itemx set disable-randomization on
2085 This option (enabled by default in @value{GDBN}) will turn off the native
2086 randomization of the virtual address space of the started program. This option
2087 is useful for multiple debugging sessions to make the execution better
2088 reproducible and memory addresses reusable across debugging sessions.
2090 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2091 On @sc{gnu}/Linux you can get the same behavior using
2094 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2097 @item set disable-randomization off
2098 Leave the behavior of the started executable unchanged. Some bugs rear their
2099 ugly heads only when the program is loaded at certain addresses. If your bug
2100 disappears when you run the program under @value{GDBN}, that might be because
2101 @value{GDBN} by default disables the address randomization on platforms, such
2102 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2103 disable-randomization off} to try to reproduce such elusive bugs.
2105 On targets where it is available, virtual address space randomization
2106 protects the programs against certain kinds of security attacks. In these
2107 cases the attacker needs to know the exact location of a concrete executable
2108 code. Randomizing its location makes it impossible to inject jumps misusing
2109 a code at its expected addresses.
2111 Prelinking shared libraries provides a startup performance advantage but it
2112 makes addresses in these libraries predictable for privileged processes by
2113 having just unprivileged access at the target system. Reading the shared
2114 library binary gives enough information for assembling the malicious code
2115 misusing it. Still even a prelinked shared library can get loaded at a new
2116 random address just requiring the regular relocation process during the
2117 startup. Shared libraries not already prelinked are always loaded at
2118 a randomly chosen address.
2120 Position independent executables (PIE) contain position independent code
2121 similar to the shared libraries and therefore such executables get loaded at
2122 a randomly chosen address upon startup. PIE executables always load even
2123 already prelinked shared libraries at a random address. You can build such
2124 executable using @command{gcc -fPIE -pie}.
2126 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2127 (as long as the randomization is enabled).
2129 @item show disable-randomization
2130 Show the current setting of the explicit disable of the native randomization of
2131 the virtual address space of the started program.
2136 @section Your Program's Arguments
2138 @cindex arguments (to your program)
2139 The arguments to your program can be specified by the arguments of the
2141 They are passed to a shell, which expands wildcard characters and
2142 performs redirection of I/O, and thence to your program. Your
2143 @code{SHELL} environment variable (if it exists) specifies what shell
2144 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2145 the default shell (@file{/bin/sh} on Unix).
2147 On non-Unix systems, the program is usually invoked directly by
2148 @value{GDBN}, which emulates I/O redirection via the appropriate system
2149 calls, and the wildcard characters are expanded by the startup code of
2150 the program, not by the shell.
2152 @code{run} with no arguments uses the same arguments used by the previous
2153 @code{run}, or those set by the @code{set args} command.
2158 Specify the arguments to be used the next time your program is run. If
2159 @code{set args} has no arguments, @code{run} executes your program
2160 with no arguments. Once you have run your program with arguments,
2161 using @code{set args} before the next @code{run} is the only way to run
2162 it again without arguments.
2166 Show the arguments to give your program when it is started.
2170 @section Your Program's Environment
2172 @cindex environment (of your program)
2173 The @dfn{environment} consists of a set of environment variables and
2174 their values. Environment variables conventionally record such things as
2175 your user name, your home directory, your terminal type, and your search
2176 path for programs to run. Usually you set up environment variables with
2177 the shell and they are inherited by all the other programs you run. When
2178 debugging, it can be useful to try running your program with a modified
2179 environment without having to start @value{GDBN} over again.
2183 @item path @var{directory}
2184 Add @var{directory} to the front of the @code{PATH} environment variable
2185 (the search path for executables) that will be passed to your program.
2186 The value of @code{PATH} used by @value{GDBN} does not change.
2187 You may specify several directory names, separated by whitespace or by a
2188 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2189 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2190 is moved to the front, so it is searched sooner.
2192 You can use the string @samp{$cwd} to refer to whatever is the current
2193 working directory at the time @value{GDBN} searches the path. If you
2194 use @samp{.} instead, it refers to the directory where you executed the
2195 @code{path} command. @value{GDBN} replaces @samp{.} in the
2196 @var{directory} argument (with the current path) before adding
2197 @var{directory} to the search path.
2198 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2199 @c document that, since repeating it would be a no-op.
2203 Display the list of search paths for executables (the @code{PATH}
2204 environment variable).
2206 @kindex show environment
2207 @item show environment @r{[}@var{varname}@r{]}
2208 Print the value of environment variable @var{varname} to be given to
2209 your program when it starts. If you do not supply @var{varname},
2210 print the names and values of all environment variables to be given to
2211 your program. You can abbreviate @code{environment} as @code{env}.
2213 @kindex set environment
2214 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2215 Set environment variable @var{varname} to @var{value}. The value
2216 changes for your program only, not for @value{GDBN} itself. @var{value} may
2217 be any string; the values of environment variables are just strings, and
2218 any interpretation is supplied by your program itself. The @var{value}
2219 parameter is optional; if it is eliminated, the variable is set to a
2221 @c "any string" here does not include leading, trailing
2222 @c blanks. Gnu asks: does anyone care?
2224 For example, this command:
2231 tells the debugged program, when subsequently run, that its user is named
2232 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2233 are not actually required.)
2235 @kindex unset environment
2236 @item unset environment @var{varname}
2237 Remove variable @var{varname} from the environment to be passed to your
2238 program. This is different from @samp{set env @var{varname} =};
2239 @code{unset environment} removes the variable from the environment,
2240 rather than assigning it an empty value.
2243 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2245 by your @code{SHELL} environment variable if it exists (or
2246 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2247 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2248 @file{.bashrc} for BASH---any variables you set in that file affect
2249 your program. You may wish to move setting of environment variables to
2250 files that are only run when you sign on, such as @file{.login} or
2253 @node Working Directory
2254 @section Your Program's Working Directory
2256 @cindex working directory (of your program)
2257 Each time you start your program with @code{run}, it inherits its
2258 working directory from the current working directory of @value{GDBN}.
2259 The @value{GDBN} working directory is initially whatever it inherited
2260 from its parent process (typically the shell), but you can specify a new
2261 working directory in @value{GDBN} with the @code{cd} command.
2263 The @value{GDBN} working directory also serves as a default for the commands
2264 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2269 @cindex change working directory
2270 @item cd @var{directory}
2271 Set the @value{GDBN} working directory to @var{directory}.
2275 Print the @value{GDBN} working directory.
2278 It is generally impossible to find the current working directory of
2279 the process being debugged (since a program can change its directory
2280 during its run). If you work on a system where @value{GDBN} is
2281 configured with the @file{/proc} support, you can use the @code{info
2282 proc} command (@pxref{SVR4 Process Information}) to find out the
2283 current working directory of the debuggee.
2286 @section Your Program's Input and Output
2291 By default, the program you run under @value{GDBN} does input and output to
2292 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2293 to its own terminal modes to interact with you, but it records the terminal
2294 modes your program was using and switches back to them when you continue
2295 running your program.
2298 @kindex info terminal
2300 Displays information recorded by @value{GDBN} about the terminal modes your
2304 You can redirect your program's input and/or output using shell
2305 redirection with the @code{run} command. For example,
2312 starts your program, diverting its output to the file @file{outfile}.
2315 @cindex controlling terminal
2316 Another way to specify where your program should do input and output is
2317 with the @code{tty} command. This command accepts a file name as
2318 argument, and causes this file to be the default for future @code{run}
2319 commands. It also resets the controlling terminal for the child
2320 process, for future @code{run} commands. For example,
2327 directs that processes started with subsequent @code{run} commands
2328 default to do input and output on the terminal @file{/dev/ttyb} and have
2329 that as their controlling terminal.
2331 An explicit redirection in @code{run} overrides the @code{tty} command's
2332 effect on the input/output device, but not its effect on the controlling
2335 When you use the @code{tty} command or redirect input in the @code{run}
2336 command, only the input @emph{for your program} is affected. The input
2337 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2338 for @code{set inferior-tty}.
2340 @cindex inferior tty
2341 @cindex set inferior controlling terminal
2342 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2343 display the name of the terminal that will be used for future runs of your
2347 @item set inferior-tty /dev/ttyb
2348 @kindex set inferior-tty
2349 Set the tty for the program being debugged to /dev/ttyb.
2351 @item show inferior-tty
2352 @kindex show inferior-tty
2353 Show the current tty for the program being debugged.
2357 @section Debugging an Already-running Process
2362 @item attach @var{process-id}
2363 This command attaches to a running process---one that was started
2364 outside @value{GDBN}. (@code{info files} shows your active
2365 targets.) The command takes as argument a process ID. The usual way to
2366 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2367 or with the @samp{jobs -l} shell command.
2369 @code{attach} does not repeat if you press @key{RET} a second time after
2370 executing the command.
2373 To use @code{attach}, your program must be running in an environment
2374 which supports processes; for example, @code{attach} does not work for
2375 programs on bare-board targets that lack an operating system. You must
2376 also have permission to send the process a signal.
2378 When you use @code{attach}, the debugger finds the program running in
2379 the process first by looking in the current working directory, then (if
2380 the program is not found) by using the source file search path
2381 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2382 the @code{file} command to load the program. @xref{Files, ,Commands to
2385 The first thing @value{GDBN} does after arranging to debug the specified
2386 process is to stop it. You can examine and modify an attached process
2387 with all the @value{GDBN} commands that are ordinarily available when
2388 you start processes with @code{run}. You can insert breakpoints; you
2389 can step and continue; you can modify storage. If you would rather the
2390 process continue running, you may use the @code{continue} command after
2391 attaching @value{GDBN} to the process.
2396 When you have finished debugging the attached process, you can use the
2397 @code{detach} command to release it from @value{GDBN} control. Detaching
2398 the process continues its execution. After the @code{detach} command,
2399 that process and @value{GDBN} become completely independent once more, and you
2400 are ready to @code{attach} another process or start one with @code{run}.
2401 @code{detach} does not repeat if you press @key{RET} again after
2402 executing the command.
2405 If you exit @value{GDBN} while you have an attached process, you detach
2406 that process. If you use the @code{run} command, you kill that process.
2407 By default, @value{GDBN} asks for confirmation if you try to do either of these
2408 things; you can control whether or not you need to confirm by using the
2409 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2413 @section Killing the Child Process
2418 Kill the child process in which your program is running under @value{GDBN}.
2421 This command is useful if you wish to debug a core dump instead of a
2422 running process. @value{GDBN} ignores any core dump file while your program
2425 On some operating systems, a program cannot be executed outside @value{GDBN}
2426 while you have breakpoints set on it inside @value{GDBN}. You can use the
2427 @code{kill} command in this situation to permit running your program
2428 outside the debugger.
2430 The @code{kill} command is also useful if you wish to recompile and
2431 relink your program, since on many systems it is impossible to modify an
2432 executable file while it is running in a process. In this case, when you
2433 next type @code{run}, @value{GDBN} notices that the file has changed, and
2434 reads the symbol table again (while trying to preserve your current
2435 breakpoint settings).
2437 @node Inferiors and Programs
2438 @section Debugging Multiple Inferiors and Programs
2440 @value{GDBN} lets you run and debug multiple programs in a single
2441 session. In addition, @value{GDBN} on some systems may let you run
2442 several programs simultaneously (otherwise you have to exit from one
2443 before starting another). In the most general case, you can have
2444 multiple threads of execution in each of multiple processes, launched
2445 from multiple executables.
2448 @value{GDBN} represents the state of each program execution with an
2449 object called an @dfn{inferior}. An inferior typically corresponds to
2450 a process, but is more general and applies also to targets that do not
2451 have processes. Inferiors may be created before a process runs, and
2452 may be retained after a process exits. Inferiors have unique
2453 identifiers that are different from process ids. Usually each
2454 inferior will also have its own distinct address space, although some
2455 embedded targets may have several inferiors running in different parts
2456 of a single address space. Each inferior may in turn have multiple
2457 threads running in it.
2459 To find out what inferiors exist at any moment, use @w{@code{info
2463 @kindex info inferiors
2464 @item info inferiors
2465 Print a list of all inferiors currently being managed by @value{GDBN}.
2467 @value{GDBN} displays for each inferior (in this order):
2471 the inferior number assigned by @value{GDBN}
2474 the target system's inferior identifier
2477 the name of the executable the inferior is running.
2482 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2483 indicates the current inferior.
2487 @c end table here to get a little more width for example
2490 (@value{GDBP}) info inferiors
2491 Num Description Executable
2492 2 process 2307 hello
2493 * 1 process 3401 goodbye
2496 To switch focus between inferiors, use the @code{inferior} command:
2499 @kindex inferior @var{infno}
2500 @item inferior @var{infno}
2501 Make inferior number @var{infno} the current inferior. The argument
2502 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2503 in the first field of the @samp{info inferiors} display.
2507 You can get multiple executables into a debugging session via the
2508 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2509 systems @value{GDBN} can add inferiors to the debug session
2510 automatically by following calls to @code{fork} and @code{exec}. To
2511 remove inferiors from the debugging session use the
2512 @w{@code{remove-inferiors}} command.
2515 @kindex add-inferior
2516 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2517 Adds @var{n} inferiors to be run using @var{executable} as the
2518 executable. @var{n} defaults to 1. If no executable is specified,
2519 the inferiors begins empty, with no program. You can still assign or
2520 change the program assigned to the inferior at any time by using the
2521 @code{file} command with the executable name as its argument.
2523 @kindex clone-inferior
2524 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2525 Adds @var{n} inferiors ready to execute the same program as inferior
2526 @var{infno}. @var{n} defaults to 1. @var{infno} defaults to the
2527 number of the current inferior. This is a convenient command when you
2528 want to run another instance of the inferior you are debugging.
2531 (@value{GDBP}) info inferiors
2532 Num Description Executable
2533 * 1 process 29964 helloworld
2534 (@value{GDBP}) clone-inferior
2537 (@value{GDBP}) info inferiors
2538 Num Description Executable
2540 * 1 process 29964 helloworld
2543 You can now simply switch focus to inferior 2 and run it.
2545 @kindex remove-inferiors
2546 @item remove-inferiors @var{infno}@dots{}
2547 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2548 possible to remove an inferior that is running with this command. For
2549 those, use the @code{kill} or @code{detach} command first.
2553 To quit debugging one of the running inferiors that is not the current
2554 inferior, you can either detach from it by using the @w{@code{detach
2555 inferior}} command (allowing it to run independently), or kill it
2556 using the @w{@code{kill inferiors}} command:
2559 @kindex detach inferiors @var{infno}@dots{}
2560 @item detach inferior @var{infno}@dots{}
2561 Detach from the inferior or inferiors identified by @value{GDBN}
2562 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2563 still stays on the list of inferiors shown by @code{info inferiors},
2564 but its Description will show @samp{<null>}.
2566 @kindex kill inferiors @var{infno}@dots{}
2567 @item kill inferiors @var{infno}@dots{}
2568 Kill the inferior or inferiors identified by @value{GDBN} inferior
2569 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2570 stays on the list of inferiors shown by @code{info inferiors}, but its
2571 Description will show @samp{<null>}.
2574 After the successful completion of a command such as @code{detach},
2575 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2576 a normal process exit, the inferior is still valid and listed with
2577 @code{info inferiors}, ready to be restarted.
2580 To be notified when inferiors are started or exit under @value{GDBN}'s
2581 control use @w{@code{set print inferior-events}}:
2584 @kindex set print inferior-events
2585 @cindex print messages on inferior start and exit
2586 @item set print inferior-events
2587 @itemx set print inferior-events on
2588 @itemx set print inferior-events off
2589 The @code{set print inferior-events} command allows you to enable or
2590 disable printing of messages when @value{GDBN} notices that new
2591 inferiors have started or that inferiors have exited or have been
2592 detached. By default, these messages will not be printed.
2594 @kindex show print inferior-events
2595 @item show print inferior-events
2596 Show whether messages will be printed when @value{GDBN} detects that
2597 inferiors have started, exited or have been detached.
2600 Many commands will work the same with multiple programs as with a
2601 single program: e.g., @code{print myglobal} will simply display the
2602 value of @code{myglobal} in the current inferior.
2605 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2606 get more info about the relationship of inferiors, programs, address
2607 spaces in a debug session. You can do that with the @w{@code{maint
2608 info program-spaces}} command.
2611 @kindex maint info program-spaces
2612 @item maint info program-spaces
2613 Print a list of all program spaces currently being managed by
2616 @value{GDBN} displays for each program space (in this order):
2620 the program space number assigned by @value{GDBN}
2623 the name of the executable loaded into the program space, with e.g.,
2624 the @code{file} command.
2629 An asterisk @samp{*} preceding the @value{GDBN} program space number
2630 indicates the current program space.
2632 In addition, below each program space line, @value{GDBN} prints extra
2633 information that isn't suitable to display in tabular form. For
2634 example, the list of inferiors bound to the program space.
2637 (@value{GDBP}) maint info program-spaces
2640 Bound inferiors: ID 1 (process 21561)
2644 Here we can see that no inferior is running the program @code{hello},
2645 while @code{process 21561} is running the program @code{goodbye}. On
2646 some targets, it is possible that multiple inferiors are bound to the
2647 same program space. The most common example is that of debugging both
2648 the parent and child processes of a @code{vfork} call. For example,
2651 (@value{GDBP}) maint info program-spaces
2654 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2657 Here, both inferior 2 and inferior 1 are running in the same program
2658 space as a result of inferior 1 having executed a @code{vfork} call.
2662 @section Debugging Programs with Multiple Threads
2664 @cindex threads of execution
2665 @cindex multiple threads
2666 @cindex switching threads
2667 In some operating systems, such as HP-UX and Solaris, a single program
2668 may have more than one @dfn{thread} of execution. The precise semantics
2669 of threads differ from one operating system to another, but in general
2670 the threads of a single program are akin to multiple processes---except
2671 that they share one address space (that is, they can all examine and
2672 modify the same variables). On the other hand, each thread has its own
2673 registers and execution stack, and perhaps private memory.
2675 @value{GDBN} provides these facilities for debugging multi-thread
2679 @item automatic notification of new threads
2680 @item @samp{thread @var{threadno}}, a command to switch among threads
2681 @item @samp{info threads}, a command to inquire about existing threads
2682 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2683 a command to apply a command to a list of threads
2684 @item thread-specific breakpoints
2685 @item @samp{set print thread-events}, which controls printing of
2686 messages on thread start and exit.
2687 @item @samp{set libthread-db-search-path @var{path}}, which lets
2688 the user specify which @code{libthread_db} to use if the default choice
2689 isn't compatible with the program.
2693 @emph{Warning:} These facilities are not yet available on every
2694 @value{GDBN} configuration where the operating system supports threads.
2695 If your @value{GDBN} does not support threads, these commands have no
2696 effect. For example, a system without thread support shows no output
2697 from @samp{info threads}, and always rejects the @code{thread} command,
2701 (@value{GDBP}) info threads
2702 (@value{GDBP}) thread 1
2703 Thread ID 1 not known. Use the "info threads" command to
2704 see the IDs of currently known threads.
2706 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2707 @c doesn't support threads"?
2710 @cindex focus of debugging
2711 @cindex current thread
2712 The @value{GDBN} thread debugging facility allows you to observe all
2713 threads while your program runs---but whenever @value{GDBN} takes
2714 control, one thread in particular is always the focus of debugging.
2715 This thread is called the @dfn{current thread}. Debugging commands show
2716 program information from the perspective of the current thread.
2718 @cindex @code{New} @var{systag} message
2719 @cindex thread identifier (system)
2720 @c FIXME-implementors!! It would be more helpful if the [New...] message
2721 @c included GDB's numeric thread handle, so you could just go to that
2722 @c thread without first checking `info threads'.
2723 Whenever @value{GDBN} detects a new thread in your program, it displays
2724 the target system's identification for the thread with a message in the
2725 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2726 whose form varies depending on the particular system. For example, on
2727 @sc{gnu}/Linux, you might see
2730 [New Thread 0x41e02940 (LWP 25582)]
2734 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2735 the @var{systag} is simply something like @samp{process 368}, with no
2738 @c FIXME!! (1) Does the [New...] message appear even for the very first
2739 @c thread of a program, or does it only appear for the
2740 @c second---i.e.@: when it becomes obvious we have a multithread
2742 @c (2) *Is* there necessarily a first thread always? Or do some
2743 @c multithread systems permit starting a program with multiple
2744 @c threads ab initio?
2746 @cindex thread number
2747 @cindex thread identifier (GDB)
2748 For debugging purposes, @value{GDBN} associates its own thread
2749 number---always a single integer---with each thread in your program.
2752 @kindex info threads
2753 @item info threads @r{[}@var{id}@dots{}@r{]}
2754 Display a summary of all threads currently in your program. Optional
2755 argument @var{id}@dots{} is one or more thread ids separated by spaces, and
2756 means to print information only about the specified thread or threads.
2757 @value{GDBN} displays for each thread (in this order):
2761 the thread number assigned by @value{GDBN}
2764 the target system's thread identifier (@var{systag})
2767 the thread's name, if one is known. A thread can either be named by
2768 the user (see @code{thread name}, below), or, in some cases, by the
2772 the current stack frame summary for that thread
2776 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2777 indicates the current thread.
2781 @c end table here to get a little more width for example
2784 (@value{GDBP}) info threads
2786 3 process 35 thread 27 0x34e5 in sigpause ()
2787 2 process 35 thread 23 0x34e5 in sigpause ()
2788 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2792 On Solaris, you can display more information about user threads with a
2793 Solaris-specific command:
2796 @item maint info sol-threads
2797 @kindex maint info sol-threads
2798 @cindex thread info (Solaris)
2799 Display info on Solaris user threads.
2803 @kindex thread @var{threadno}
2804 @item thread @var{threadno}
2805 Make thread number @var{threadno} the current thread. The command
2806 argument @var{threadno} is the internal @value{GDBN} thread number, as
2807 shown in the first field of the @samp{info threads} display.
2808 @value{GDBN} responds by displaying the system identifier of the thread
2809 you selected, and its current stack frame summary:
2812 (@value{GDBP}) thread 2
2813 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
2814 #0 some_function (ignore=0x0) at example.c:8
2815 8 printf ("hello\n");
2819 As with the @samp{[New @dots{}]} message, the form of the text after
2820 @samp{Switching to} depends on your system's conventions for identifying
2823 @vindex $_thread@r{, convenience variable}
2824 The debugger convenience variable @samp{$_thread} contains the number
2825 of the current thread. You may find this useful in writing breakpoint
2826 conditional expressions, command scripts, and so forth. See
2827 @xref{Convenience Vars,, Convenience Variables}, for general
2828 information on convenience variables.
2830 @kindex thread apply
2831 @cindex apply command to several threads
2832 @item thread apply [@var{threadno} | all] @var{command}
2833 The @code{thread apply} command allows you to apply the named
2834 @var{command} to one or more threads. Specify the numbers of the
2835 threads that you want affected with the command argument
2836 @var{threadno}. It can be a single thread number, one of the numbers
2837 shown in the first field of the @samp{info threads} display; or it
2838 could be a range of thread numbers, as in @code{2-4}. To apply a
2839 command to all threads, type @kbd{thread apply all @var{command}}.
2842 @cindex name a thread
2843 @item thread name [@var{name}]
2844 This command assigns a name to the current thread. If no argument is
2845 given, any existing user-specified name is removed. The thread name
2846 appears in the @samp{info threads} display.
2848 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
2849 determine the name of the thread as given by the OS. On these
2850 systems, a name specified with @samp{thread name} will override the
2851 system-give name, and removing the user-specified name will cause
2852 @value{GDBN} to once again display the system-specified name.
2855 @cindex search for a thread
2856 @item thread find [@var{regexp}]
2857 Search for and display thread ids whose name or @var{systag}
2858 matches the supplied regular expression.
2860 As well as being the complement to the @samp{thread name} command,
2861 this command also allows you to identify a thread by its target
2862 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
2866 (@value{GDBN}) thread find 26688
2867 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
2868 (@value{GDBN}) info thread 4
2870 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
2873 @kindex set print thread-events
2874 @cindex print messages on thread start and exit
2875 @item set print thread-events
2876 @itemx set print thread-events on
2877 @itemx set print thread-events off
2878 The @code{set print thread-events} command allows you to enable or
2879 disable printing of messages when @value{GDBN} notices that new threads have
2880 started or that threads have exited. By default, these messages will
2881 be printed if detection of these events is supported by the target.
2882 Note that these messages cannot be disabled on all targets.
2884 @kindex show print thread-events
2885 @item show print thread-events
2886 Show whether messages will be printed when @value{GDBN} detects that threads
2887 have started and exited.
2890 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2891 more information about how @value{GDBN} behaves when you stop and start
2892 programs with multiple threads.
2894 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2895 watchpoints in programs with multiple threads.
2897 @anchor{set libthread-db-search-path}
2899 @kindex set libthread-db-search-path
2900 @cindex search path for @code{libthread_db}
2901 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2902 If this variable is set, @var{path} is a colon-separated list of
2903 directories @value{GDBN} will use to search for @code{libthread_db}.
2904 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2905 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
2906 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
2909 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2910 @code{libthread_db} library to obtain information about threads in the
2911 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2912 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
2913 specific thread debugging library loading is enabled
2914 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
2916 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
2917 refers to the default system directories that are
2918 normally searched for loading shared libraries. The @samp{$sdir} entry
2919 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
2920 (@pxref{libthread_db.so.1 file}).
2922 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
2923 refers to the directory from which @code{libpthread}
2924 was loaded in the inferior process.
2926 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2927 @value{GDBN} attempts to initialize it with the current inferior process.
2928 If this initialization fails (which could happen because of a version
2929 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2930 will unload @code{libthread_db}, and continue with the next directory.
2931 If none of @code{libthread_db} libraries initialize successfully,
2932 @value{GDBN} will issue a warning and thread debugging will be disabled.
2934 Setting @code{libthread-db-search-path} is currently implemented
2935 only on some platforms.
2937 @kindex show libthread-db-search-path
2938 @item show libthread-db-search-path
2939 Display current libthread_db search path.
2941 @kindex set debug libthread-db
2942 @kindex show debug libthread-db
2943 @cindex debugging @code{libthread_db}
2944 @item set debug libthread-db
2945 @itemx show debug libthread-db
2946 Turns on or off display of @code{libthread_db}-related events.
2947 Use @code{1} to enable, @code{0} to disable.
2951 @section Debugging Forks
2953 @cindex fork, debugging programs which call
2954 @cindex multiple processes
2955 @cindex processes, multiple
2956 On most systems, @value{GDBN} has no special support for debugging
2957 programs which create additional processes using the @code{fork}
2958 function. When a program forks, @value{GDBN} will continue to debug the
2959 parent process and the child process will run unimpeded. If you have
2960 set a breakpoint in any code which the child then executes, the child
2961 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2962 will cause it to terminate.
2964 However, if you want to debug the child process there is a workaround
2965 which isn't too painful. Put a call to @code{sleep} in the code which
2966 the child process executes after the fork. It may be useful to sleep
2967 only if a certain environment variable is set, or a certain file exists,
2968 so that the delay need not occur when you don't want to run @value{GDBN}
2969 on the child. While the child is sleeping, use the @code{ps} program to
2970 get its process ID. Then tell @value{GDBN} (a new invocation of
2971 @value{GDBN} if you are also debugging the parent process) to attach to
2972 the child process (@pxref{Attach}). From that point on you can debug
2973 the child process just like any other process which you attached to.
2975 On some systems, @value{GDBN} provides support for debugging programs that
2976 create additional processes using the @code{fork} or @code{vfork} functions.
2977 Currently, the only platforms with this feature are HP-UX (11.x and later
2978 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2980 By default, when a program forks, @value{GDBN} will continue to debug
2981 the parent process and the child process will run unimpeded.
2983 If you want to follow the child process instead of the parent process,
2984 use the command @w{@code{set follow-fork-mode}}.
2987 @kindex set follow-fork-mode
2988 @item set follow-fork-mode @var{mode}
2989 Set the debugger response to a program call of @code{fork} or
2990 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2991 process. The @var{mode} argument can be:
2995 The original process is debugged after a fork. The child process runs
2996 unimpeded. This is the default.
2999 The new process is debugged after a fork. The parent process runs
3004 @kindex show follow-fork-mode
3005 @item show follow-fork-mode
3006 Display the current debugger response to a @code{fork} or @code{vfork} call.
3009 @cindex debugging multiple processes
3010 On Linux, if you want to debug both the parent and child processes, use the
3011 command @w{@code{set detach-on-fork}}.
3014 @kindex set detach-on-fork
3015 @item set detach-on-fork @var{mode}
3016 Tells gdb whether to detach one of the processes after a fork, or
3017 retain debugger control over them both.
3021 The child process (or parent process, depending on the value of
3022 @code{follow-fork-mode}) will be detached and allowed to run
3023 independently. This is the default.
3026 Both processes will be held under the control of @value{GDBN}.
3027 One process (child or parent, depending on the value of
3028 @code{follow-fork-mode}) is debugged as usual, while the other
3033 @kindex show detach-on-fork
3034 @item show detach-on-fork
3035 Show whether detach-on-fork mode is on/off.
3038 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3039 will retain control of all forked processes (including nested forks).
3040 You can list the forked processes under the control of @value{GDBN} by
3041 using the @w{@code{info inferiors}} command, and switch from one fork
3042 to another by using the @code{inferior} command (@pxref{Inferiors and
3043 Programs, ,Debugging Multiple Inferiors and Programs}).
3045 To quit debugging one of the forked processes, you can either detach
3046 from it by using the @w{@code{detach inferiors}} command (allowing it
3047 to run independently), or kill it using the @w{@code{kill inferiors}}
3048 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3051 If you ask to debug a child process and a @code{vfork} is followed by an
3052 @code{exec}, @value{GDBN} executes the new target up to the first
3053 breakpoint in the new target. If you have a breakpoint set on
3054 @code{main} in your original program, the breakpoint will also be set on
3055 the child process's @code{main}.
3057 On some systems, when a child process is spawned by @code{vfork}, you
3058 cannot debug the child or parent until an @code{exec} call completes.
3060 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3061 call executes, the new target restarts. To restart the parent
3062 process, use the @code{file} command with the parent executable name
3063 as its argument. By default, after an @code{exec} call executes,
3064 @value{GDBN} discards the symbols of the previous executable image.
3065 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3069 @kindex set follow-exec-mode
3070 @item set follow-exec-mode @var{mode}
3072 Set debugger response to a program call of @code{exec}. An
3073 @code{exec} call replaces the program image of a process.
3075 @code{follow-exec-mode} can be:
3079 @value{GDBN} creates a new inferior and rebinds the process to this
3080 new inferior. The program the process was running before the
3081 @code{exec} call can be restarted afterwards by restarting the
3087 (@value{GDBP}) info inferiors
3089 Id Description Executable
3092 process 12020 is executing new program: prog2
3093 Program exited normally.
3094 (@value{GDBP}) info inferiors
3095 Id Description Executable
3101 @value{GDBN} keeps the process bound to the same inferior. The new
3102 executable image replaces the previous executable loaded in the
3103 inferior. Restarting the inferior after the @code{exec} call, with
3104 e.g., the @code{run} command, restarts the executable the process was
3105 running after the @code{exec} call. This is the default mode.
3110 (@value{GDBP}) info inferiors
3111 Id Description Executable
3114 process 12020 is executing new program: prog2
3115 Program exited normally.
3116 (@value{GDBP}) info inferiors
3117 Id Description Executable
3124 You can use the @code{catch} command to make @value{GDBN} stop whenever
3125 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3126 Catchpoints, ,Setting Catchpoints}.
3128 @node Checkpoint/Restart
3129 @section Setting a @emph{Bookmark} to Return to Later
3134 @cindex snapshot of a process
3135 @cindex rewind program state
3137 On certain operating systems@footnote{Currently, only
3138 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3139 program's state, called a @dfn{checkpoint}, and come back to it
3142 Returning to a checkpoint effectively undoes everything that has
3143 happened in the program since the @code{checkpoint} was saved. This
3144 includes changes in memory, registers, and even (within some limits)
3145 system state. Effectively, it is like going back in time to the
3146 moment when the checkpoint was saved.
3148 Thus, if you're stepping thru a program and you think you're
3149 getting close to the point where things go wrong, you can save
3150 a checkpoint. Then, if you accidentally go too far and miss
3151 the critical statement, instead of having to restart your program
3152 from the beginning, you can just go back to the checkpoint and
3153 start again from there.
3155 This can be especially useful if it takes a lot of time or
3156 steps to reach the point where you think the bug occurs.
3158 To use the @code{checkpoint}/@code{restart} method of debugging:
3163 Save a snapshot of the debugged program's current execution state.
3164 The @code{checkpoint} command takes no arguments, but each checkpoint
3165 is assigned a small integer id, similar to a breakpoint id.
3167 @kindex info checkpoints
3168 @item info checkpoints
3169 List the checkpoints that have been saved in the current debugging
3170 session. For each checkpoint, the following information will be
3177 @item Source line, or label
3180 @kindex restart @var{checkpoint-id}
3181 @item restart @var{checkpoint-id}
3182 Restore the program state that was saved as checkpoint number
3183 @var{checkpoint-id}. All program variables, registers, stack frames
3184 etc.@: will be returned to the values that they had when the checkpoint
3185 was saved. In essence, gdb will ``wind back the clock'' to the point
3186 in time when the checkpoint was saved.
3188 Note that breakpoints, @value{GDBN} variables, command history etc.
3189 are not affected by restoring a checkpoint. In general, a checkpoint
3190 only restores things that reside in the program being debugged, not in
3193 @kindex delete checkpoint @var{checkpoint-id}
3194 @item delete checkpoint @var{checkpoint-id}
3195 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3199 Returning to a previously saved checkpoint will restore the user state
3200 of the program being debugged, plus a significant subset of the system
3201 (OS) state, including file pointers. It won't ``un-write'' data from
3202 a file, but it will rewind the file pointer to the previous location,
3203 so that the previously written data can be overwritten. For files
3204 opened in read mode, the pointer will also be restored so that the
3205 previously read data can be read again.
3207 Of course, characters that have been sent to a printer (or other
3208 external device) cannot be ``snatched back'', and characters received
3209 from eg.@: a serial device can be removed from internal program buffers,
3210 but they cannot be ``pushed back'' into the serial pipeline, ready to
3211 be received again. Similarly, the actual contents of files that have
3212 been changed cannot be restored (at this time).
3214 However, within those constraints, you actually can ``rewind'' your
3215 program to a previously saved point in time, and begin debugging it
3216 again --- and you can change the course of events so as to debug a
3217 different execution path this time.
3219 @cindex checkpoints and process id
3220 Finally, there is one bit of internal program state that will be
3221 different when you return to a checkpoint --- the program's process
3222 id. Each checkpoint will have a unique process id (or @var{pid}),
3223 and each will be different from the program's original @var{pid}.
3224 If your program has saved a local copy of its process id, this could
3225 potentially pose a problem.
3227 @subsection A Non-obvious Benefit of Using Checkpoints
3229 On some systems such as @sc{gnu}/Linux, address space randomization
3230 is performed on new processes for security reasons. This makes it
3231 difficult or impossible to set a breakpoint, or watchpoint, on an
3232 absolute address if you have to restart the program, since the
3233 absolute location of a symbol will change from one execution to the
3236 A checkpoint, however, is an @emph{identical} copy of a process.
3237 Therefore if you create a checkpoint at (eg.@:) the start of main,
3238 and simply return to that checkpoint instead of restarting the
3239 process, you can avoid the effects of address randomization and
3240 your symbols will all stay in the same place.
3243 @chapter Stopping and Continuing
3245 The principal purposes of using a debugger are so that you can stop your
3246 program before it terminates; or so that, if your program runs into
3247 trouble, you can investigate and find out why.
3249 Inside @value{GDBN}, your program may stop for any of several reasons,
3250 such as a signal, a breakpoint, or reaching a new line after a
3251 @value{GDBN} command such as @code{step}. You may then examine and
3252 change variables, set new breakpoints or remove old ones, and then
3253 continue execution. Usually, the messages shown by @value{GDBN} provide
3254 ample explanation of the status of your program---but you can also
3255 explicitly request this information at any time.
3258 @kindex info program
3260 Display information about the status of your program: whether it is
3261 running or not, what process it is, and why it stopped.
3265 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3266 * Continuing and Stepping:: Resuming execution
3267 * Skipping Over Functions and Files::
3268 Skipping over functions and files
3270 * Thread Stops:: Stopping and starting multi-thread programs
3274 @section Breakpoints, Watchpoints, and Catchpoints
3277 A @dfn{breakpoint} makes your program stop whenever a certain point in
3278 the program is reached. For each breakpoint, you can add conditions to
3279 control in finer detail whether your program stops. You can set
3280 breakpoints with the @code{break} command and its variants (@pxref{Set
3281 Breaks, ,Setting Breakpoints}), to specify the place where your program
3282 should stop by line number, function name or exact address in the
3285 On some systems, you can set breakpoints in shared libraries before
3286 the executable is run. There is a minor limitation on HP-UX systems:
3287 you must wait until the executable is run in order to set breakpoints
3288 in shared library routines that are not called directly by the program
3289 (for example, routines that are arguments in a @code{pthread_create}
3293 @cindex data breakpoints
3294 @cindex memory tracing
3295 @cindex breakpoint on memory address
3296 @cindex breakpoint on variable modification
3297 A @dfn{watchpoint} is a special breakpoint that stops your program
3298 when the value of an expression changes. The expression may be a value
3299 of a variable, or it could involve values of one or more variables
3300 combined by operators, such as @samp{a + b}. This is sometimes called
3301 @dfn{data breakpoints}. You must use a different command to set
3302 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3303 from that, you can manage a watchpoint like any other breakpoint: you
3304 enable, disable, and delete both breakpoints and watchpoints using the
3307 You can arrange to have values from your program displayed automatically
3308 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3312 @cindex breakpoint on events
3313 A @dfn{catchpoint} is another special breakpoint that stops your program
3314 when a certain kind of event occurs, such as the throwing of a C@t{++}
3315 exception or the loading of a library. As with watchpoints, you use a
3316 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3317 Catchpoints}), but aside from that, you can manage a catchpoint like any
3318 other breakpoint. (To stop when your program receives a signal, use the
3319 @code{handle} command; see @ref{Signals, ,Signals}.)
3321 @cindex breakpoint numbers
3322 @cindex numbers for breakpoints
3323 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3324 catchpoint when you create it; these numbers are successive integers
3325 starting with one. In many of the commands for controlling various
3326 features of breakpoints you use the breakpoint number to say which
3327 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3328 @dfn{disabled}; if disabled, it has no effect on your program until you
3331 @cindex breakpoint ranges
3332 @cindex ranges of breakpoints
3333 Some @value{GDBN} commands accept a range of breakpoints on which to
3334 operate. A breakpoint range is either a single breakpoint number, like
3335 @samp{5}, or two such numbers, in increasing order, separated by a
3336 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3337 all breakpoints in that range are operated on.
3340 * Set Breaks:: Setting breakpoints
3341 * Set Watchpoints:: Setting watchpoints
3342 * Set Catchpoints:: Setting catchpoints
3343 * Delete Breaks:: Deleting breakpoints
3344 * Disabling:: Disabling breakpoints
3345 * Conditions:: Break conditions
3346 * Break Commands:: Breakpoint command lists
3347 * Dynamic Printf:: Dynamic printf
3348 * Save Breakpoints:: How to save breakpoints in a file
3349 * Static Probe Points:: Listing static probe points
3350 * Error in Breakpoints:: ``Cannot insert breakpoints''
3351 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3355 @subsection Setting Breakpoints
3357 @c FIXME LMB what does GDB do if no code on line of breakpt?
3358 @c consider in particular declaration with/without initialization.
3360 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3363 @kindex b @r{(@code{break})}
3364 @vindex $bpnum@r{, convenience variable}
3365 @cindex latest breakpoint
3366 Breakpoints are set with the @code{break} command (abbreviated
3367 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3368 number of the breakpoint you've set most recently; see @ref{Convenience
3369 Vars,, Convenience Variables}, for a discussion of what you can do with
3370 convenience variables.
3373 @item break @var{location}
3374 Set a breakpoint at the given @var{location}, which can specify a
3375 function name, a line number, or an address of an instruction.
3376 (@xref{Specify Location}, for a list of all the possible ways to
3377 specify a @var{location}.) The breakpoint will stop your program just
3378 before it executes any of the code in the specified @var{location}.
3380 When using source languages that permit overloading of symbols, such as
3381 C@t{++}, a function name may refer to more than one possible place to break.
3382 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3385 It is also possible to insert a breakpoint that will stop the program
3386 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3387 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3390 When called without any arguments, @code{break} sets a breakpoint at
3391 the next instruction to be executed in the selected stack frame
3392 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3393 innermost, this makes your program stop as soon as control
3394 returns to that frame. This is similar to the effect of a
3395 @code{finish} command in the frame inside the selected frame---except
3396 that @code{finish} does not leave an active breakpoint. If you use
3397 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3398 the next time it reaches the current location; this may be useful
3401 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3402 least one instruction has been executed. If it did not do this, you
3403 would be unable to proceed past a breakpoint without first disabling the
3404 breakpoint. This rule applies whether or not the breakpoint already
3405 existed when your program stopped.
3407 @item break @dots{} if @var{cond}
3408 Set a breakpoint with condition @var{cond}; evaluate the expression
3409 @var{cond} each time the breakpoint is reached, and stop only if the
3410 value is nonzero---that is, if @var{cond} evaluates as true.
3411 @samp{@dots{}} stands for one of the possible arguments described
3412 above (or no argument) specifying where to break. @xref{Conditions,
3413 ,Break Conditions}, for more information on breakpoint conditions.
3416 @item tbreak @var{args}
3417 Set a breakpoint enabled only for one stop. @var{args} are the
3418 same as for the @code{break} command, and the breakpoint is set in the same
3419 way, but the breakpoint is automatically deleted after the first time your
3420 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3423 @cindex hardware breakpoints
3424 @item hbreak @var{args}
3425 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3426 @code{break} command and the breakpoint is set in the same way, but the
3427 breakpoint requires hardware support and some target hardware may not
3428 have this support. The main purpose of this is EPROM/ROM code
3429 debugging, so you can set a breakpoint at an instruction without
3430 changing the instruction. This can be used with the new trap-generation
3431 provided by SPARClite DSU and most x86-based targets. These targets
3432 will generate traps when a program accesses some data or instruction
3433 address that is assigned to the debug registers. However the hardware
3434 breakpoint registers can take a limited number of breakpoints. For
3435 example, on the DSU, only two data breakpoints can be set at a time, and
3436 @value{GDBN} will reject this command if more than two are used. Delete
3437 or disable unused hardware breakpoints before setting new ones
3438 (@pxref{Disabling, ,Disabling Breakpoints}).
3439 @xref{Conditions, ,Break Conditions}.
3440 For remote targets, you can restrict the number of hardware
3441 breakpoints @value{GDBN} will use, see @ref{set remote
3442 hardware-breakpoint-limit}.
3445 @item thbreak @var{args}
3446 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3447 are the same as for the @code{hbreak} command and the breakpoint is set in
3448 the same way. However, like the @code{tbreak} command,
3449 the breakpoint is automatically deleted after the
3450 first time your program stops there. Also, like the @code{hbreak}
3451 command, the breakpoint requires hardware support and some target hardware
3452 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3453 See also @ref{Conditions, ,Break Conditions}.
3456 @cindex regular expression
3457 @cindex breakpoints at functions matching a regexp
3458 @cindex set breakpoints in many functions
3459 @item rbreak @var{regex}
3460 Set breakpoints on all functions matching the regular expression
3461 @var{regex}. This command sets an unconditional breakpoint on all
3462 matches, printing a list of all breakpoints it set. Once these
3463 breakpoints are set, they are treated just like the breakpoints set with
3464 the @code{break} command. You can delete them, disable them, or make
3465 them conditional the same way as any other breakpoint.
3467 The syntax of the regular expression is the standard one used with tools
3468 like @file{grep}. Note that this is different from the syntax used by
3469 shells, so for instance @code{foo*} matches all functions that include
3470 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3471 @code{.*} leading and trailing the regular expression you supply, so to
3472 match only functions that begin with @code{foo}, use @code{^foo}.
3474 @cindex non-member C@t{++} functions, set breakpoint in
3475 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3476 breakpoints on overloaded functions that are not members of any special
3479 @cindex set breakpoints on all functions
3480 The @code{rbreak} command can be used to set breakpoints in
3481 @strong{all} the functions in a program, like this:
3484 (@value{GDBP}) rbreak .
3487 @item rbreak @var{file}:@var{regex}
3488 If @code{rbreak} is called with a filename qualification, it limits
3489 the search for functions matching the given regular expression to the
3490 specified @var{file}. This can be used, for example, to set breakpoints on
3491 every function in a given file:
3494 (@value{GDBP}) rbreak file.c:.
3497 The colon separating the filename qualifier from the regex may
3498 optionally be surrounded by spaces.
3500 @kindex info breakpoints
3501 @cindex @code{$_} and @code{info breakpoints}
3502 @item info breakpoints @r{[}@var{n}@dots{}@r{]}
3503 @itemx info break @r{[}@var{n}@dots{}@r{]}
3504 Print a table of all breakpoints, watchpoints, and catchpoints set and
3505 not deleted. Optional argument @var{n} means print information only
3506 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
3507 For each breakpoint, following columns are printed:
3510 @item Breakpoint Numbers
3512 Breakpoint, watchpoint, or catchpoint.
3514 Whether the breakpoint is marked to be disabled or deleted when hit.
3515 @item Enabled or Disabled
3516 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3517 that are not enabled.
3519 Where the breakpoint is in your program, as a memory address. For a
3520 pending breakpoint whose address is not yet known, this field will
3521 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3522 library that has the symbol or line referred by breakpoint is loaded.
3523 See below for details. A breakpoint with several locations will
3524 have @samp{<MULTIPLE>} in this field---see below for details.
3526 Where the breakpoint is in the source for your program, as a file and
3527 line number. For a pending breakpoint, the original string passed to
3528 the breakpoint command will be listed as it cannot be resolved until
3529 the appropriate shared library is loaded in the future.
3533 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
3534 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
3535 @value{GDBN} on the host's side. If it is ``target'', then the condition
3536 is evaluated by the target. The @code{info break} command shows
3537 the condition on the line following the affected breakpoint, together with
3538 its condition evaluation mode in between parentheses.
3540 Breakpoint commands, if any, are listed after that. A pending breakpoint is
3541 allowed to have a condition specified for it. The condition is not parsed for
3542 validity until a shared library is loaded that allows the pending
3543 breakpoint to resolve to a valid location.
3546 @code{info break} with a breakpoint
3547 number @var{n} as argument lists only that breakpoint. The
3548 convenience variable @code{$_} and the default examining-address for
3549 the @code{x} command are set to the address of the last breakpoint
3550 listed (@pxref{Memory, ,Examining Memory}).
3553 @code{info break} displays a count of the number of times the breakpoint
3554 has been hit. This is especially useful in conjunction with the
3555 @code{ignore} command. You can ignore a large number of breakpoint
3556 hits, look at the breakpoint info to see how many times the breakpoint
3557 was hit, and then run again, ignoring one less than that number. This
3558 will get you quickly to the last hit of that breakpoint.
3561 For a breakpoints with an enable count (xref) greater than 1,
3562 @code{info break} also displays that count.
3566 @value{GDBN} allows you to set any number of breakpoints at the same place in
3567 your program. There is nothing silly or meaningless about this. When
3568 the breakpoints are conditional, this is even useful
3569 (@pxref{Conditions, ,Break Conditions}).
3571 @cindex multiple locations, breakpoints
3572 @cindex breakpoints, multiple locations
3573 It is possible that a breakpoint corresponds to several locations
3574 in your program. Examples of this situation are:
3578 Multiple functions in the program may have the same name.
3581 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3582 instances of the function body, used in different cases.
3585 For a C@t{++} template function, a given line in the function can
3586 correspond to any number of instantiations.
3589 For an inlined function, a given source line can correspond to
3590 several places where that function is inlined.
3593 In all those cases, @value{GDBN} will insert a breakpoint at all
3594 the relevant locations.
3596 A breakpoint with multiple locations is displayed in the breakpoint
3597 table using several rows---one header row, followed by one row for
3598 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3599 address column. The rows for individual locations contain the actual
3600 addresses for locations, and show the functions to which those
3601 locations belong. The number column for a location is of the form
3602 @var{breakpoint-number}.@var{location-number}.
3607 Num Type Disp Enb Address What
3608 1 breakpoint keep y <MULTIPLE>
3610 breakpoint already hit 1 time
3611 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3612 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3615 Each location can be individually enabled or disabled by passing
3616 @var{breakpoint-number}.@var{location-number} as argument to the
3617 @code{enable} and @code{disable} commands. Note that you cannot
3618 delete the individual locations from the list, you can only delete the
3619 entire list of locations that belong to their parent breakpoint (with
3620 the @kbd{delete @var{num}} command, where @var{num} is the number of
3621 the parent breakpoint, 1 in the above example). Disabling or enabling
3622 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3623 that belong to that breakpoint.
3625 @cindex pending breakpoints
3626 It's quite common to have a breakpoint inside a shared library.
3627 Shared libraries can be loaded and unloaded explicitly,
3628 and possibly repeatedly, as the program is executed. To support
3629 this use case, @value{GDBN} updates breakpoint locations whenever
3630 any shared library is loaded or unloaded. Typically, you would
3631 set a breakpoint in a shared library at the beginning of your
3632 debugging session, when the library is not loaded, and when the
3633 symbols from the library are not available. When you try to set
3634 breakpoint, @value{GDBN} will ask you if you want to set
3635 a so called @dfn{pending breakpoint}---breakpoint whose address
3636 is not yet resolved.
3638 After the program is run, whenever a new shared library is loaded,
3639 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3640 shared library contains the symbol or line referred to by some
3641 pending breakpoint, that breakpoint is resolved and becomes an
3642 ordinary breakpoint. When a library is unloaded, all breakpoints
3643 that refer to its symbols or source lines become pending again.
3645 This logic works for breakpoints with multiple locations, too. For
3646 example, if you have a breakpoint in a C@t{++} template function, and
3647 a newly loaded shared library has an instantiation of that template,
3648 a new location is added to the list of locations for the breakpoint.
3650 Except for having unresolved address, pending breakpoints do not
3651 differ from regular breakpoints. You can set conditions or commands,
3652 enable and disable them and perform other breakpoint operations.
3654 @value{GDBN} provides some additional commands for controlling what
3655 happens when the @samp{break} command cannot resolve breakpoint
3656 address specification to an address:
3658 @kindex set breakpoint pending
3659 @kindex show breakpoint pending
3661 @item set breakpoint pending auto
3662 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3663 location, it queries you whether a pending breakpoint should be created.
3665 @item set breakpoint pending on
3666 This indicates that an unrecognized breakpoint location should automatically
3667 result in a pending breakpoint being created.
3669 @item set breakpoint pending off
3670 This indicates that pending breakpoints are not to be created. Any
3671 unrecognized breakpoint location results in an error. This setting does
3672 not affect any pending breakpoints previously created.
3674 @item show breakpoint pending
3675 Show the current behavior setting for creating pending breakpoints.
3678 The settings above only affect the @code{break} command and its
3679 variants. Once breakpoint is set, it will be automatically updated
3680 as shared libraries are loaded and unloaded.
3682 @cindex automatic hardware breakpoints
3683 For some targets, @value{GDBN} can automatically decide if hardware or
3684 software breakpoints should be used, depending on whether the
3685 breakpoint address is read-only or read-write. This applies to
3686 breakpoints set with the @code{break} command as well as to internal
3687 breakpoints set by commands like @code{next} and @code{finish}. For
3688 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3691 You can control this automatic behaviour with the following commands::
3693 @kindex set breakpoint auto-hw
3694 @kindex show breakpoint auto-hw
3696 @item set breakpoint auto-hw on
3697 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3698 will try to use the target memory map to decide if software or hardware
3699 breakpoint must be used.
3701 @item set breakpoint auto-hw off
3702 This indicates @value{GDBN} should not automatically select breakpoint
3703 type. If the target provides a memory map, @value{GDBN} will warn when
3704 trying to set software breakpoint at a read-only address.
3707 @value{GDBN} normally implements breakpoints by replacing the program code
3708 at the breakpoint address with a special instruction, which, when
3709 executed, given control to the debugger. By default, the program
3710 code is so modified only when the program is resumed. As soon as
3711 the program stops, @value{GDBN} restores the original instructions. This
3712 behaviour guards against leaving breakpoints inserted in the
3713 target should gdb abrubptly disconnect. However, with slow remote
3714 targets, inserting and removing breakpoint can reduce the performance.
3715 This behavior can be controlled with the following commands::
3717 @kindex set breakpoint always-inserted
3718 @kindex show breakpoint always-inserted
3720 @item set breakpoint always-inserted off
3721 All breakpoints, including newly added by the user, are inserted in
3722 the target only when the target is resumed. All breakpoints are
3723 removed from the target when it stops.
3725 @item set breakpoint always-inserted on
3726 Causes all breakpoints to be inserted in the target at all times. If
3727 the user adds a new breakpoint, or changes an existing breakpoint, the
3728 breakpoints in the target are updated immediately. A breakpoint is
3729 removed from the target only when breakpoint itself is removed.
3731 @cindex non-stop mode, and @code{breakpoint always-inserted}
3732 @item set breakpoint always-inserted auto
3733 This is the default mode. If @value{GDBN} is controlling the inferior
3734 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3735 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3736 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3737 @code{breakpoint always-inserted} mode is off.
3740 @value{GDBN} handles conditional breakpoints by evaluating these conditions
3741 when a breakpoint breaks. If the condition is true, then the process being
3742 debugged stops, otherwise the process is resumed.
3744 If the target supports evaluating conditions on its end, @value{GDBN} may
3745 download the breakpoint, together with its conditions, to it.
3747 This feature can be controlled via the following commands:
3749 @kindex set breakpoint condition-evaluation
3750 @kindex show breakpoint condition-evaluation
3752 @item set breakpoint condition-evaluation host
3753 This option commands @value{GDBN} to evaluate the breakpoint
3754 conditions on the host's side. Unconditional breakpoints are sent to
3755 the target which in turn receives the triggers and reports them back to GDB
3756 for condition evaluation. This is the standard evaluation mode.
3758 @item set breakpoint condition-evaluation target
3759 This option commands @value{GDBN} to download breakpoint conditions
3760 to the target at the moment of their insertion. The target
3761 is responsible for evaluating the conditional expression and reporting
3762 breakpoint stop events back to @value{GDBN} whenever the condition
3763 is true. Due to limitations of target-side evaluation, some conditions
3764 cannot be evaluated there, e.g., conditions that depend on local data
3765 that is only known to the host. Examples include
3766 conditional expressions involving convenience variables, complex types
3767 that cannot be handled by the agent expression parser and expressions
3768 that are too long to be sent over to the target, specially when the
3769 target is a remote system. In these cases, the conditions will be
3770 evaluated by @value{GDBN}.
3772 @item set breakpoint condition-evaluation auto
3773 This is the default mode. If the target supports evaluating breakpoint
3774 conditions on its end, @value{GDBN} will download breakpoint conditions to
3775 the target (limitations mentioned previously apply). If the target does
3776 not support breakpoint condition evaluation, then @value{GDBN} will fallback
3777 to evaluating all these conditions on the host's side.
3781 @cindex negative breakpoint numbers
3782 @cindex internal @value{GDBN} breakpoints
3783 @value{GDBN} itself sometimes sets breakpoints in your program for
3784 special purposes, such as proper handling of @code{longjmp} (in C
3785 programs). These internal breakpoints are assigned negative numbers,
3786 starting with @code{-1}; @samp{info breakpoints} does not display them.
3787 You can see these breakpoints with the @value{GDBN} maintenance command
3788 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3791 @node Set Watchpoints
3792 @subsection Setting Watchpoints
3794 @cindex setting watchpoints
3795 You can use a watchpoint to stop execution whenever the value of an
3796 expression changes, without having to predict a particular place where
3797 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3798 The expression may be as simple as the value of a single variable, or
3799 as complex as many variables combined by operators. Examples include:
3803 A reference to the value of a single variable.
3806 An address cast to an appropriate data type. For example,
3807 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3808 address (assuming an @code{int} occupies 4 bytes).
3811 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3812 expression can use any operators valid in the program's native
3813 language (@pxref{Languages}).
3816 You can set a watchpoint on an expression even if the expression can
3817 not be evaluated yet. For instance, you can set a watchpoint on
3818 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3819 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3820 the expression produces a valid value. If the expression becomes
3821 valid in some other way than changing a variable (e.g.@: if the memory
3822 pointed to by @samp{*global_ptr} becomes readable as the result of a
3823 @code{malloc} call), @value{GDBN} may not stop until the next time
3824 the expression changes.
3826 @cindex software watchpoints
3827 @cindex hardware watchpoints
3828 Depending on your system, watchpoints may be implemented in software or
3829 hardware. @value{GDBN} does software watchpointing by single-stepping your
3830 program and testing the variable's value each time, which is hundreds of
3831 times slower than normal execution. (But this may still be worth it, to
3832 catch errors where you have no clue what part of your program is the
3835 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3836 x86-based targets, @value{GDBN} includes support for hardware
3837 watchpoints, which do not slow down the running of your program.
3841 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3842 Set a watchpoint for an expression. @value{GDBN} will break when the
3843 expression @var{expr} is written into by the program and its value
3844 changes. The simplest (and the most popular) use of this command is
3845 to watch the value of a single variable:
3848 (@value{GDBP}) watch foo
3851 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3852 argument, @value{GDBN} breaks only when the thread identified by
3853 @var{threadnum} changes the value of @var{expr}. If any other threads
3854 change the value of @var{expr}, @value{GDBN} will not break. Note
3855 that watchpoints restricted to a single thread in this way only work
3856 with Hardware Watchpoints.
3858 Ordinarily a watchpoint respects the scope of variables in @var{expr}
3859 (see below). The @code{-location} argument tells @value{GDBN} to
3860 instead watch the memory referred to by @var{expr}. In this case,
3861 @value{GDBN} will evaluate @var{expr}, take the address of the result,
3862 and watch the memory at that address. The type of the result is used
3863 to determine the size of the watched memory. If the expression's
3864 result does not have an address, then @value{GDBN} will print an
3867 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
3868 of masked watchpoints, if the current architecture supports this
3869 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
3870 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
3871 to an address to watch. The mask specifies that some bits of an address
3872 (the bits which are reset in the mask) should be ignored when matching
3873 the address accessed by the inferior against the watchpoint address.
3874 Thus, a masked watchpoint watches many addresses simultaneously---those
3875 addresses whose unmasked bits are identical to the unmasked bits in the
3876 watchpoint address. The @code{mask} argument implies @code{-location}.
3880 (@value{GDBP}) watch foo mask 0xffff00ff
3881 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
3885 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3886 Set a watchpoint that will break when the value of @var{expr} is read
3890 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3891 Set a watchpoint that will break when @var{expr} is either read from
3892 or written into by the program.
3894 @kindex info watchpoints @r{[}@var{n}@dots{}@r{]}
3895 @item info watchpoints @r{[}@var{n}@dots{}@r{]}
3896 This command prints a list of watchpoints, using the same format as
3897 @code{info break} (@pxref{Set Breaks}).
3900 If you watch for a change in a numerically entered address you need to
3901 dereference it, as the address itself is just a constant number which will
3902 never change. @value{GDBN} refuses to create a watchpoint that watches
3903 a never-changing value:
3906 (@value{GDBP}) watch 0x600850
3907 Cannot watch constant value 0x600850.
3908 (@value{GDBP}) watch *(int *) 0x600850
3909 Watchpoint 1: *(int *) 6293584
3912 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3913 watchpoints execute very quickly, and the debugger reports a change in
3914 value at the exact instruction where the change occurs. If @value{GDBN}
3915 cannot set a hardware watchpoint, it sets a software watchpoint, which
3916 executes more slowly and reports the change in value at the next
3917 @emph{statement}, not the instruction, after the change occurs.
3919 @cindex use only software watchpoints
3920 You can force @value{GDBN} to use only software watchpoints with the
3921 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3922 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3923 the underlying system supports them. (Note that hardware-assisted
3924 watchpoints that were set @emph{before} setting
3925 @code{can-use-hw-watchpoints} to zero will still use the hardware
3926 mechanism of watching expression values.)
3929 @item set can-use-hw-watchpoints
3930 @kindex set can-use-hw-watchpoints
3931 Set whether or not to use hardware watchpoints.
3933 @item show can-use-hw-watchpoints
3934 @kindex show can-use-hw-watchpoints
3935 Show the current mode of using hardware watchpoints.
3938 For remote targets, you can restrict the number of hardware
3939 watchpoints @value{GDBN} will use, see @ref{set remote
3940 hardware-breakpoint-limit}.
3942 When you issue the @code{watch} command, @value{GDBN} reports
3945 Hardware watchpoint @var{num}: @var{expr}
3949 if it was able to set a hardware watchpoint.
3951 Currently, the @code{awatch} and @code{rwatch} commands can only set
3952 hardware watchpoints, because accesses to data that don't change the
3953 value of the watched expression cannot be detected without examining
3954 every instruction as it is being executed, and @value{GDBN} does not do
3955 that currently. If @value{GDBN} finds that it is unable to set a
3956 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3957 will print a message like this:
3960 Expression cannot be implemented with read/access watchpoint.
3963 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3964 data type of the watched expression is wider than what a hardware
3965 watchpoint on the target machine can handle. For example, some systems
3966 can only watch regions that are up to 4 bytes wide; on such systems you
3967 cannot set hardware watchpoints for an expression that yields a
3968 double-precision floating-point number (which is typically 8 bytes
3969 wide). As a work-around, it might be possible to break the large region
3970 into a series of smaller ones and watch them with separate watchpoints.
3972 If you set too many hardware watchpoints, @value{GDBN} might be unable
3973 to insert all of them when you resume the execution of your program.
3974 Since the precise number of active watchpoints is unknown until such
3975 time as the program is about to be resumed, @value{GDBN} might not be
3976 able to warn you about this when you set the watchpoints, and the
3977 warning will be printed only when the program is resumed:
3980 Hardware watchpoint @var{num}: Could not insert watchpoint
3984 If this happens, delete or disable some of the watchpoints.
3986 Watching complex expressions that reference many variables can also
3987 exhaust the resources available for hardware-assisted watchpoints.
3988 That's because @value{GDBN} needs to watch every variable in the
3989 expression with separately allocated resources.
3991 If you call a function interactively using @code{print} or @code{call},
3992 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3993 kind of breakpoint or the call completes.
3995 @value{GDBN} automatically deletes watchpoints that watch local
3996 (automatic) variables, or expressions that involve such variables, when
3997 they go out of scope, that is, when the execution leaves the block in
3998 which these variables were defined. In particular, when the program
3999 being debugged terminates, @emph{all} local variables go out of scope,
4000 and so only watchpoints that watch global variables remain set. If you
4001 rerun the program, you will need to set all such watchpoints again. One
4002 way of doing that would be to set a code breakpoint at the entry to the
4003 @code{main} function and when it breaks, set all the watchpoints.
4005 @cindex watchpoints and threads
4006 @cindex threads and watchpoints
4007 In multi-threaded programs, watchpoints will detect changes to the
4008 watched expression from every thread.
4011 @emph{Warning:} In multi-threaded programs, software watchpoints
4012 have only limited usefulness. If @value{GDBN} creates a software
4013 watchpoint, it can only watch the value of an expression @emph{in a
4014 single thread}. If you are confident that the expression can only
4015 change due to the current thread's activity (and if you are also
4016 confident that no other thread can become current), then you can use
4017 software watchpoints as usual. However, @value{GDBN} may not notice
4018 when a non-current thread's activity changes the expression. (Hardware
4019 watchpoints, in contrast, watch an expression in all threads.)
4022 @xref{set remote hardware-watchpoint-limit}.
4024 @node Set Catchpoints
4025 @subsection Setting Catchpoints
4026 @cindex catchpoints, setting
4027 @cindex exception handlers
4028 @cindex event handling
4030 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4031 kinds of program events, such as C@t{++} exceptions or the loading of a
4032 shared library. Use the @code{catch} command to set a catchpoint.
4036 @item catch @var{event}
4037 Stop when @var{event} occurs. @var{event} can be any of the following:
4040 @cindex stop on C@t{++} exceptions
4041 The throwing of a C@t{++} exception.
4044 The catching of a C@t{++} exception.
4047 @cindex Ada exception catching
4048 @cindex catch Ada exceptions
4049 An Ada exception being raised. If an exception name is specified
4050 at the end of the command (eg @code{catch exception Program_Error}),
4051 the debugger will stop only when this specific exception is raised.
4052 Otherwise, the debugger stops execution when any Ada exception is raised.
4054 When inserting an exception catchpoint on a user-defined exception whose
4055 name is identical to one of the exceptions defined by the language, the
4056 fully qualified name must be used as the exception name. Otherwise,
4057 @value{GDBN} will assume that it should stop on the pre-defined exception
4058 rather than the user-defined one. For instance, assuming an exception
4059 called @code{Constraint_Error} is defined in package @code{Pck}, then
4060 the command to use to catch such exceptions is @kbd{catch exception
4061 Pck.Constraint_Error}.
4063 @item exception unhandled
4064 An exception that was raised but is not handled by the program.
4067 A failed Ada assertion.
4070 @cindex break on fork/exec
4071 A call to @code{exec}. This is currently only available for HP-UX
4075 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @dots{}
4076 @cindex break on a system call.
4077 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4078 syscall is a mechanism for application programs to request a service
4079 from the operating system (OS) or one of the OS system services.
4080 @value{GDBN} can catch some or all of the syscalls issued by the
4081 debuggee, and show the related information for each syscall. If no
4082 argument is specified, calls to and returns from all system calls
4085 @var{name} can be any system call name that is valid for the
4086 underlying OS. Just what syscalls are valid depends on the OS. On
4087 GNU and Unix systems, you can find the full list of valid syscall
4088 names on @file{/usr/include/asm/unistd.h}.
4090 @c For MS-Windows, the syscall names and the corresponding numbers
4091 @c can be found, e.g., on this URL:
4092 @c http://www.metasploit.com/users/opcode/syscalls.html
4093 @c but we don't support Windows syscalls yet.
4095 Normally, @value{GDBN} knows in advance which syscalls are valid for
4096 each OS, so you can use the @value{GDBN} command-line completion
4097 facilities (@pxref{Completion,, command completion}) to list the
4100 You may also specify the system call numerically. A syscall's
4101 number is the value passed to the OS's syscall dispatcher to
4102 identify the requested service. When you specify the syscall by its
4103 name, @value{GDBN} uses its database of syscalls to convert the name
4104 into the corresponding numeric code, but using the number directly
4105 may be useful if @value{GDBN}'s database does not have the complete
4106 list of syscalls on your system (e.g., because @value{GDBN} lags
4107 behind the OS upgrades).
4109 The example below illustrates how this command works if you don't provide
4113 (@value{GDBP}) catch syscall
4114 Catchpoint 1 (syscall)
4116 Starting program: /tmp/catch-syscall
4118 Catchpoint 1 (call to syscall 'close'), \
4119 0xffffe424 in __kernel_vsyscall ()
4123 Catchpoint 1 (returned from syscall 'close'), \
4124 0xffffe424 in __kernel_vsyscall ()
4128 Here is an example of catching a system call by name:
4131 (@value{GDBP}) catch syscall chroot
4132 Catchpoint 1 (syscall 'chroot' [61])
4134 Starting program: /tmp/catch-syscall
4136 Catchpoint 1 (call to syscall 'chroot'), \
4137 0xffffe424 in __kernel_vsyscall ()
4141 Catchpoint 1 (returned from syscall 'chroot'), \
4142 0xffffe424 in __kernel_vsyscall ()
4146 An example of specifying a system call numerically. In the case
4147 below, the syscall number has a corresponding entry in the XML
4148 file, so @value{GDBN} finds its name and prints it:
4151 (@value{GDBP}) catch syscall 252
4152 Catchpoint 1 (syscall(s) 'exit_group')
4154 Starting program: /tmp/catch-syscall
4156 Catchpoint 1 (call to syscall 'exit_group'), \
4157 0xffffe424 in __kernel_vsyscall ()
4161 Program exited normally.
4165 However, there can be situations when there is no corresponding name
4166 in XML file for that syscall number. In this case, @value{GDBN} prints
4167 a warning message saying that it was not able to find the syscall name,
4168 but the catchpoint will be set anyway. See the example below:
4171 (@value{GDBP}) catch syscall 764
4172 warning: The number '764' does not represent a known syscall.
4173 Catchpoint 2 (syscall 764)
4177 If you configure @value{GDBN} using the @samp{--without-expat} option,
4178 it will not be able to display syscall names. Also, if your
4179 architecture does not have an XML file describing its system calls,
4180 you will not be able to see the syscall names. It is important to
4181 notice that these two features are used for accessing the syscall
4182 name database. In either case, you will see a warning like this:
4185 (@value{GDBP}) catch syscall
4186 warning: Could not open "syscalls/i386-linux.xml"
4187 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4188 GDB will not be able to display syscall names.
4189 Catchpoint 1 (syscall)
4193 Of course, the file name will change depending on your architecture and system.
4195 Still using the example above, you can also try to catch a syscall by its
4196 number. In this case, you would see something like:
4199 (@value{GDBP}) catch syscall 252
4200 Catchpoint 1 (syscall(s) 252)
4203 Again, in this case @value{GDBN} would not be able to display syscall's names.
4206 A call to @code{fork}. This is currently only available for HP-UX
4210 A call to @code{vfork}. This is currently only available for HP-UX
4213 @item load @r{[}regexp@r{]}
4214 @itemx unload @r{[}regexp@r{]}
4215 The loading or unloading of a shared library. If @var{regexp} is
4216 given, then the catchpoint will stop only if the regular expression
4217 matches one of the affected libraries.
4221 @item tcatch @var{event}
4222 Set a catchpoint that is enabled only for one stop. The catchpoint is
4223 automatically deleted after the first time the event is caught.
4227 Use the @code{info break} command to list the current catchpoints.
4229 There are currently some limitations to C@t{++} exception handling
4230 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
4234 If you call a function interactively, @value{GDBN} normally returns
4235 control to you when the function has finished executing. If the call
4236 raises an exception, however, the call may bypass the mechanism that
4237 returns control to you and cause your program either to abort or to
4238 simply continue running until it hits a breakpoint, catches a signal
4239 that @value{GDBN} is listening for, or exits. This is the case even if
4240 you set a catchpoint for the exception; catchpoints on exceptions are
4241 disabled within interactive calls.
4244 You cannot raise an exception interactively.
4247 You cannot install an exception handler interactively.
4250 @cindex raise exceptions
4251 Sometimes @code{catch} is not the best way to debug exception handling:
4252 if you need to know exactly where an exception is raised, it is better to
4253 stop @emph{before} the exception handler is called, since that way you
4254 can see the stack before any unwinding takes place. If you set a
4255 breakpoint in an exception handler instead, it may not be easy to find
4256 out where the exception was raised.
4258 To stop just before an exception handler is called, you need some
4259 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
4260 raised by calling a library function named @code{__raise_exception}
4261 which has the following ANSI C interface:
4264 /* @var{addr} is where the exception identifier is stored.
4265 @var{id} is the exception identifier. */
4266 void __raise_exception (void **addr, void *id);
4270 To make the debugger catch all exceptions before any stack
4271 unwinding takes place, set a breakpoint on @code{__raise_exception}
4272 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
4274 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
4275 that depends on the value of @var{id}, you can stop your program when
4276 a specific exception is raised. You can use multiple conditional
4277 breakpoints to stop your program when any of a number of exceptions are
4282 @subsection Deleting Breakpoints
4284 @cindex clearing breakpoints, watchpoints, catchpoints
4285 @cindex deleting breakpoints, watchpoints, catchpoints
4286 It is often necessary to eliminate a breakpoint, watchpoint, or
4287 catchpoint once it has done its job and you no longer want your program
4288 to stop there. This is called @dfn{deleting} the breakpoint. A
4289 breakpoint that has been deleted no longer exists; it is forgotten.
4291 With the @code{clear} command you can delete breakpoints according to
4292 where they are in your program. With the @code{delete} command you can
4293 delete individual breakpoints, watchpoints, or catchpoints by specifying
4294 their breakpoint numbers.
4296 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4297 automatically ignores breakpoints on the first instruction to be executed
4298 when you continue execution without changing the execution address.
4303 Delete any breakpoints at the next instruction to be executed in the
4304 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4305 the innermost frame is selected, this is a good way to delete a
4306 breakpoint where your program just stopped.
4308 @item clear @var{location}
4309 Delete any breakpoints set at the specified @var{location}.
4310 @xref{Specify Location}, for the various forms of @var{location}; the
4311 most useful ones are listed below:
4314 @item clear @var{function}
4315 @itemx clear @var{filename}:@var{function}
4316 Delete any breakpoints set at entry to the named @var{function}.
4318 @item clear @var{linenum}
4319 @itemx clear @var{filename}:@var{linenum}
4320 Delete any breakpoints set at or within the code of the specified
4321 @var{linenum} of the specified @var{filename}.
4324 @cindex delete breakpoints
4326 @kindex d @r{(@code{delete})}
4327 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4328 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4329 ranges specified as arguments. If no argument is specified, delete all
4330 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4331 confirm off}). You can abbreviate this command as @code{d}.
4335 @subsection Disabling Breakpoints
4337 @cindex enable/disable a breakpoint
4338 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4339 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4340 it had been deleted, but remembers the information on the breakpoint so
4341 that you can @dfn{enable} it again later.
4343 You disable and enable breakpoints, watchpoints, and catchpoints with
4344 the @code{enable} and @code{disable} commands, optionally specifying
4345 one or more breakpoint numbers as arguments. Use @code{info break} to
4346 print a list of all breakpoints, watchpoints, and catchpoints if you
4347 do not know which numbers to use.
4349 Disabling and enabling a breakpoint that has multiple locations
4350 affects all of its locations.
4352 A breakpoint, watchpoint, or catchpoint can have any of several
4353 different states of enablement:
4357 Enabled. The breakpoint stops your program. A breakpoint set
4358 with the @code{break} command starts out in this state.
4360 Disabled. The breakpoint has no effect on your program.
4362 Enabled once. The breakpoint stops your program, but then becomes
4365 Enabled for a count. The breakpoint stops your program for the next
4366 N times, then becomes disabled.
4368 Enabled for deletion. The breakpoint stops your program, but
4369 immediately after it does so it is deleted permanently. A breakpoint
4370 set with the @code{tbreak} command starts out in this state.
4373 You can use the following commands to enable or disable breakpoints,
4374 watchpoints, and catchpoints:
4378 @kindex dis @r{(@code{disable})}
4379 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4380 Disable the specified breakpoints---or all breakpoints, if none are
4381 listed. A disabled breakpoint has no effect but is not forgotten. All
4382 options such as ignore-counts, conditions and commands are remembered in
4383 case the breakpoint is enabled again later. You may abbreviate
4384 @code{disable} as @code{dis}.
4387 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4388 Enable the specified breakpoints (or all defined breakpoints). They
4389 become effective once again in stopping your program.
4391 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4392 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4393 of these breakpoints immediately after stopping your program.
4395 @item enable @r{[}breakpoints@r{]} count @var{count} @var{range}@dots{}
4396 Enable the specified breakpoints temporarily. @value{GDBN} records
4397 @var{count} with each of the specified breakpoints, and decrements a
4398 breakpoint's count when it is hit. When any count reaches 0,
4399 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
4400 count (@pxref{Conditions, ,Break Conditions}), that will be
4401 decremented to 0 before @var{count} is affected.
4403 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4404 Enable the specified breakpoints to work once, then die. @value{GDBN}
4405 deletes any of these breakpoints as soon as your program stops there.
4406 Breakpoints set by the @code{tbreak} command start out in this state.
4409 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4410 @c confusing: tbreak is also initially enabled.
4411 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4412 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4413 subsequently, they become disabled or enabled only when you use one of
4414 the commands above. (The command @code{until} can set and delete a
4415 breakpoint of its own, but it does not change the state of your other
4416 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4420 @subsection Break Conditions
4421 @cindex conditional breakpoints
4422 @cindex breakpoint conditions
4424 @c FIXME what is scope of break condition expr? Context where wanted?
4425 @c in particular for a watchpoint?
4426 The simplest sort of breakpoint breaks every time your program reaches a
4427 specified place. You can also specify a @dfn{condition} for a
4428 breakpoint. A condition is just a Boolean expression in your
4429 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4430 a condition evaluates the expression each time your program reaches it,
4431 and your program stops only if the condition is @emph{true}.
4433 This is the converse of using assertions for program validation; in that
4434 situation, you want to stop when the assertion is violated---that is,
4435 when the condition is false. In C, if you want to test an assertion expressed
4436 by the condition @var{assert}, you should set the condition
4437 @samp{! @var{assert}} on the appropriate breakpoint.
4439 Conditions are also accepted for watchpoints; you may not need them,
4440 since a watchpoint is inspecting the value of an expression anyhow---but
4441 it might be simpler, say, to just set a watchpoint on a variable name,
4442 and specify a condition that tests whether the new value is an interesting
4445 Break conditions can have side effects, and may even call functions in
4446 your program. This can be useful, for example, to activate functions
4447 that log program progress, or to use your own print functions to
4448 format special data structures. The effects are completely predictable
4449 unless there is another enabled breakpoint at the same address. (In
4450 that case, @value{GDBN} might see the other breakpoint first and stop your
4451 program without checking the condition of this one.) Note that
4452 breakpoint commands are usually more convenient and flexible than break
4454 purpose of performing side effects when a breakpoint is reached
4455 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4457 Breakpoint conditions can also be evaluated on the target's side if
4458 the target supports it. Instead of evaluating the conditions locally,
4459 @value{GDBN} encodes the expression into an agent expression
4460 (@pxref{Agent Expressions}) suitable for execution on the target,
4461 independently of @value{GDBN}. Global variables become raw memory
4462 locations, locals become stack accesses, and so forth.
4464 In this case, @value{GDBN} will only be notified of a breakpoint trigger
4465 when its condition evaluates to true. This mechanism may provide faster
4466 response times depending on the performance characteristics of the target
4467 since it does not need to keep @value{GDBN} informed about
4468 every breakpoint trigger, even those with false conditions.
4470 Break conditions can be specified when a breakpoint is set, by using
4471 @samp{if} in the arguments to the @code{break} command. @xref{Set
4472 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4473 with the @code{condition} command.
4475 You can also use the @code{if} keyword with the @code{watch} command.
4476 The @code{catch} command does not recognize the @code{if} keyword;
4477 @code{condition} is the only way to impose a further condition on a
4482 @item condition @var{bnum} @var{expression}
4483 Specify @var{expression} as the break condition for breakpoint,
4484 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4485 breakpoint @var{bnum} stops your program only if the value of
4486 @var{expression} is true (nonzero, in C). When you use
4487 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4488 syntactic correctness, and to determine whether symbols in it have
4489 referents in the context of your breakpoint. If @var{expression} uses
4490 symbols not referenced in the context of the breakpoint, @value{GDBN}
4491 prints an error message:
4494 No symbol "foo" in current context.
4499 not actually evaluate @var{expression} at the time the @code{condition}
4500 command (or a command that sets a breakpoint with a condition, like
4501 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4503 @item condition @var{bnum}
4504 Remove the condition from breakpoint number @var{bnum}. It becomes
4505 an ordinary unconditional breakpoint.
4508 @cindex ignore count (of breakpoint)
4509 A special case of a breakpoint condition is to stop only when the
4510 breakpoint has been reached a certain number of times. This is so
4511 useful that there is a special way to do it, using the @dfn{ignore
4512 count} of the breakpoint. Every breakpoint has an ignore count, which
4513 is an integer. Most of the time, the ignore count is zero, and
4514 therefore has no effect. But if your program reaches a breakpoint whose
4515 ignore count is positive, then instead of stopping, it just decrements
4516 the ignore count by one and continues. As a result, if the ignore count
4517 value is @var{n}, the breakpoint does not stop the next @var{n} times
4518 your program reaches it.
4522 @item ignore @var{bnum} @var{count}
4523 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4524 The next @var{count} times the breakpoint is reached, your program's
4525 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4528 To make the breakpoint stop the next time it is reached, specify
4531 When you use @code{continue} to resume execution of your program from a
4532 breakpoint, you can specify an ignore count directly as an argument to
4533 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4534 Stepping,,Continuing and Stepping}.
4536 If a breakpoint has a positive ignore count and a condition, the
4537 condition is not checked. Once the ignore count reaches zero,
4538 @value{GDBN} resumes checking the condition.
4540 You could achieve the effect of the ignore count with a condition such
4541 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4542 is decremented each time. @xref{Convenience Vars, ,Convenience
4546 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4549 @node Break Commands
4550 @subsection Breakpoint Command Lists
4552 @cindex breakpoint commands
4553 You can give any breakpoint (or watchpoint or catchpoint) a series of
4554 commands to execute when your program stops due to that breakpoint. For
4555 example, you might want to print the values of certain expressions, or
4556 enable other breakpoints.
4560 @kindex end@r{ (breakpoint commands)}
4561 @item commands @r{[}@var{range}@dots{}@r{]}
4562 @itemx @dots{} @var{command-list} @dots{}
4564 Specify a list of commands for the given breakpoints. The commands
4565 themselves appear on the following lines. Type a line containing just
4566 @code{end} to terminate the commands.
4568 To remove all commands from a breakpoint, type @code{commands} and
4569 follow it immediately with @code{end}; that is, give no commands.
4571 With no argument, @code{commands} refers to the last breakpoint,
4572 watchpoint, or catchpoint set (not to the breakpoint most recently
4573 encountered). If the most recent breakpoints were set with a single
4574 command, then the @code{commands} will apply to all the breakpoints
4575 set by that command. This applies to breakpoints set by
4576 @code{rbreak}, and also applies when a single @code{break} command
4577 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4581 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4582 disabled within a @var{command-list}.
4584 You can use breakpoint commands to start your program up again. Simply
4585 use the @code{continue} command, or @code{step}, or any other command
4586 that resumes execution.
4588 Any other commands in the command list, after a command that resumes
4589 execution, are ignored. This is because any time you resume execution
4590 (even with a simple @code{next} or @code{step}), you may encounter
4591 another breakpoint---which could have its own command list, leading to
4592 ambiguities about which list to execute.
4595 If the first command you specify in a command list is @code{silent}, the
4596 usual message about stopping at a breakpoint is not printed. This may
4597 be desirable for breakpoints that are to print a specific message and
4598 then continue. If none of the remaining commands print anything, you
4599 see no sign that the breakpoint was reached. @code{silent} is
4600 meaningful only at the beginning of a breakpoint command list.
4602 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4603 print precisely controlled output, and are often useful in silent
4604 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4606 For example, here is how you could use breakpoint commands to print the
4607 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4613 printf "x is %d\n",x
4618 One application for breakpoint commands is to compensate for one bug so
4619 you can test for another. Put a breakpoint just after the erroneous line
4620 of code, give it a condition to detect the case in which something
4621 erroneous has been done, and give it commands to assign correct values
4622 to any variables that need them. End with the @code{continue} command
4623 so that your program does not stop, and start with the @code{silent}
4624 command so that no output is produced. Here is an example:
4635 @node Dynamic Printf
4636 @subsection Dynamic Printf
4638 @cindex dynamic printf
4640 The dynamic printf command @code{dprintf} combines a breakpoint with
4641 formatted printing of your program's data to give you the effect of
4642 inserting @code{printf} calls into your program on-the-fly, without
4643 having to recompile it.
4645 In its most basic form, the output goes to the GDB console. However,
4646 you can set the variable @code{dprintf-style} for alternate handling.
4647 For instance, you can ask to format the output by calling your
4648 program's @code{printf} function. This has the advantage that the
4649 characters go to the program's output device, so they can recorded in
4650 redirects to files and so forth.
4652 If you are doing remote debugging with a stub or agent, you can also
4653 ask to have the printf handled by the remote agent. In addition to
4654 ensuring that the output goes to the remote program's device along
4655 with any other output the program might produce, you can also ask that
4656 the dprintf remain active even after disconnecting from the remote
4657 target. Using the stub/agent is also more efficient, as it can do
4658 everything without needing to communicate with @value{GDBN}.
4662 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
4663 Whenever execution reaches @var{location}, print the values of one or
4664 more @var{expressions} under the control of the string @var{template}.
4665 To print several values, separate them with commas.
4667 @item set dprintf-style @var{style}
4668 Set the dprintf output to be handled in one of several different
4669 styles enumerated below. A change of style affects all existing
4670 dynamic printfs immediately. (If you need individual control over the
4671 print commands, simply define normal breakpoints with
4672 explicitly-supplied command lists.)
4675 @kindex dprintf-style gdb
4676 Handle the output using the @value{GDBN} @code{printf} command.
4679 @kindex dprintf-style call
4680 Handle the output by calling a function in your program (normally
4684 @kindex dprintf-style agent
4685 Have the remote debugging agent (such as @code{gdbserver}) handle
4686 the output itself. This style is only available for agents that
4687 support running commands on the target.
4689 @item set dprintf-function @var{function}
4690 Set the function to call if the dprintf style is @code{call}. By
4691 default its value is @code{printf}. You may set it to any expression.
4692 that @value{GDBN} can evaluate to a function, as per the @code{call}
4695 @item set dprintf-channel @var{channel}
4696 Set a ``channel'' for dprintf. If set to a non-empty value,
4697 @value{GDBN} will evaluate it as an expression and pass the result as
4698 a first argument to the @code{dprintf-function}, in the manner of
4699 @code{fprintf} and similar functions. Otherwise, the dprintf format
4700 string will be the first argument, in the manner of @code{printf}.
4702 As an example, if you wanted @code{dprintf} output to go to a logfile
4703 that is a standard I/O stream assigned to the variable @code{mylog},
4704 you could do the following:
4707 (gdb) set dprintf-style call
4708 (gdb) set dprintf-function fprintf
4709 (gdb) set dprintf-channel mylog
4710 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
4711 Dprintf 1 at 0x123456: file main.c, line 25.
4713 1 dprintf keep y 0x00123456 in main at main.c:25
4714 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
4719 Note that the @code{info break} displays the dynamic printf commands
4720 as normal breakpoint commands; you can thus easily see the effect of
4721 the variable settings.
4723 @item set disconnected-dprintf on
4724 @itemx set disconnected-dprintf off
4725 @kindex set disconnected-dprintf
4726 Choose whether @code{dprintf} commands should continue to run if
4727 @value{GDBN} has disconnected from the target. This only applies
4728 if the @code{dprintf-style} is @code{agent}.
4730 @item show disconnected-dprintf off
4731 @kindex show disconnected-dprintf
4732 Show the current choice for disconnected @code{dprintf}.
4736 @value{GDBN} does not check the validity of function and channel,
4737 relying on you to supply values that are meaningful for the contexts
4738 in which they are being used. For instance, the function and channel
4739 may be the values of local variables, but if that is the case, then
4740 all enabled dynamic prints must be at locations within the scope of
4741 those locals. If evaluation fails, @value{GDBN} will report an error.
4743 @node Save Breakpoints
4744 @subsection How to save breakpoints to a file
4746 To save breakpoint definitions to a file use the @w{@code{save
4747 breakpoints}} command.
4750 @kindex save breakpoints
4751 @cindex save breakpoints to a file for future sessions
4752 @item save breakpoints [@var{filename}]
4753 This command saves all current breakpoint definitions together with
4754 their commands and ignore counts, into a file @file{@var{filename}}
4755 suitable for use in a later debugging session. This includes all
4756 types of breakpoints (breakpoints, watchpoints, catchpoints,
4757 tracepoints). To read the saved breakpoint definitions, use the
4758 @code{source} command (@pxref{Command Files}). Note that watchpoints
4759 with expressions involving local variables may fail to be recreated
4760 because it may not be possible to access the context where the
4761 watchpoint is valid anymore. Because the saved breakpoint definitions
4762 are simply a sequence of @value{GDBN} commands that recreate the
4763 breakpoints, you can edit the file in your favorite editing program,
4764 and remove the breakpoint definitions you're not interested in, or
4765 that can no longer be recreated.
4768 @node Static Probe Points
4769 @subsection Static Probe Points
4771 @cindex static probe point, SystemTap
4772 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
4773 for Statically Defined Tracing, and the probes are designed to have a tiny
4774 runtime code and data footprint, and no dynamic relocations. They are
4775 usable from assembly, C and C@t{++} languages. See
4776 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
4777 for a good reference on how the @acronym{SDT} probes are implemented.
4779 Currently, @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
4780 @acronym{SDT} probes are supported on ELF-compatible systems. See
4781 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
4782 for more information on how to add @code{SystemTap} @acronym{SDT} probes
4783 in your applications.
4785 @cindex semaphores on static probe points
4786 Some probes have an associated semaphore variable; for instance, this
4787 happens automatically if you defined your probe using a DTrace-style
4788 @file{.d} file. If your probe has a semaphore, @value{GDBN} will
4789 automatically enable it when you specify a breakpoint using the
4790 @samp{-probe-stap} notation. But, if you put a breakpoint at a probe's
4791 location by some other method (e.g., @code{break file:line}), then
4792 @value{GDBN} will not automatically set the semaphore.
4794 You can examine the available static static probes using @code{info
4795 probes}, with optional arguments:
4799 @item info probes stap @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
4800 If given, @var{provider} is a regular expression used to match against provider
4801 names when selecting which probes to list. If omitted, probes by all
4802 probes from all providers are listed.
4804 If given, @var{name} is a regular expression to match against probe names
4805 when selecting which probes to list. If omitted, probe names are not
4806 considered when deciding whether to display them.
4808 If given, @var{objfile} is a regular expression used to select which
4809 object files (executable or shared libraries) to examine. If not
4810 given, all object files are considered.
4812 @item info probes all
4813 List the available static probes, from all types.
4816 @vindex $_probe_arg@r{, convenience variable}
4817 A probe may specify up to twelve arguments. These are available at the
4818 point at which the probe is defined---that is, when the current PC is
4819 at the probe's location. The arguments are available using the
4820 convenience variables (@pxref{Convenience Vars})
4821 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. Each probe argument is
4822 an integer of the appropriate size; types are not preserved. The
4823 convenience variable @code{$_probe_argc} holds the number of arguments
4824 at the current probe point.
4826 These variables are always available, but attempts to access them at
4827 any location other than a probe point will cause @value{GDBN} to give
4831 @c @ifclear BARETARGET
4832 @node Error in Breakpoints
4833 @subsection ``Cannot insert breakpoints''
4835 If you request too many active hardware-assisted breakpoints and
4836 watchpoints, you will see this error message:
4838 @c FIXME: the precise wording of this message may change; the relevant
4839 @c source change is not committed yet (Sep 3, 1999).
4841 Stopped; cannot insert breakpoints.
4842 You may have requested too many hardware breakpoints and watchpoints.
4846 This message is printed when you attempt to resume the program, since
4847 only then @value{GDBN} knows exactly how many hardware breakpoints and
4848 watchpoints it needs to insert.
4850 When this message is printed, you need to disable or remove some of the
4851 hardware-assisted breakpoints and watchpoints, and then continue.
4853 @node Breakpoint-related Warnings
4854 @subsection ``Breakpoint address adjusted...''
4855 @cindex breakpoint address adjusted
4857 Some processor architectures place constraints on the addresses at
4858 which breakpoints may be placed. For architectures thus constrained,
4859 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4860 with the constraints dictated by the architecture.
4862 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4863 a VLIW architecture in which a number of RISC-like instructions may be
4864 bundled together for parallel execution. The FR-V architecture
4865 constrains the location of a breakpoint instruction within such a
4866 bundle to the instruction with the lowest address. @value{GDBN}
4867 honors this constraint by adjusting a breakpoint's address to the
4868 first in the bundle.
4870 It is not uncommon for optimized code to have bundles which contain
4871 instructions from different source statements, thus it may happen that
4872 a breakpoint's address will be adjusted from one source statement to
4873 another. Since this adjustment may significantly alter @value{GDBN}'s
4874 breakpoint related behavior from what the user expects, a warning is
4875 printed when the breakpoint is first set and also when the breakpoint
4878 A warning like the one below is printed when setting a breakpoint
4879 that's been subject to address adjustment:
4882 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4885 Such warnings are printed both for user settable and @value{GDBN}'s
4886 internal breakpoints. If you see one of these warnings, you should
4887 verify that a breakpoint set at the adjusted address will have the
4888 desired affect. If not, the breakpoint in question may be removed and
4889 other breakpoints may be set which will have the desired behavior.
4890 E.g., it may be sufficient to place the breakpoint at a later
4891 instruction. A conditional breakpoint may also be useful in some
4892 cases to prevent the breakpoint from triggering too often.
4894 @value{GDBN} will also issue a warning when stopping at one of these
4895 adjusted breakpoints:
4898 warning: Breakpoint 1 address previously adjusted from 0x00010414
4902 When this warning is encountered, it may be too late to take remedial
4903 action except in cases where the breakpoint is hit earlier or more
4904 frequently than expected.
4906 @node Continuing and Stepping
4907 @section Continuing and Stepping
4911 @cindex resuming execution
4912 @dfn{Continuing} means resuming program execution until your program
4913 completes normally. In contrast, @dfn{stepping} means executing just
4914 one more ``step'' of your program, where ``step'' may mean either one
4915 line of source code, or one machine instruction (depending on what
4916 particular command you use). Either when continuing or when stepping,
4917 your program may stop even sooner, due to a breakpoint or a signal. (If
4918 it stops due to a signal, you may want to use @code{handle}, or use
4919 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4923 @kindex c @r{(@code{continue})}
4924 @kindex fg @r{(resume foreground execution)}
4925 @item continue @r{[}@var{ignore-count}@r{]}
4926 @itemx c @r{[}@var{ignore-count}@r{]}
4927 @itemx fg @r{[}@var{ignore-count}@r{]}
4928 Resume program execution, at the address where your program last stopped;
4929 any breakpoints set at that address are bypassed. The optional argument
4930 @var{ignore-count} allows you to specify a further number of times to
4931 ignore a breakpoint at this location; its effect is like that of
4932 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4934 The argument @var{ignore-count} is meaningful only when your program
4935 stopped due to a breakpoint. At other times, the argument to
4936 @code{continue} is ignored.
4938 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4939 debugged program is deemed to be the foreground program) are provided
4940 purely for convenience, and have exactly the same behavior as
4944 To resume execution at a different place, you can use @code{return}
4945 (@pxref{Returning, ,Returning from a Function}) to go back to the
4946 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4947 Different Address}) to go to an arbitrary location in your program.
4949 A typical technique for using stepping is to set a breakpoint
4950 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4951 beginning of the function or the section of your program where a problem
4952 is believed to lie, run your program until it stops at that breakpoint,
4953 and then step through the suspect area, examining the variables that are
4954 interesting, until you see the problem happen.
4958 @kindex s @r{(@code{step})}
4960 Continue running your program until control reaches a different source
4961 line, then stop it and return control to @value{GDBN}. This command is
4962 abbreviated @code{s}.
4965 @c "without debugging information" is imprecise; actually "without line
4966 @c numbers in the debugging information". (gcc -g1 has debugging info but
4967 @c not line numbers). But it seems complex to try to make that
4968 @c distinction here.
4969 @emph{Warning:} If you use the @code{step} command while control is
4970 within a function that was compiled without debugging information,
4971 execution proceeds until control reaches a function that does have
4972 debugging information. Likewise, it will not step into a function which
4973 is compiled without debugging information. To step through functions
4974 without debugging information, use the @code{stepi} command, described
4978 The @code{step} command only stops at the first instruction of a source
4979 line. This prevents the multiple stops that could otherwise occur in
4980 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4981 to stop if a function that has debugging information is called within
4982 the line. In other words, @code{step} @emph{steps inside} any functions
4983 called within the line.
4985 Also, the @code{step} command only enters a function if there is line
4986 number information for the function. Otherwise it acts like the
4987 @code{next} command. This avoids problems when using @code{cc -gl}
4988 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
4989 was any debugging information about the routine.
4991 @item step @var{count}
4992 Continue running as in @code{step}, but do so @var{count} times. If a
4993 breakpoint is reached, or a signal not related to stepping occurs before
4994 @var{count} steps, stepping stops right away.
4997 @kindex n @r{(@code{next})}
4998 @item next @r{[}@var{count}@r{]}
4999 Continue to the next source line in the current (innermost) stack frame.
5000 This is similar to @code{step}, but function calls that appear within
5001 the line of code are executed without stopping. Execution stops when
5002 control reaches a different line of code at the original stack level
5003 that was executing when you gave the @code{next} command. This command
5004 is abbreviated @code{n}.
5006 An argument @var{count} is a repeat count, as for @code{step}.
5009 @c FIX ME!! Do we delete this, or is there a way it fits in with
5010 @c the following paragraph? --- Vctoria
5012 @c @code{next} within a function that lacks debugging information acts like
5013 @c @code{step}, but any function calls appearing within the code of the
5014 @c function are executed without stopping.
5016 The @code{next} command only stops at the first instruction of a
5017 source line. This prevents multiple stops that could otherwise occur in
5018 @code{switch} statements, @code{for} loops, etc.
5020 @kindex set step-mode
5022 @cindex functions without line info, and stepping
5023 @cindex stepping into functions with no line info
5024 @itemx set step-mode on
5025 The @code{set step-mode on} command causes the @code{step} command to
5026 stop at the first instruction of a function which contains no debug line
5027 information rather than stepping over it.
5029 This is useful in cases where you may be interested in inspecting the
5030 machine instructions of a function which has no symbolic info and do not
5031 want @value{GDBN} to automatically skip over this function.
5033 @item set step-mode off
5034 Causes the @code{step} command to step over any functions which contains no
5035 debug information. This is the default.
5037 @item show step-mode
5038 Show whether @value{GDBN} will stop in or step over functions without
5039 source line debug information.
5042 @kindex fin @r{(@code{finish})}
5044 Continue running until just after function in the selected stack frame
5045 returns. Print the returned value (if any). This command can be
5046 abbreviated as @code{fin}.
5048 Contrast this with the @code{return} command (@pxref{Returning,
5049 ,Returning from a Function}).
5052 @kindex u @r{(@code{until})}
5053 @cindex run until specified location
5056 Continue running until a source line past the current line, in the
5057 current stack frame, is reached. This command is used to avoid single
5058 stepping through a loop more than once. It is like the @code{next}
5059 command, except that when @code{until} encounters a jump, it
5060 automatically continues execution until the program counter is greater
5061 than the address of the jump.
5063 This means that when you reach the end of a loop after single stepping
5064 though it, @code{until} makes your program continue execution until it
5065 exits the loop. In contrast, a @code{next} command at the end of a loop
5066 simply steps back to the beginning of the loop, which forces you to step
5067 through the next iteration.
5069 @code{until} always stops your program if it attempts to exit the current
5072 @code{until} may produce somewhat counterintuitive results if the order
5073 of machine code does not match the order of the source lines. For
5074 example, in the following excerpt from a debugging session, the @code{f}
5075 (@code{frame}) command shows that execution is stopped at line
5076 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5080 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5082 (@value{GDBP}) until
5083 195 for ( ; argc > 0; NEXTARG) @{
5086 This happened because, for execution efficiency, the compiler had
5087 generated code for the loop closure test at the end, rather than the
5088 start, of the loop---even though the test in a C @code{for}-loop is
5089 written before the body of the loop. The @code{until} command appeared
5090 to step back to the beginning of the loop when it advanced to this
5091 expression; however, it has not really gone to an earlier
5092 statement---not in terms of the actual machine code.
5094 @code{until} with no argument works by means of single
5095 instruction stepping, and hence is slower than @code{until} with an
5098 @item until @var{location}
5099 @itemx u @var{location}
5100 Continue running your program until either the specified location is
5101 reached, or the current stack frame returns. @var{location} is any of
5102 the forms described in @ref{Specify Location}.
5103 This form of the command uses temporary breakpoints, and
5104 hence is quicker than @code{until} without an argument. The specified
5105 location is actually reached only if it is in the current frame. This
5106 implies that @code{until} can be used to skip over recursive function
5107 invocations. For instance in the code below, if the current location is
5108 line @code{96}, issuing @code{until 99} will execute the program up to
5109 line @code{99} in the same invocation of factorial, i.e., after the inner
5110 invocations have returned.
5113 94 int factorial (int value)
5115 96 if (value > 1) @{
5116 97 value *= factorial (value - 1);
5123 @kindex advance @var{location}
5124 @item advance @var{location}
5125 Continue running the program up to the given @var{location}. An argument is
5126 required, which should be of one of the forms described in
5127 @ref{Specify Location}.
5128 Execution will also stop upon exit from the current stack
5129 frame. This command is similar to @code{until}, but @code{advance} will
5130 not skip over recursive function calls, and the target location doesn't
5131 have to be in the same frame as the current one.
5135 @kindex si @r{(@code{stepi})}
5137 @itemx stepi @var{arg}
5139 Execute one machine instruction, then stop and return to the debugger.
5141 It is often useful to do @samp{display/i $pc} when stepping by machine
5142 instructions. This makes @value{GDBN} automatically display the next
5143 instruction to be executed, each time your program stops. @xref{Auto
5144 Display,, Automatic Display}.
5146 An argument is a repeat count, as in @code{step}.
5150 @kindex ni @r{(@code{nexti})}
5152 @itemx nexti @var{arg}
5154 Execute one machine instruction, but if it is a function call,
5155 proceed until the function returns.
5157 An argument is a repeat count, as in @code{next}.
5160 @node Skipping Over Functions and Files
5161 @section Skipping Over Functions and Files
5162 @cindex skipping over functions and files
5164 The program you are debugging may contain some functions which are
5165 uninteresting to debug. The @code{skip} comand lets you tell @value{GDBN} to
5166 skip a function or all functions in a file when stepping.
5168 For example, consider the following C function:
5179 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5180 are not interested in stepping through @code{boring}. If you run @code{step}
5181 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5182 step over both @code{foo} and @code{boring}!
5184 One solution is to @code{step} into @code{boring} and use the @code{finish}
5185 command to immediately exit it. But this can become tedious if @code{boring}
5186 is called from many places.
5188 A more flexible solution is to execute @kbd{skip boring}. This instructs
5189 @value{GDBN} never to step into @code{boring}. Now when you execute
5190 @code{step} at line 103, you'll step over @code{boring} and directly into
5193 You can also instruct @value{GDBN} to skip all functions in a file, with, for
5194 example, @code{skip file boring.c}.
5197 @kindex skip function
5198 @item skip @r{[}@var{linespec}@r{]}
5199 @itemx skip function @r{[}@var{linespec}@r{]}
5200 After running this command, the function named by @var{linespec} or the
5201 function containing the line named by @var{linespec} will be skipped over when
5202 stepping. @xref{Specify Location}.
5204 If you do not specify @var{linespec}, the function you're currently debugging
5207 (If you have a function called @code{file} that you want to skip, use
5208 @kbd{skip function file}.)
5211 @item skip file @r{[}@var{filename}@r{]}
5212 After running this command, any function whose source lives in @var{filename}
5213 will be skipped over when stepping.
5215 If you do not specify @var{filename}, functions whose source lives in the file
5216 you're currently debugging will be skipped.
5219 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
5220 These are the commands for managing your list of skips:
5224 @item info skip @r{[}@var{range}@r{]}
5225 Print details about the specified skip(s). If @var{range} is not specified,
5226 print a table with details about all functions and files marked for skipping.
5227 @code{info skip} prints the following information about each skip:
5231 A number identifying this skip.
5233 The type of this skip, either @samp{function} or @samp{file}.
5234 @item Enabled or Disabled
5235 Enabled skips are marked with @samp{y}. Disabled skips are marked with @samp{n}.
5237 For function skips, this column indicates the address in memory of the function
5238 being skipped. If you've set a function skip on a function which has not yet
5239 been loaded, this field will contain @samp{<PENDING>}. Once a shared library
5240 which has the function is loaded, @code{info skip} will show the function's
5243 For file skips, this field contains the filename being skipped. For functions
5244 skips, this field contains the function name and its line number in the file
5245 where it is defined.
5249 @item skip delete @r{[}@var{range}@r{]}
5250 Delete the specified skip(s). If @var{range} is not specified, delete all
5254 @item skip enable @r{[}@var{range}@r{]}
5255 Enable the specified skip(s). If @var{range} is not specified, enable all
5258 @kindex skip disable
5259 @item skip disable @r{[}@var{range}@r{]}
5260 Disable the specified skip(s). If @var{range} is not specified, disable all
5269 A signal is an asynchronous event that can happen in a program. The
5270 operating system defines the possible kinds of signals, and gives each
5271 kind a name and a number. For example, in Unix @code{SIGINT} is the
5272 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
5273 @code{SIGSEGV} is the signal a program gets from referencing a place in
5274 memory far away from all the areas in use; @code{SIGALRM} occurs when
5275 the alarm clock timer goes off (which happens only if your program has
5276 requested an alarm).
5278 @cindex fatal signals
5279 Some signals, including @code{SIGALRM}, are a normal part of the
5280 functioning of your program. Others, such as @code{SIGSEGV}, indicate
5281 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
5282 program has not specified in advance some other way to handle the signal.
5283 @code{SIGINT} does not indicate an error in your program, but it is normally
5284 fatal so it can carry out the purpose of the interrupt: to kill the program.
5286 @value{GDBN} has the ability to detect any occurrence of a signal in your
5287 program. You can tell @value{GDBN} in advance what to do for each kind of
5290 @cindex handling signals
5291 Normally, @value{GDBN} is set up to let the non-erroneous signals like
5292 @code{SIGALRM} be silently passed to your program
5293 (so as not to interfere with their role in the program's functioning)
5294 but to stop your program immediately whenever an error signal happens.
5295 You can change these settings with the @code{handle} command.
5298 @kindex info signals
5302 Print a table of all the kinds of signals and how @value{GDBN} has been told to
5303 handle each one. You can use this to see the signal numbers of all
5304 the defined types of signals.
5306 @item info signals @var{sig}
5307 Similar, but print information only about the specified signal number.
5309 @code{info handle} is an alias for @code{info signals}.
5312 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
5313 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
5314 can be the number of a signal or its name (with or without the
5315 @samp{SIG} at the beginning); a list of signal numbers of the form
5316 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
5317 known signals. Optional arguments @var{keywords}, described below,
5318 say what change to make.
5322 The keywords allowed by the @code{handle} command can be abbreviated.
5323 Their full names are:
5327 @value{GDBN} should not stop your program when this signal happens. It may
5328 still print a message telling you that the signal has come in.
5331 @value{GDBN} should stop your program when this signal happens. This implies
5332 the @code{print} keyword as well.
5335 @value{GDBN} should print a message when this signal happens.
5338 @value{GDBN} should not mention the occurrence of the signal at all. This
5339 implies the @code{nostop} keyword as well.
5343 @value{GDBN} should allow your program to see this signal; your program
5344 can handle the signal, or else it may terminate if the signal is fatal
5345 and not handled. @code{pass} and @code{noignore} are synonyms.
5349 @value{GDBN} should not allow your program to see this signal.
5350 @code{nopass} and @code{ignore} are synonyms.
5354 When a signal stops your program, the signal is not visible to the
5356 continue. Your program sees the signal then, if @code{pass} is in
5357 effect for the signal in question @emph{at that time}. In other words,
5358 after @value{GDBN} reports a signal, you can use the @code{handle}
5359 command with @code{pass} or @code{nopass} to control whether your
5360 program sees that signal when you continue.
5362 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
5363 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
5364 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
5367 You can also use the @code{signal} command to prevent your program from
5368 seeing a signal, or cause it to see a signal it normally would not see,
5369 or to give it any signal at any time. For example, if your program stopped
5370 due to some sort of memory reference error, you might store correct
5371 values into the erroneous variables and continue, hoping to see more
5372 execution; but your program would probably terminate immediately as
5373 a result of the fatal signal once it saw the signal. To prevent this,
5374 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
5377 @cindex extra signal information
5378 @anchor{extra signal information}
5380 On some targets, @value{GDBN} can inspect extra signal information
5381 associated with the intercepted signal, before it is actually
5382 delivered to the program being debugged. This information is exported
5383 by the convenience variable @code{$_siginfo}, and consists of data
5384 that is passed by the kernel to the signal handler at the time of the
5385 receipt of a signal. The data type of the information itself is
5386 target dependent. You can see the data type using the @code{ptype
5387 $_siginfo} command. On Unix systems, it typically corresponds to the
5388 standard @code{siginfo_t} type, as defined in the @file{signal.h}
5391 Here's an example, on a @sc{gnu}/Linux system, printing the stray
5392 referenced address that raised a segmentation fault.
5396 (@value{GDBP}) continue
5397 Program received signal SIGSEGV, Segmentation fault.
5398 0x0000000000400766 in main ()
5400 (@value{GDBP}) ptype $_siginfo
5407 struct @{...@} _kill;
5408 struct @{...@} _timer;
5410 struct @{...@} _sigchld;
5411 struct @{...@} _sigfault;
5412 struct @{...@} _sigpoll;
5415 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
5419 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
5420 $1 = (void *) 0x7ffff7ff7000
5424 Depending on target support, @code{$_siginfo} may also be writable.
5427 @section Stopping and Starting Multi-thread Programs
5429 @cindex stopped threads
5430 @cindex threads, stopped
5432 @cindex continuing threads
5433 @cindex threads, continuing
5435 @value{GDBN} supports debugging programs with multiple threads
5436 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
5437 are two modes of controlling execution of your program within the
5438 debugger. In the default mode, referred to as @dfn{all-stop mode},
5439 when any thread in your program stops (for example, at a breakpoint
5440 or while being stepped), all other threads in the program are also stopped by
5441 @value{GDBN}. On some targets, @value{GDBN} also supports
5442 @dfn{non-stop mode}, in which other threads can continue to run freely while
5443 you examine the stopped thread in the debugger.
5446 * All-Stop Mode:: All threads stop when GDB takes control
5447 * Non-Stop Mode:: Other threads continue to execute
5448 * Background Execution:: Running your program asynchronously
5449 * Thread-Specific Breakpoints:: Controlling breakpoints
5450 * Interrupted System Calls:: GDB may interfere with system calls
5451 * Observer Mode:: GDB does not alter program behavior
5455 @subsection All-Stop Mode
5457 @cindex all-stop mode
5459 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
5460 @emph{all} threads of execution stop, not just the current thread. This
5461 allows you to examine the overall state of the program, including
5462 switching between threads, without worrying that things may change
5465 Conversely, whenever you restart the program, @emph{all} threads start
5466 executing. @emph{This is true even when single-stepping} with commands
5467 like @code{step} or @code{next}.
5469 In particular, @value{GDBN} cannot single-step all threads in lockstep.
5470 Since thread scheduling is up to your debugging target's operating
5471 system (not controlled by @value{GDBN}), other threads may
5472 execute more than one statement while the current thread completes a
5473 single step. Moreover, in general other threads stop in the middle of a
5474 statement, rather than at a clean statement boundary, when the program
5477 You might even find your program stopped in another thread after
5478 continuing or even single-stepping. This happens whenever some other
5479 thread runs into a breakpoint, a signal, or an exception before the
5480 first thread completes whatever you requested.
5482 @cindex automatic thread selection
5483 @cindex switching threads automatically
5484 @cindex threads, automatic switching
5485 Whenever @value{GDBN} stops your program, due to a breakpoint or a
5486 signal, it automatically selects the thread where that breakpoint or
5487 signal happened. @value{GDBN} alerts you to the context switch with a
5488 message such as @samp{[Switching to Thread @var{n}]} to identify the
5491 On some OSes, you can modify @value{GDBN}'s default behavior by
5492 locking the OS scheduler to allow only a single thread to run.
5495 @item set scheduler-locking @var{mode}
5496 @cindex scheduler locking mode
5497 @cindex lock scheduler
5498 Set the scheduler locking mode. If it is @code{off}, then there is no
5499 locking and any thread may run at any time. If @code{on}, then only the
5500 current thread may run when the inferior is resumed. The @code{step}
5501 mode optimizes for single-stepping; it prevents other threads
5502 from preempting the current thread while you are stepping, so that
5503 the focus of debugging does not change unexpectedly.
5504 Other threads only rarely (or never) get a chance to run
5505 when you step. They are more likely to run when you @samp{next} over a
5506 function call, and they are completely free to run when you use commands
5507 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
5508 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
5509 the current thread away from the thread that you are debugging.
5511 @item show scheduler-locking
5512 Display the current scheduler locking mode.
5515 @cindex resume threads of multiple processes simultaneously
5516 By default, when you issue one of the execution commands such as
5517 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
5518 threads of the current inferior to run. For example, if @value{GDBN}
5519 is attached to two inferiors, each with two threads, the
5520 @code{continue} command resumes only the two threads of the current
5521 inferior. This is useful, for example, when you debug a program that
5522 forks and you want to hold the parent stopped (so that, for instance,
5523 it doesn't run to exit), while you debug the child. In other
5524 situations, you may not be interested in inspecting the current state
5525 of any of the processes @value{GDBN} is attached to, and you may want
5526 to resume them all until some breakpoint is hit. In the latter case,
5527 you can instruct @value{GDBN} to allow all threads of all the
5528 inferiors to run with the @w{@code{set schedule-multiple}} command.
5531 @kindex set schedule-multiple
5532 @item set schedule-multiple
5533 Set the mode for allowing threads of multiple processes to be resumed
5534 when an execution command is issued. When @code{on}, all threads of
5535 all processes are allowed to run. When @code{off}, only the threads
5536 of the current process are resumed. The default is @code{off}. The
5537 @code{scheduler-locking} mode takes precedence when set to @code{on},
5538 or while you are stepping and set to @code{step}.
5540 @item show schedule-multiple
5541 Display the current mode for resuming the execution of threads of
5546 @subsection Non-Stop Mode
5548 @cindex non-stop mode
5550 @c This section is really only a place-holder, and needs to be expanded
5551 @c with more details.
5553 For some multi-threaded targets, @value{GDBN} supports an optional
5554 mode of operation in which you can examine stopped program threads in
5555 the debugger while other threads continue to execute freely. This
5556 minimizes intrusion when debugging live systems, such as programs
5557 where some threads have real-time constraints or must continue to
5558 respond to external events. This is referred to as @dfn{non-stop} mode.
5560 In non-stop mode, when a thread stops to report a debugging event,
5561 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5562 threads as well, in contrast to the all-stop mode behavior. Additionally,
5563 execution commands such as @code{continue} and @code{step} apply by default
5564 only to the current thread in non-stop mode, rather than all threads as
5565 in all-stop mode. This allows you to control threads explicitly in
5566 ways that are not possible in all-stop mode --- for example, stepping
5567 one thread while allowing others to run freely, stepping
5568 one thread while holding all others stopped, or stepping several threads
5569 independently and simultaneously.
5571 To enter non-stop mode, use this sequence of commands before you run
5572 or attach to your program:
5575 # Enable the async interface.
5578 # If using the CLI, pagination breaks non-stop.
5581 # Finally, turn it on!
5585 You can use these commands to manipulate the non-stop mode setting:
5588 @kindex set non-stop
5589 @item set non-stop on
5590 Enable selection of non-stop mode.
5591 @item set non-stop off
5592 Disable selection of non-stop mode.
5593 @kindex show non-stop
5595 Show the current non-stop enablement setting.
5598 Note these commands only reflect whether non-stop mode is enabled,
5599 not whether the currently-executing program is being run in non-stop mode.
5600 In particular, the @code{set non-stop} preference is only consulted when
5601 @value{GDBN} starts or connects to the target program, and it is generally
5602 not possible to switch modes once debugging has started. Furthermore,
5603 since not all targets support non-stop mode, even when you have enabled
5604 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
5607 In non-stop mode, all execution commands apply only to the current thread
5608 by default. That is, @code{continue} only continues one thread.
5609 To continue all threads, issue @code{continue -a} or @code{c -a}.
5611 You can use @value{GDBN}'s background execution commands
5612 (@pxref{Background Execution}) to run some threads in the background
5613 while you continue to examine or step others from @value{GDBN}.
5614 The MI execution commands (@pxref{GDB/MI Program Execution}) are
5615 always executed asynchronously in non-stop mode.
5617 Suspending execution is done with the @code{interrupt} command when
5618 running in the background, or @kbd{Ctrl-c} during foreground execution.
5619 In all-stop mode, this stops the whole process;
5620 but in non-stop mode the interrupt applies only to the current thread.
5621 To stop the whole program, use @code{interrupt -a}.
5623 Other execution commands do not currently support the @code{-a} option.
5625 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
5626 that thread current, as it does in all-stop mode. This is because the
5627 thread stop notifications are asynchronous with respect to @value{GDBN}'s
5628 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
5629 changed to a different thread just as you entered a command to operate on the
5630 previously current thread.
5632 @node Background Execution
5633 @subsection Background Execution
5635 @cindex foreground execution
5636 @cindex background execution
5637 @cindex asynchronous execution
5638 @cindex execution, foreground, background and asynchronous
5640 @value{GDBN}'s execution commands have two variants: the normal
5641 foreground (synchronous) behavior, and a background
5642 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
5643 the program to report that some thread has stopped before prompting for
5644 another command. In background execution, @value{GDBN} immediately gives
5645 a command prompt so that you can issue other commands while your program runs.
5647 You need to explicitly enable asynchronous mode before you can use
5648 background execution commands. You can use these commands to
5649 manipulate the asynchronous mode setting:
5652 @kindex set target-async
5653 @item set target-async on
5654 Enable asynchronous mode.
5655 @item set target-async off
5656 Disable asynchronous mode.
5657 @kindex show target-async
5658 @item show target-async
5659 Show the current target-async setting.
5662 If the target doesn't support async mode, @value{GDBN} issues an error
5663 message if you attempt to use the background execution commands.
5665 To specify background execution, add a @code{&} to the command. For example,
5666 the background form of the @code{continue} command is @code{continue&}, or
5667 just @code{c&}. The execution commands that accept background execution
5673 @xref{Starting, , Starting your Program}.
5677 @xref{Attach, , Debugging an Already-running Process}.
5681 @xref{Continuing and Stepping, step}.
5685 @xref{Continuing and Stepping, stepi}.
5689 @xref{Continuing and Stepping, next}.
5693 @xref{Continuing and Stepping, nexti}.
5697 @xref{Continuing and Stepping, continue}.
5701 @xref{Continuing and Stepping, finish}.
5705 @xref{Continuing and Stepping, until}.
5709 Background execution is especially useful in conjunction with non-stop
5710 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
5711 However, you can also use these commands in the normal all-stop mode with
5712 the restriction that you cannot issue another execution command until the
5713 previous one finishes. Examples of commands that are valid in all-stop
5714 mode while the program is running include @code{help} and @code{info break}.
5716 You can interrupt your program while it is running in the background by
5717 using the @code{interrupt} command.
5724 Suspend execution of the running program. In all-stop mode,
5725 @code{interrupt} stops the whole process, but in non-stop mode, it stops
5726 only the current thread. To stop the whole program in non-stop mode,
5727 use @code{interrupt -a}.
5730 @node Thread-Specific Breakpoints
5731 @subsection Thread-Specific Breakpoints
5733 When your program has multiple threads (@pxref{Threads,, Debugging
5734 Programs with Multiple Threads}), you can choose whether to set
5735 breakpoints on all threads, or on a particular thread.
5738 @cindex breakpoints and threads
5739 @cindex thread breakpoints
5740 @kindex break @dots{} thread @var{threadno}
5741 @item break @var{linespec} thread @var{threadno}
5742 @itemx break @var{linespec} thread @var{threadno} if @dots{}
5743 @var{linespec} specifies source lines; there are several ways of
5744 writing them (@pxref{Specify Location}), but the effect is always to
5745 specify some source line.
5747 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
5748 to specify that you only want @value{GDBN} to stop the program when a
5749 particular thread reaches this breakpoint. @var{threadno} is one of the
5750 numeric thread identifiers assigned by @value{GDBN}, shown in the first
5751 column of the @samp{info threads} display.
5753 If you do not specify @samp{thread @var{threadno}} when you set a
5754 breakpoint, the breakpoint applies to @emph{all} threads of your
5757 You can use the @code{thread} qualifier on conditional breakpoints as
5758 well; in this case, place @samp{thread @var{threadno}} before or
5759 after the breakpoint condition, like this:
5762 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
5767 @node Interrupted System Calls
5768 @subsection Interrupted System Calls
5770 @cindex thread breakpoints and system calls
5771 @cindex system calls and thread breakpoints
5772 @cindex premature return from system calls
5773 There is an unfortunate side effect when using @value{GDBN} to debug
5774 multi-threaded programs. If one thread stops for a
5775 breakpoint, or for some other reason, and another thread is blocked in a
5776 system call, then the system call may return prematurely. This is a
5777 consequence of the interaction between multiple threads and the signals
5778 that @value{GDBN} uses to implement breakpoints and other events that
5781 To handle this problem, your program should check the return value of
5782 each system call and react appropriately. This is good programming
5785 For example, do not write code like this:
5791 The call to @code{sleep} will return early if a different thread stops
5792 at a breakpoint or for some other reason.
5794 Instead, write this:
5799 unslept = sleep (unslept);
5802 A system call is allowed to return early, so the system is still
5803 conforming to its specification. But @value{GDBN} does cause your
5804 multi-threaded program to behave differently than it would without
5807 Also, @value{GDBN} uses internal breakpoints in the thread library to
5808 monitor certain events such as thread creation and thread destruction.
5809 When such an event happens, a system call in another thread may return
5810 prematurely, even though your program does not appear to stop.
5813 @subsection Observer Mode
5815 If you want to build on non-stop mode and observe program behavior
5816 without any chance of disruption by @value{GDBN}, you can set
5817 variables to disable all of the debugger's attempts to modify state,
5818 whether by writing memory, inserting breakpoints, etc. These operate
5819 at a low level, intercepting operations from all commands.
5821 When all of these are set to @code{off}, then @value{GDBN} is said to
5822 be @dfn{observer mode}. As a convenience, the variable
5823 @code{observer} can be set to disable these, plus enable non-stop
5826 Note that @value{GDBN} will not prevent you from making nonsensical
5827 combinations of these settings. For instance, if you have enabled
5828 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
5829 then breakpoints that work by writing trap instructions into the code
5830 stream will still not be able to be placed.
5835 @item set observer on
5836 @itemx set observer off
5837 When set to @code{on}, this disables all the permission variables
5838 below (except for @code{insert-fast-tracepoints}), plus enables
5839 non-stop debugging. Setting this to @code{off} switches back to
5840 normal debugging, though remaining in non-stop mode.
5843 Show whether observer mode is on or off.
5845 @kindex may-write-registers
5846 @item set may-write-registers on
5847 @itemx set may-write-registers off
5848 This controls whether @value{GDBN} will attempt to alter the values of
5849 registers, such as with assignment expressions in @code{print}, or the
5850 @code{jump} command. It defaults to @code{on}.
5852 @item show may-write-registers
5853 Show the current permission to write registers.
5855 @kindex may-write-memory
5856 @item set may-write-memory on
5857 @itemx set may-write-memory off
5858 This controls whether @value{GDBN} will attempt to alter the contents
5859 of memory, such as with assignment expressions in @code{print}. It
5860 defaults to @code{on}.
5862 @item show may-write-memory
5863 Show the current permission to write memory.
5865 @kindex may-insert-breakpoints
5866 @item set may-insert-breakpoints on
5867 @itemx set may-insert-breakpoints off
5868 This controls whether @value{GDBN} will attempt to insert breakpoints.
5869 This affects all breakpoints, including internal breakpoints defined
5870 by @value{GDBN}. It defaults to @code{on}.
5872 @item show may-insert-breakpoints
5873 Show the current permission to insert breakpoints.
5875 @kindex may-insert-tracepoints
5876 @item set may-insert-tracepoints on
5877 @itemx set may-insert-tracepoints off
5878 This controls whether @value{GDBN} will attempt to insert (regular)
5879 tracepoints at the beginning of a tracing experiment. It affects only
5880 non-fast tracepoints, fast tracepoints being under the control of
5881 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
5883 @item show may-insert-tracepoints
5884 Show the current permission to insert tracepoints.
5886 @kindex may-insert-fast-tracepoints
5887 @item set may-insert-fast-tracepoints on
5888 @itemx set may-insert-fast-tracepoints off
5889 This controls whether @value{GDBN} will attempt to insert fast
5890 tracepoints at the beginning of a tracing experiment. It affects only
5891 fast tracepoints, regular (non-fast) tracepoints being under the
5892 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
5894 @item show may-insert-fast-tracepoints
5895 Show the current permission to insert fast tracepoints.
5897 @kindex may-interrupt
5898 @item set may-interrupt on
5899 @itemx set may-interrupt off
5900 This controls whether @value{GDBN} will attempt to interrupt or stop
5901 program execution. When this variable is @code{off}, the
5902 @code{interrupt} command will have no effect, nor will
5903 @kbd{Ctrl-c}. It defaults to @code{on}.
5905 @item show may-interrupt
5906 Show the current permission to interrupt or stop the program.
5910 @node Reverse Execution
5911 @chapter Running programs backward
5912 @cindex reverse execution
5913 @cindex running programs backward
5915 When you are debugging a program, it is not unusual to realize that
5916 you have gone too far, and some event of interest has already happened.
5917 If the target environment supports it, @value{GDBN} can allow you to
5918 ``rewind'' the program by running it backward.
5920 A target environment that supports reverse execution should be able
5921 to ``undo'' the changes in machine state that have taken place as the
5922 program was executing normally. Variables, registers etc.@: should
5923 revert to their previous values. Obviously this requires a great
5924 deal of sophistication on the part of the target environment; not
5925 all target environments can support reverse execution.
5927 When a program is executed in reverse, the instructions that
5928 have most recently been executed are ``un-executed'', in reverse
5929 order. The program counter runs backward, following the previous
5930 thread of execution in reverse. As each instruction is ``un-executed'',
5931 the values of memory and/or registers that were changed by that
5932 instruction are reverted to their previous states. After executing
5933 a piece of source code in reverse, all side effects of that code
5934 should be ``undone'', and all variables should be returned to their
5935 prior values@footnote{
5936 Note that some side effects are easier to undo than others. For instance,
5937 memory and registers are relatively easy, but device I/O is hard. Some
5938 targets may be able undo things like device I/O, and some may not.
5940 The contract between @value{GDBN} and the reverse executing target
5941 requires only that the target do something reasonable when
5942 @value{GDBN} tells it to execute backwards, and then report the
5943 results back to @value{GDBN}. Whatever the target reports back to
5944 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
5945 assumes that the memory and registers that the target reports are in a
5946 consistant state, but @value{GDBN} accepts whatever it is given.
5949 If you are debugging in a target environment that supports
5950 reverse execution, @value{GDBN} provides the following commands.
5953 @kindex reverse-continue
5954 @kindex rc @r{(@code{reverse-continue})}
5955 @item reverse-continue @r{[}@var{ignore-count}@r{]}
5956 @itemx rc @r{[}@var{ignore-count}@r{]}
5957 Beginning at the point where your program last stopped, start executing
5958 in reverse. Reverse execution will stop for breakpoints and synchronous
5959 exceptions (signals), just like normal execution. Behavior of
5960 asynchronous signals depends on the target environment.
5962 @kindex reverse-step
5963 @kindex rs @r{(@code{step})}
5964 @item reverse-step @r{[}@var{count}@r{]}
5965 Run the program backward until control reaches the start of a
5966 different source line; then stop it, and return control to @value{GDBN}.
5968 Like the @code{step} command, @code{reverse-step} will only stop
5969 at the beginning of a source line. It ``un-executes'' the previously
5970 executed source line. If the previous source line included calls to
5971 debuggable functions, @code{reverse-step} will step (backward) into
5972 the called function, stopping at the beginning of the @emph{last}
5973 statement in the called function (typically a return statement).
5975 Also, as with the @code{step} command, if non-debuggable functions are
5976 called, @code{reverse-step} will run thru them backward without stopping.
5978 @kindex reverse-stepi
5979 @kindex rsi @r{(@code{reverse-stepi})}
5980 @item reverse-stepi @r{[}@var{count}@r{]}
5981 Reverse-execute one machine instruction. Note that the instruction
5982 to be reverse-executed is @emph{not} the one pointed to by the program
5983 counter, but the instruction executed prior to that one. For instance,
5984 if the last instruction was a jump, @code{reverse-stepi} will take you
5985 back from the destination of the jump to the jump instruction itself.
5987 @kindex reverse-next
5988 @kindex rn @r{(@code{reverse-next})}
5989 @item reverse-next @r{[}@var{count}@r{]}
5990 Run backward to the beginning of the previous line executed in
5991 the current (innermost) stack frame. If the line contains function
5992 calls, they will be ``un-executed'' without stopping. Starting from
5993 the first line of a function, @code{reverse-next} will take you back
5994 to the caller of that function, @emph{before} the function was called,
5995 just as the normal @code{next} command would take you from the last
5996 line of a function back to its return to its caller
5997 @footnote{Unless the code is too heavily optimized.}.
5999 @kindex reverse-nexti
6000 @kindex rni @r{(@code{reverse-nexti})}
6001 @item reverse-nexti @r{[}@var{count}@r{]}
6002 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
6003 in reverse, except that called functions are ``un-executed'' atomically.
6004 That is, if the previously executed instruction was a return from
6005 another function, @code{reverse-nexti} will continue to execute
6006 in reverse until the call to that function (from the current stack
6009 @kindex reverse-finish
6010 @item reverse-finish
6011 Just as the @code{finish} command takes you to the point where the
6012 current function returns, @code{reverse-finish} takes you to the point
6013 where it was called. Instead of ending up at the end of the current
6014 function invocation, you end up at the beginning.
6016 @kindex set exec-direction
6017 @item set exec-direction
6018 Set the direction of target execution.
6019 @item set exec-direction reverse
6020 @cindex execute forward or backward in time
6021 @value{GDBN} will perform all execution commands in reverse, until the
6022 exec-direction mode is changed to ``forward''. Affected commands include
6023 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
6024 command cannot be used in reverse mode.
6025 @item set exec-direction forward
6026 @value{GDBN} will perform all execution commands in the normal fashion.
6027 This is the default.
6031 @node Process Record and Replay
6032 @chapter Recording Inferior's Execution and Replaying It
6033 @cindex process record and replay
6034 @cindex recording inferior's execution and replaying it
6036 On some platforms, @value{GDBN} provides a special @dfn{process record
6037 and replay} target that can record a log of the process execution, and
6038 replay it later with both forward and reverse execution commands.
6041 When this target is in use, if the execution log includes the record
6042 for the next instruction, @value{GDBN} will debug in @dfn{replay
6043 mode}. In the replay mode, the inferior does not really execute code
6044 instructions. Instead, all the events that normally happen during
6045 code execution are taken from the execution log. While code is not
6046 really executed in replay mode, the values of registers (including the
6047 program counter register) and the memory of the inferior are still
6048 changed as they normally would. Their contents are taken from the
6052 If the record for the next instruction is not in the execution log,
6053 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6054 inferior executes normally, and @value{GDBN} records the execution log
6057 The process record and replay target supports reverse execution
6058 (@pxref{Reverse Execution}), even if the platform on which the
6059 inferior runs does not. However, the reverse execution is limited in
6060 this case by the range of the instructions recorded in the execution
6061 log. In other words, reverse execution on platforms that don't
6062 support it directly can only be done in the replay mode.
6064 When debugging in the reverse direction, @value{GDBN} will work in
6065 replay mode as long as the execution log includes the record for the
6066 previous instruction; otherwise, it will work in record mode, if the
6067 platform supports reverse execution, or stop if not.
6069 For architecture environments that support process record and replay,
6070 @value{GDBN} provides the following commands:
6073 @kindex target record
6077 This command starts the process record and replay target. The process
6078 record and replay target can only debug a process that is already
6079 running. Therefore, you need first to start the process with the
6080 @kbd{run} or @kbd{start} commands, and then start the recording with
6081 the @kbd{target record} command.
6083 Both @code{record} and @code{rec} are aliases of @code{target record}.
6085 @cindex displaced stepping, and process record and replay
6086 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
6087 will be automatically disabled when process record and replay target
6088 is started. That's because the process record and replay target
6089 doesn't support displaced stepping.
6091 @cindex non-stop mode, and process record and replay
6092 @cindex asynchronous execution, and process record and replay
6093 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
6094 the asynchronous execution mode (@pxref{Background Execution}), the
6095 process record and replay target cannot be started because it doesn't
6096 support these two modes.
6101 Stop the process record and replay target. When process record and
6102 replay target stops, the entire execution log will be deleted and the
6103 inferior will either be terminated, or will remain in its final state.
6105 When you stop the process record and replay target in record mode (at
6106 the end of the execution log), the inferior will be stopped at the
6107 next instruction that would have been recorded. In other words, if
6108 you record for a while and then stop recording, the inferior process
6109 will be left in the same state as if the recording never happened.
6111 On the other hand, if the process record and replay target is stopped
6112 while in replay mode (that is, not at the end of the execution log,
6113 but at some earlier point), the inferior process will become ``live''
6114 at that earlier state, and it will then be possible to continue the
6115 usual ``live'' debugging of the process from that state.
6117 When the inferior process exits, or @value{GDBN} detaches from it,
6118 process record and replay target will automatically stop itself.
6121 @item record save @var{filename}
6122 Save the execution log to a file @file{@var{filename}}.
6123 Default filename is @file{gdb_record.@var{process_id}}, where
6124 @var{process_id} is the process ID of the inferior.
6126 @kindex record restore
6127 @item record restore @var{filename}
6128 Restore the execution log from a file @file{@var{filename}}.
6129 File must have been created with @code{record save}.
6131 @kindex set record insn-number-max
6132 @item set record insn-number-max @var{limit}
6133 Set the limit of instructions to be recorded. Default value is 200000.
6135 If @var{limit} is a positive number, then @value{GDBN} will start
6136 deleting instructions from the log once the number of the record
6137 instructions becomes greater than @var{limit}. For every new recorded
6138 instruction, @value{GDBN} will delete the earliest recorded
6139 instruction to keep the number of recorded instructions at the limit.
6140 (Since deleting recorded instructions loses information, @value{GDBN}
6141 lets you control what happens when the limit is reached, by means of
6142 the @code{stop-at-limit} option, described below.)
6144 If @var{limit} is zero, @value{GDBN} will never delete recorded
6145 instructions from the execution log. The number of recorded
6146 instructions is unlimited in this case.
6148 @kindex show record insn-number-max
6149 @item show record insn-number-max
6150 Show the limit of instructions to be recorded.
6152 @kindex set record stop-at-limit
6153 @item set record stop-at-limit
6154 Control the behavior when the number of recorded instructions reaches
6155 the limit. If ON (the default), @value{GDBN} will stop when the limit
6156 is reached for the first time and ask you whether you want to stop the
6157 inferior or continue running it and recording the execution log. If
6158 you decide to continue recording, each new recorded instruction will
6159 cause the oldest one to be deleted.
6161 If this option is OFF, @value{GDBN} will automatically delete the
6162 oldest record to make room for each new one, without asking.
6164 @kindex show record stop-at-limit
6165 @item show record stop-at-limit
6166 Show the current setting of @code{stop-at-limit}.
6168 @kindex set record memory-query
6169 @item set record memory-query
6170 Control the behavior when @value{GDBN} is unable to record memory
6171 changes caused by an instruction. If ON, @value{GDBN} will query
6172 whether to stop the inferior in that case.
6174 If this option is OFF (the default), @value{GDBN} will automatically
6175 ignore the effect of such instructions on memory. Later, when
6176 @value{GDBN} replays this execution log, it will mark the log of this
6177 instruction as not accessible, and it will not affect the replay
6180 @kindex show record memory-query
6181 @item show record memory-query
6182 Show the current setting of @code{memory-query}.
6186 Show various statistics about the state of process record and its
6187 in-memory execution log buffer, including:
6191 Whether in record mode or replay mode.
6193 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
6195 Highest recorded instruction number.
6197 Current instruction about to be replayed (if in replay mode).
6199 Number of instructions contained in the execution log.
6201 Maximum number of instructions that may be contained in the execution log.
6204 @kindex record delete
6207 When record target runs in replay mode (``in the past''), delete the
6208 subsequent execution log and begin to record a new execution log starting
6209 from the current address. This means you will abandon the previously
6210 recorded ``future'' and begin recording a new ``future''.
6215 @chapter Examining the Stack
6217 When your program has stopped, the first thing you need to know is where it
6218 stopped and how it got there.
6221 Each time your program performs a function call, information about the call
6223 That information includes the location of the call in your program,
6224 the arguments of the call,
6225 and the local variables of the function being called.
6226 The information is saved in a block of data called a @dfn{stack frame}.
6227 The stack frames are allocated in a region of memory called the @dfn{call
6230 When your program stops, the @value{GDBN} commands for examining the
6231 stack allow you to see all of this information.
6233 @cindex selected frame
6234 One of the stack frames is @dfn{selected} by @value{GDBN} and many
6235 @value{GDBN} commands refer implicitly to the selected frame. In
6236 particular, whenever you ask @value{GDBN} for the value of a variable in
6237 your program, the value is found in the selected frame. There are
6238 special @value{GDBN} commands to select whichever frame you are
6239 interested in. @xref{Selection, ,Selecting a Frame}.
6241 When your program stops, @value{GDBN} automatically selects the
6242 currently executing frame and describes it briefly, similar to the
6243 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
6246 * Frames:: Stack frames
6247 * Backtrace:: Backtraces
6248 * Selection:: Selecting a frame
6249 * Frame Info:: Information on a frame
6254 @section Stack Frames
6256 @cindex frame, definition
6258 The call stack is divided up into contiguous pieces called @dfn{stack
6259 frames}, or @dfn{frames} for short; each frame is the data associated
6260 with one call to one function. The frame contains the arguments given
6261 to the function, the function's local variables, and the address at
6262 which the function is executing.
6264 @cindex initial frame
6265 @cindex outermost frame
6266 @cindex innermost frame
6267 When your program is started, the stack has only one frame, that of the
6268 function @code{main}. This is called the @dfn{initial} frame or the
6269 @dfn{outermost} frame. Each time a function is called, a new frame is
6270 made. Each time a function returns, the frame for that function invocation
6271 is eliminated. If a function is recursive, there can be many frames for
6272 the same function. The frame for the function in which execution is
6273 actually occurring is called the @dfn{innermost} frame. This is the most
6274 recently created of all the stack frames that still exist.
6276 @cindex frame pointer
6277 Inside your program, stack frames are identified by their addresses. A
6278 stack frame consists of many bytes, each of which has its own address; each
6279 kind of computer has a convention for choosing one byte whose
6280 address serves as the address of the frame. Usually this address is kept
6281 in a register called the @dfn{frame pointer register}
6282 (@pxref{Registers, $fp}) while execution is going on in that frame.
6284 @cindex frame number
6285 @value{GDBN} assigns numbers to all existing stack frames, starting with
6286 zero for the innermost frame, one for the frame that called it,
6287 and so on upward. These numbers do not really exist in your program;
6288 they are assigned by @value{GDBN} to give you a way of designating stack
6289 frames in @value{GDBN} commands.
6291 @c The -fomit-frame-pointer below perennially causes hbox overflow
6292 @c underflow problems.
6293 @cindex frameless execution
6294 Some compilers provide a way to compile functions so that they operate
6295 without stack frames. (For example, the @value{NGCC} option
6297 @samp{-fomit-frame-pointer}
6299 generates functions without a frame.)
6300 This is occasionally done with heavily used library functions to save
6301 the frame setup time. @value{GDBN} has limited facilities for dealing
6302 with these function invocations. If the innermost function invocation
6303 has no stack frame, @value{GDBN} nevertheless regards it as though
6304 it had a separate frame, which is numbered zero as usual, allowing
6305 correct tracing of the function call chain. However, @value{GDBN} has
6306 no provision for frameless functions elsewhere in the stack.
6309 @kindex frame@r{, command}
6310 @cindex current stack frame
6311 @item frame @var{args}
6312 The @code{frame} command allows you to move from one stack frame to another,
6313 and to print the stack frame you select. @var{args} may be either the
6314 address of the frame or the stack frame number. Without an argument,
6315 @code{frame} prints the current stack frame.
6317 @kindex select-frame
6318 @cindex selecting frame silently
6320 The @code{select-frame} command allows you to move from one stack frame
6321 to another without printing the frame. This is the silent version of
6329 @cindex call stack traces
6330 A backtrace is a summary of how your program got where it is. It shows one
6331 line per frame, for many frames, starting with the currently executing
6332 frame (frame zero), followed by its caller (frame one), and on up the
6337 @kindex bt @r{(@code{backtrace})}
6340 Print a backtrace of the entire stack: one line per frame for all
6341 frames in the stack.
6343 You can stop the backtrace at any time by typing the system interrupt
6344 character, normally @kbd{Ctrl-c}.
6346 @item backtrace @var{n}
6348 Similar, but print only the innermost @var{n} frames.
6350 @item backtrace -@var{n}
6352 Similar, but print only the outermost @var{n} frames.
6354 @item backtrace full
6356 @itemx bt full @var{n}
6357 @itemx bt full -@var{n}
6358 Print the values of the local variables also. @var{n} specifies the
6359 number of frames to print, as described above.
6364 The names @code{where} and @code{info stack} (abbreviated @code{info s})
6365 are additional aliases for @code{backtrace}.
6367 @cindex multiple threads, backtrace
6368 In a multi-threaded program, @value{GDBN} by default shows the
6369 backtrace only for the current thread. To display the backtrace for
6370 several or all of the threads, use the command @code{thread apply}
6371 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
6372 apply all backtrace}, @value{GDBN} will display the backtrace for all
6373 the threads; this is handy when you debug a core dump of a
6374 multi-threaded program.
6376 Each line in the backtrace shows the frame number and the function name.
6377 The program counter value is also shown---unless you use @code{set
6378 print address off}. The backtrace also shows the source file name and
6379 line number, as well as the arguments to the function. The program
6380 counter value is omitted if it is at the beginning of the code for that
6383 Here is an example of a backtrace. It was made with the command
6384 @samp{bt 3}, so it shows the innermost three frames.
6388 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6390 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
6391 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
6393 (More stack frames follow...)
6398 The display for frame zero does not begin with a program counter
6399 value, indicating that your program has stopped at the beginning of the
6400 code for line @code{993} of @code{builtin.c}.
6403 The value of parameter @code{data} in frame 1 has been replaced by
6404 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
6405 only if it is a scalar (integer, pointer, enumeration, etc). See command
6406 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
6407 on how to configure the way function parameter values are printed.
6409 @cindex optimized out, in backtrace
6410 @cindex function call arguments, optimized out
6411 If your program was compiled with optimizations, some compilers will
6412 optimize away arguments passed to functions if those arguments are
6413 never used after the call. Such optimizations generate code that
6414 passes arguments through registers, but doesn't store those arguments
6415 in the stack frame. @value{GDBN} has no way of displaying such
6416 arguments in stack frames other than the innermost one. Here's what
6417 such a backtrace might look like:
6421 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6423 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
6424 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
6426 (More stack frames follow...)
6431 The values of arguments that were not saved in their stack frames are
6432 shown as @samp{<optimized out>}.
6434 If you need to display the values of such optimized-out arguments,
6435 either deduce that from other variables whose values depend on the one
6436 you are interested in, or recompile without optimizations.
6438 @cindex backtrace beyond @code{main} function
6439 @cindex program entry point
6440 @cindex startup code, and backtrace
6441 Most programs have a standard user entry point---a place where system
6442 libraries and startup code transition into user code. For C this is
6443 @code{main}@footnote{
6444 Note that embedded programs (the so-called ``free-standing''
6445 environment) are not required to have a @code{main} function as the
6446 entry point. They could even have multiple entry points.}.
6447 When @value{GDBN} finds the entry function in a backtrace
6448 it will terminate the backtrace, to avoid tracing into highly
6449 system-specific (and generally uninteresting) code.
6451 If you need to examine the startup code, or limit the number of levels
6452 in a backtrace, you can change this behavior:
6455 @item set backtrace past-main
6456 @itemx set backtrace past-main on
6457 @kindex set backtrace
6458 Backtraces will continue past the user entry point.
6460 @item set backtrace past-main off
6461 Backtraces will stop when they encounter the user entry point. This is the
6464 @item show backtrace past-main
6465 @kindex show backtrace
6466 Display the current user entry point backtrace policy.
6468 @item set backtrace past-entry
6469 @itemx set backtrace past-entry on
6470 Backtraces will continue past the internal entry point of an application.
6471 This entry point is encoded by the linker when the application is built,
6472 and is likely before the user entry point @code{main} (or equivalent) is called.
6474 @item set backtrace past-entry off
6475 Backtraces will stop when they encounter the internal entry point of an
6476 application. This is the default.
6478 @item show backtrace past-entry
6479 Display the current internal entry point backtrace policy.
6481 @item set backtrace limit @var{n}
6482 @itemx set backtrace limit 0
6483 @cindex backtrace limit
6484 Limit the backtrace to @var{n} levels. A value of zero means
6487 @item show backtrace limit
6488 Display the current limit on backtrace levels.
6492 @section Selecting a Frame
6494 Most commands for examining the stack and other data in your program work on
6495 whichever stack frame is selected at the moment. Here are the commands for
6496 selecting a stack frame; all of them finish by printing a brief description
6497 of the stack frame just selected.
6500 @kindex frame@r{, selecting}
6501 @kindex f @r{(@code{frame})}
6504 Select frame number @var{n}. Recall that frame zero is the innermost
6505 (currently executing) frame, frame one is the frame that called the
6506 innermost one, and so on. The highest-numbered frame is the one for
6509 @item frame @var{addr}
6511 Select the frame at address @var{addr}. This is useful mainly if the
6512 chaining of stack frames has been damaged by a bug, making it
6513 impossible for @value{GDBN} to assign numbers properly to all frames. In
6514 addition, this can be useful when your program has multiple stacks and
6515 switches between them.
6517 On the SPARC architecture, @code{frame} needs two addresses to
6518 select an arbitrary frame: a frame pointer and a stack pointer.
6520 On the @acronym{MIPS} and Alpha architecture, it needs two addresses: a stack
6521 pointer and a program counter.
6523 On the 29k architecture, it needs three addresses: a register stack
6524 pointer, a program counter, and a memory stack pointer.
6528 Move @var{n} frames up the stack. For positive numbers @var{n}, this
6529 advances toward the outermost frame, to higher frame numbers, to frames
6530 that have existed longer. @var{n} defaults to one.
6533 @kindex do @r{(@code{down})}
6535 Move @var{n} frames down the stack. For positive numbers @var{n}, this
6536 advances toward the innermost frame, to lower frame numbers, to frames
6537 that were created more recently. @var{n} defaults to one. You may
6538 abbreviate @code{down} as @code{do}.
6541 All of these commands end by printing two lines of output describing the
6542 frame. The first line shows the frame number, the function name, the
6543 arguments, and the source file and line number of execution in that
6544 frame. The second line shows the text of that source line.
6552 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
6554 10 read_input_file (argv[i]);
6558 After such a printout, the @code{list} command with no arguments
6559 prints ten lines centered on the point of execution in the frame.
6560 You can also edit the program at the point of execution with your favorite
6561 editing program by typing @code{edit}.
6562 @xref{List, ,Printing Source Lines},
6566 @kindex down-silently
6568 @item up-silently @var{n}
6569 @itemx down-silently @var{n}
6570 These two commands are variants of @code{up} and @code{down},
6571 respectively; they differ in that they do their work silently, without
6572 causing display of the new frame. They are intended primarily for use
6573 in @value{GDBN} command scripts, where the output might be unnecessary and
6578 @section Information About a Frame
6580 There are several other commands to print information about the selected
6586 When used without any argument, this command does not change which
6587 frame is selected, but prints a brief description of the currently
6588 selected stack frame. It can be abbreviated @code{f}. With an
6589 argument, this command is used to select a stack frame.
6590 @xref{Selection, ,Selecting a Frame}.
6593 @kindex info f @r{(@code{info frame})}
6596 This command prints a verbose description of the selected stack frame,
6601 the address of the frame
6603 the address of the next frame down (called by this frame)
6605 the address of the next frame up (caller of this frame)
6607 the language in which the source code corresponding to this frame is written
6609 the address of the frame's arguments
6611 the address of the frame's local variables
6613 the program counter saved in it (the address of execution in the caller frame)
6615 which registers were saved in the frame
6618 @noindent The verbose description is useful when
6619 something has gone wrong that has made the stack format fail to fit
6620 the usual conventions.
6622 @item info frame @var{addr}
6623 @itemx info f @var{addr}
6624 Print a verbose description of the frame at address @var{addr}, without
6625 selecting that frame. The selected frame remains unchanged by this
6626 command. This requires the same kind of address (more than one for some
6627 architectures) that you specify in the @code{frame} command.
6628 @xref{Selection, ,Selecting a Frame}.
6632 Print the arguments of the selected frame, each on a separate line.
6636 Print the local variables of the selected frame, each on a separate
6637 line. These are all variables (declared either static or automatic)
6638 accessible at the point of execution of the selected frame.
6644 @chapter Examining Source Files
6646 @value{GDBN} can print parts of your program's source, since the debugging
6647 information recorded in the program tells @value{GDBN} what source files were
6648 used to build it. When your program stops, @value{GDBN} spontaneously prints
6649 the line where it stopped. Likewise, when you select a stack frame
6650 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
6651 execution in that frame has stopped. You can print other portions of
6652 source files by explicit command.
6654 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
6655 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
6656 @value{GDBN} under @sc{gnu} Emacs}.
6659 * List:: Printing source lines
6660 * Specify Location:: How to specify code locations
6661 * Edit:: Editing source files
6662 * Search:: Searching source files
6663 * Source Path:: Specifying source directories
6664 * Machine Code:: Source and machine code
6668 @section Printing Source Lines
6671 @kindex l @r{(@code{list})}
6672 To print lines from a source file, use the @code{list} command
6673 (abbreviated @code{l}). By default, ten lines are printed.
6674 There are several ways to specify what part of the file you want to
6675 print; see @ref{Specify Location}, for the full list.
6677 Here are the forms of the @code{list} command most commonly used:
6680 @item list @var{linenum}
6681 Print lines centered around line number @var{linenum} in the
6682 current source file.
6684 @item list @var{function}
6685 Print lines centered around the beginning of function
6689 Print more lines. If the last lines printed were printed with a
6690 @code{list} command, this prints lines following the last lines
6691 printed; however, if the last line printed was a solitary line printed
6692 as part of displaying a stack frame (@pxref{Stack, ,Examining the
6693 Stack}), this prints lines centered around that line.
6696 Print lines just before the lines last printed.
6699 @cindex @code{list}, how many lines to display
6700 By default, @value{GDBN} prints ten source lines with any of these forms of
6701 the @code{list} command. You can change this using @code{set listsize}:
6704 @kindex set listsize
6705 @item set listsize @var{count}
6706 Make the @code{list} command display @var{count} source lines (unless
6707 the @code{list} argument explicitly specifies some other number).
6709 @kindex show listsize
6711 Display the number of lines that @code{list} prints.
6714 Repeating a @code{list} command with @key{RET} discards the argument,
6715 so it is equivalent to typing just @code{list}. This is more useful
6716 than listing the same lines again. An exception is made for an
6717 argument of @samp{-}; that argument is preserved in repetition so that
6718 each repetition moves up in the source file.
6720 In general, the @code{list} command expects you to supply zero, one or two
6721 @dfn{linespecs}. Linespecs specify source lines; there are several ways
6722 of writing them (@pxref{Specify Location}), but the effect is always
6723 to specify some source line.
6725 Here is a complete description of the possible arguments for @code{list}:
6728 @item list @var{linespec}
6729 Print lines centered around the line specified by @var{linespec}.
6731 @item list @var{first},@var{last}
6732 Print lines from @var{first} to @var{last}. Both arguments are
6733 linespecs. When a @code{list} command has two linespecs, and the
6734 source file of the second linespec is omitted, this refers to
6735 the same source file as the first linespec.
6737 @item list ,@var{last}
6738 Print lines ending with @var{last}.
6740 @item list @var{first},
6741 Print lines starting with @var{first}.
6744 Print lines just after the lines last printed.
6747 Print lines just before the lines last printed.
6750 As described in the preceding table.
6753 @node Specify Location
6754 @section Specifying a Location
6755 @cindex specifying location
6758 Several @value{GDBN} commands accept arguments that specify a location
6759 of your program's code. Since @value{GDBN} is a source-level
6760 debugger, a location usually specifies some line in the source code;
6761 for that reason, locations are also known as @dfn{linespecs}.
6763 Here are all the different ways of specifying a code location that
6764 @value{GDBN} understands:
6768 Specifies the line number @var{linenum} of the current source file.
6771 @itemx +@var{offset}
6772 Specifies the line @var{offset} lines before or after the @dfn{current
6773 line}. For the @code{list} command, the current line is the last one
6774 printed; for the breakpoint commands, this is the line at which
6775 execution stopped in the currently selected @dfn{stack frame}
6776 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
6777 used as the second of the two linespecs in a @code{list} command,
6778 this specifies the line @var{offset} lines up or down from the first
6781 @item @var{filename}:@var{linenum}
6782 Specifies the line @var{linenum} in the source file @var{filename}.
6783 If @var{filename} is a relative file name, then it will match any
6784 source file name with the same trailing components. For example, if
6785 @var{filename} is @samp{gcc/expr.c}, then it will match source file
6786 name of @file{/build/trunk/gcc/expr.c}, but not
6787 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
6789 @item @var{function}
6790 Specifies the line that begins the body of the function @var{function}.
6791 For example, in C, this is the line with the open brace.
6793 @item @var{function}:@var{label}
6794 Specifies the line where @var{label} appears in @var{function}.
6796 @item @var{filename}:@var{function}
6797 Specifies the line that begins the body of the function @var{function}
6798 in the file @var{filename}. You only need the file name with a
6799 function name to avoid ambiguity when there are identically named
6800 functions in different source files.
6803 Specifies the line at which the label named @var{label} appears.
6804 @value{GDBN} searches for the label in the function corresponding to
6805 the currently selected stack frame. If there is no current selected
6806 stack frame (for instance, if the inferior is not running), then
6807 @value{GDBN} will not search for a label.
6809 @item *@var{address}
6810 Specifies the program address @var{address}. For line-oriented
6811 commands, such as @code{list} and @code{edit}, this specifies a source
6812 line that contains @var{address}. For @code{break} and other
6813 breakpoint oriented commands, this can be used to set breakpoints in
6814 parts of your program which do not have debugging information or
6817 Here @var{address} may be any expression valid in the current working
6818 language (@pxref{Languages, working language}) that specifies a code
6819 address. In addition, as a convenience, @value{GDBN} extends the
6820 semantics of expressions used in locations to cover the situations
6821 that frequently happen during debugging. Here are the various forms
6825 @item @var{expression}
6826 Any expression valid in the current working language.
6828 @item @var{funcaddr}
6829 An address of a function or procedure derived from its name. In C,
6830 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
6831 simply the function's name @var{function} (and actually a special case
6832 of a valid expression). In Pascal and Modula-2, this is
6833 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
6834 (although the Pascal form also works).
6836 This form specifies the address of the function's first instruction,
6837 before the stack frame and arguments have been set up.
6839 @item '@var{filename}'::@var{funcaddr}
6840 Like @var{funcaddr} above, but also specifies the name of the source
6841 file explicitly. This is useful if the name of the function does not
6842 specify the function unambiguously, e.g., if there are several
6843 functions with identical names in different source files.
6846 @cindex breakpoint at static probe point
6847 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
6848 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
6849 applications to embed static probes. @xref{Static Probe Points}, for more
6850 information on finding and using static probes. This form of linespec
6851 specifies the location of such a static probe.
6853 If @var{objfile} is given, only probes coming from that shared library
6854 or executable matching @var{objfile} as a regular expression are considered.
6855 If @var{provider} is given, then only probes from that provider are considered.
6856 If several probes match the spec, @value{GDBN} will insert a breakpoint at
6857 each one of those probes.
6863 @section Editing Source Files
6864 @cindex editing source files
6867 @kindex e @r{(@code{edit})}
6868 To edit the lines in a source file, use the @code{edit} command.
6869 The editing program of your choice
6870 is invoked with the current line set to
6871 the active line in the program.
6872 Alternatively, there are several ways to specify what part of the file you
6873 want to print if you want to see other parts of the program:
6876 @item edit @var{location}
6877 Edit the source file specified by @code{location}. Editing starts at
6878 that @var{location}, e.g., at the specified source line of the
6879 specified file. @xref{Specify Location}, for all the possible forms
6880 of the @var{location} argument; here are the forms of the @code{edit}
6881 command most commonly used:
6884 @item edit @var{number}
6885 Edit the current source file with @var{number} as the active line number.
6887 @item edit @var{function}
6888 Edit the file containing @var{function} at the beginning of its definition.
6893 @subsection Choosing your Editor
6894 You can customize @value{GDBN} to use any editor you want
6896 The only restriction is that your editor (say @code{ex}), recognizes the
6897 following command-line syntax:
6899 ex +@var{number} file
6901 The optional numeric value +@var{number} specifies the number of the line in
6902 the file where to start editing.}.
6903 By default, it is @file{@value{EDITOR}}, but you can change this
6904 by setting the environment variable @code{EDITOR} before using
6905 @value{GDBN}. For example, to configure @value{GDBN} to use the
6906 @code{vi} editor, you could use these commands with the @code{sh} shell:
6912 or in the @code{csh} shell,
6914 setenv EDITOR /usr/bin/vi
6919 @section Searching Source Files
6920 @cindex searching source files
6922 There are two commands for searching through the current source file for a
6927 @kindex forward-search
6928 @item forward-search @var{regexp}
6929 @itemx search @var{regexp}
6930 The command @samp{forward-search @var{regexp}} checks each line,
6931 starting with the one following the last line listed, for a match for
6932 @var{regexp}. It lists the line that is found. You can use the
6933 synonym @samp{search @var{regexp}} or abbreviate the command name as
6936 @kindex reverse-search
6937 @item reverse-search @var{regexp}
6938 The command @samp{reverse-search @var{regexp}} checks each line, starting
6939 with the one before the last line listed and going backward, for a match
6940 for @var{regexp}. It lists the line that is found. You can abbreviate
6941 this command as @code{rev}.
6945 @section Specifying Source Directories
6948 @cindex directories for source files
6949 Executable programs sometimes do not record the directories of the source
6950 files from which they were compiled, just the names. Even when they do,
6951 the directories could be moved between the compilation and your debugging
6952 session. @value{GDBN} has a list of directories to search for source files;
6953 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
6954 it tries all the directories in the list, in the order they are present
6955 in the list, until it finds a file with the desired name.
6957 For example, suppose an executable references the file
6958 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
6959 @file{/mnt/cross}. The file is first looked up literally; if this
6960 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
6961 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
6962 message is printed. @value{GDBN} does not look up the parts of the
6963 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
6964 Likewise, the subdirectories of the source path are not searched: if
6965 the source path is @file{/mnt/cross}, and the binary refers to
6966 @file{foo.c}, @value{GDBN} would not find it under
6967 @file{/mnt/cross/usr/src/foo-1.0/lib}.
6969 Plain file names, relative file names with leading directories, file
6970 names containing dots, etc.@: are all treated as described above; for
6971 instance, if the source path is @file{/mnt/cross}, and the source file
6972 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
6973 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
6974 that---@file{/mnt/cross/foo.c}.
6976 Note that the executable search path is @emph{not} used to locate the
6979 Whenever you reset or rearrange the source path, @value{GDBN} clears out
6980 any information it has cached about where source files are found and where
6981 each line is in the file.
6985 When you start @value{GDBN}, its source path includes only @samp{cdir}
6986 and @samp{cwd}, in that order.
6987 To add other directories, use the @code{directory} command.
6989 The search path is used to find both program source files and @value{GDBN}
6990 script files (read using the @samp{-command} option and @samp{source} command).
6992 In addition to the source path, @value{GDBN} provides a set of commands
6993 that manage a list of source path substitution rules. A @dfn{substitution
6994 rule} specifies how to rewrite source directories stored in the program's
6995 debug information in case the sources were moved to a different
6996 directory between compilation and debugging. A rule is made of
6997 two strings, the first specifying what needs to be rewritten in
6998 the path, and the second specifying how it should be rewritten.
6999 In @ref{set substitute-path}, we name these two parts @var{from} and
7000 @var{to} respectively. @value{GDBN} does a simple string replacement
7001 of @var{from} with @var{to} at the start of the directory part of the
7002 source file name, and uses that result instead of the original file
7003 name to look up the sources.
7005 Using the previous example, suppose the @file{foo-1.0} tree has been
7006 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
7007 @value{GDBN} to replace @file{/usr/src} in all source path names with
7008 @file{/mnt/cross}. The first lookup will then be
7009 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
7010 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
7011 substitution rule, use the @code{set substitute-path} command
7012 (@pxref{set substitute-path}).
7014 To avoid unexpected substitution results, a rule is applied only if the
7015 @var{from} part of the directory name ends at a directory separator.
7016 For instance, a rule substituting @file{/usr/source} into
7017 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
7018 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
7019 is applied only at the beginning of the directory name, this rule will
7020 not be applied to @file{/root/usr/source/baz.c} either.
7022 In many cases, you can achieve the same result using the @code{directory}
7023 command. However, @code{set substitute-path} can be more efficient in
7024 the case where the sources are organized in a complex tree with multiple
7025 subdirectories. With the @code{directory} command, you need to add each
7026 subdirectory of your project. If you moved the entire tree while
7027 preserving its internal organization, then @code{set substitute-path}
7028 allows you to direct the debugger to all the sources with one single
7031 @code{set substitute-path} is also more than just a shortcut command.
7032 The source path is only used if the file at the original location no
7033 longer exists. On the other hand, @code{set substitute-path} modifies
7034 the debugger behavior to look at the rewritten location instead. So, if
7035 for any reason a source file that is not relevant to your executable is
7036 located at the original location, a substitution rule is the only
7037 method available to point @value{GDBN} at the new location.
7039 @cindex @samp{--with-relocated-sources}
7040 @cindex default source path substitution
7041 You can configure a default source path substitution rule by
7042 configuring @value{GDBN} with the
7043 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
7044 should be the name of a directory under @value{GDBN}'s configured
7045 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
7046 directory names in debug information under @var{dir} will be adjusted
7047 automatically if the installed @value{GDBN} is moved to a new
7048 location. This is useful if @value{GDBN}, libraries or executables
7049 with debug information and corresponding source code are being moved
7053 @item directory @var{dirname} @dots{}
7054 @item dir @var{dirname} @dots{}
7055 Add directory @var{dirname} to the front of the source path. Several
7056 directory names may be given to this command, separated by @samp{:}
7057 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
7058 part of absolute file names) or
7059 whitespace. You may specify a directory that is already in the source
7060 path; this moves it forward, so @value{GDBN} searches it sooner.
7064 @vindex $cdir@r{, convenience variable}
7065 @vindex $cwd@r{, convenience variable}
7066 @cindex compilation directory
7067 @cindex current directory
7068 @cindex working directory
7069 @cindex directory, current
7070 @cindex directory, compilation
7071 You can use the string @samp{$cdir} to refer to the compilation
7072 directory (if one is recorded), and @samp{$cwd} to refer to the current
7073 working directory. @samp{$cwd} is not the same as @samp{.}---the former
7074 tracks the current working directory as it changes during your @value{GDBN}
7075 session, while the latter is immediately expanded to the current
7076 directory at the time you add an entry to the source path.
7079 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
7081 @c RET-repeat for @code{directory} is explicitly disabled, but since
7082 @c repeating it would be a no-op we do not say that. (thanks to RMS)
7084 @item set directories @var{path-list}
7085 @kindex set directories
7086 Set the source path to @var{path-list}.
7087 @samp{$cdir:$cwd} are added if missing.
7089 @item show directories
7090 @kindex show directories
7091 Print the source path: show which directories it contains.
7093 @anchor{set substitute-path}
7094 @item set substitute-path @var{from} @var{to}
7095 @kindex set substitute-path
7096 Define a source path substitution rule, and add it at the end of the
7097 current list of existing substitution rules. If a rule with the same
7098 @var{from} was already defined, then the old rule is also deleted.
7100 For example, if the file @file{/foo/bar/baz.c} was moved to
7101 @file{/mnt/cross/baz.c}, then the command
7104 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
7108 will tell @value{GDBN} to replace @samp{/usr/src} with
7109 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
7110 @file{baz.c} even though it was moved.
7112 In the case when more than one substitution rule have been defined,
7113 the rules are evaluated one by one in the order where they have been
7114 defined. The first one matching, if any, is selected to perform
7117 For instance, if we had entered the following commands:
7120 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
7121 (@value{GDBP}) set substitute-path /usr/src /mnt/src
7125 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
7126 @file{/mnt/include/defs.h} by using the first rule. However, it would
7127 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
7128 @file{/mnt/src/lib/foo.c}.
7131 @item unset substitute-path [path]
7132 @kindex unset substitute-path
7133 If a path is specified, search the current list of substitution rules
7134 for a rule that would rewrite that path. Delete that rule if found.
7135 A warning is emitted by the debugger if no rule could be found.
7137 If no path is specified, then all substitution rules are deleted.
7139 @item show substitute-path [path]
7140 @kindex show substitute-path
7141 If a path is specified, then print the source path substitution rule
7142 which would rewrite that path, if any.
7144 If no path is specified, then print all existing source path substitution
7149 If your source path is cluttered with directories that are no longer of
7150 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
7151 versions of source. You can correct the situation as follows:
7155 Use @code{directory} with no argument to reset the source path to its default value.
7158 Use @code{directory} with suitable arguments to reinstall the
7159 directories you want in the source path. You can add all the
7160 directories in one command.
7164 @section Source and Machine Code
7165 @cindex source line and its code address
7167 You can use the command @code{info line} to map source lines to program
7168 addresses (and vice versa), and the command @code{disassemble} to display
7169 a range of addresses as machine instructions. You can use the command
7170 @code{set disassemble-next-line} to set whether to disassemble next
7171 source line when execution stops. When run under @sc{gnu} Emacs
7172 mode, the @code{info line} command causes the arrow to point to the
7173 line specified. Also, @code{info line} prints addresses in symbolic form as
7178 @item info line @var{linespec}
7179 Print the starting and ending addresses of the compiled code for
7180 source line @var{linespec}. You can specify source lines in any of
7181 the ways documented in @ref{Specify Location}.
7184 For example, we can use @code{info line} to discover the location of
7185 the object code for the first line of function
7186 @code{m4_changequote}:
7188 @c FIXME: I think this example should also show the addresses in
7189 @c symbolic form, as they usually would be displayed.
7191 (@value{GDBP}) info line m4_changequote
7192 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
7196 @cindex code address and its source line
7197 We can also inquire (using @code{*@var{addr}} as the form for
7198 @var{linespec}) what source line covers a particular address:
7200 (@value{GDBP}) info line *0x63ff
7201 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
7204 @cindex @code{$_} and @code{info line}
7205 @cindex @code{x} command, default address
7206 @kindex x@r{(examine), and} info line
7207 After @code{info line}, the default address for the @code{x} command
7208 is changed to the starting address of the line, so that @samp{x/i} is
7209 sufficient to begin examining the machine code (@pxref{Memory,
7210 ,Examining Memory}). Also, this address is saved as the value of the
7211 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
7216 @cindex assembly instructions
7217 @cindex instructions, assembly
7218 @cindex machine instructions
7219 @cindex listing machine instructions
7221 @itemx disassemble /m
7222 @itemx disassemble /r
7223 This specialized command dumps a range of memory as machine
7224 instructions. It can also print mixed source+disassembly by specifying
7225 the @code{/m} modifier and print the raw instructions in hex as well as
7226 in symbolic form by specifying the @code{/r}.
7227 The default memory range is the function surrounding the
7228 program counter of the selected frame. A single argument to this
7229 command is a program counter value; @value{GDBN} dumps the function
7230 surrounding this value. When two arguments are given, they should
7231 be separated by a comma, possibly surrounded by whitespace. The
7232 arguments specify a range of addresses to dump, in one of two forms:
7235 @item @var{start},@var{end}
7236 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
7237 @item @var{start},+@var{length}
7238 the addresses from @var{start} (inclusive) to
7239 @code{@var{start}+@var{length}} (exclusive).
7243 When 2 arguments are specified, the name of the function is also
7244 printed (since there could be several functions in the given range).
7246 The argument(s) can be any expression yielding a numeric value, such as
7247 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
7249 If the range of memory being disassembled contains current program counter,
7250 the instruction at that location is shown with a @code{=>} marker.
7253 The following example shows the disassembly of a range of addresses of
7254 HP PA-RISC 2.0 code:
7257 (@value{GDBP}) disas 0x32c4, 0x32e4
7258 Dump of assembler code from 0x32c4 to 0x32e4:
7259 0x32c4 <main+204>: addil 0,dp
7260 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
7261 0x32cc <main+212>: ldil 0x3000,r31
7262 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
7263 0x32d4 <main+220>: ldo 0(r31),rp
7264 0x32d8 <main+224>: addil -0x800,dp
7265 0x32dc <main+228>: ldo 0x588(r1),r26
7266 0x32e0 <main+232>: ldil 0x3000,r31
7267 End of assembler dump.
7270 Here is an example showing mixed source+assembly for Intel x86, when the
7271 program is stopped just after function prologue:
7274 (@value{GDBP}) disas /m main
7275 Dump of assembler code for function main:
7277 0x08048330 <+0>: push %ebp
7278 0x08048331 <+1>: mov %esp,%ebp
7279 0x08048333 <+3>: sub $0x8,%esp
7280 0x08048336 <+6>: and $0xfffffff0,%esp
7281 0x08048339 <+9>: sub $0x10,%esp
7283 6 printf ("Hello.\n");
7284 => 0x0804833c <+12>: movl $0x8048440,(%esp)
7285 0x08048343 <+19>: call 0x8048284 <puts@@plt>
7289 0x08048348 <+24>: mov $0x0,%eax
7290 0x0804834d <+29>: leave
7291 0x0804834e <+30>: ret
7293 End of assembler dump.
7296 Here is another example showing raw instructions in hex for AMD x86-64,
7299 (gdb) disas /r 0x400281,+10
7300 Dump of assembler code from 0x400281 to 0x40028b:
7301 0x0000000000400281: 38 36 cmp %dh,(%rsi)
7302 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
7303 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
7304 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
7305 End of assembler dump.
7308 Some architectures have more than one commonly-used set of instruction
7309 mnemonics or other syntax.
7311 For programs that were dynamically linked and use shared libraries,
7312 instructions that call functions or branch to locations in the shared
7313 libraries might show a seemingly bogus location---it's actually a
7314 location of the relocation table. On some architectures, @value{GDBN}
7315 might be able to resolve these to actual function names.
7318 @kindex set disassembly-flavor
7319 @cindex Intel disassembly flavor
7320 @cindex AT&T disassembly flavor
7321 @item set disassembly-flavor @var{instruction-set}
7322 Select the instruction set to use when disassembling the
7323 program via the @code{disassemble} or @code{x/i} commands.
7325 Currently this command is only defined for the Intel x86 family. You
7326 can set @var{instruction-set} to either @code{intel} or @code{att}.
7327 The default is @code{att}, the AT&T flavor used by default by Unix
7328 assemblers for x86-based targets.
7330 @kindex show disassembly-flavor
7331 @item show disassembly-flavor
7332 Show the current setting of the disassembly flavor.
7336 @kindex set disassemble-next-line
7337 @kindex show disassemble-next-line
7338 @item set disassemble-next-line
7339 @itemx show disassemble-next-line
7340 Control whether or not @value{GDBN} will disassemble the next source
7341 line or instruction when execution stops. If ON, @value{GDBN} will
7342 display disassembly of the next source line when execution of the
7343 program being debugged stops. This is @emph{in addition} to
7344 displaying the source line itself, which @value{GDBN} always does if
7345 possible. If the next source line cannot be displayed for some reason
7346 (e.g., if @value{GDBN} cannot find the source file, or there's no line
7347 info in the debug info), @value{GDBN} will display disassembly of the
7348 next @emph{instruction} instead of showing the next source line. If
7349 AUTO, @value{GDBN} will display disassembly of next instruction only
7350 if the source line cannot be displayed. This setting causes
7351 @value{GDBN} to display some feedback when you step through a function
7352 with no line info or whose source file is unavailable. The default is
7353 OFF, which means never display the disassembly of the next line or
7359 @chapter Examining Data
7361 @cindex printing data
7362 @cindex examining data
7365 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
7366 @c document because it is nonstandard... Under Epoch it displays in a
7367 @c different window or something like that.
7368 The usual way to examine data in your program is with the @code{print}
7369 command (abbreviated @code{p}), or its synonym @code{inspect}. It
7370 evaluates and prints the value of an expression of the language your
7371 program is written in (@pxref{Languages, ,Using @value{GDBN} with
7372 Different Languages}). It may also print the expression using a
7373 Python-based pretty-printer (@pxref{Pretty Printing}).
7376 @item print @var{expr}
7377 @itemx print /@var{f} @var{expr}
7378 @var{expr} is an expression (in the source language). By default the
7379 value of @var{expr} is printed in a format appropriate to its data type;
7380 you can choose a different format by specifying @samp{/@var{f}}, where
7381 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
7385 @itemx print /@var{f}
7386 @cindex reprint the last value
7387 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
7388 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
7389 conveniently inspect the same value in an alternative format.
7392 A more low-level way of examining data is with the @code{x} command.
7393 It examines data in memory at a specified address and prints it in a
7394 specified format. @xref{Memory, ,Examining Memory}.
7396 If you are interested in information about types, or about how the
7397 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
7398 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
7401 @cindex exploring hierarchical data structures
7403 Another way of examining values of expressions and type information is
7404 through the Python extension command @code{explore} (available only if
7405 the @value{GDBN} build is configured with @code{--with-python}). It
7406 offers an interactive way to start at the highest level (or, the most
7407 abstract level) of the data type of an expression (or, the data type
7408 itself) and explore all the way down to leaf scalar values/fields
7409 embedded in the higher level data types.
7412 @item explore @var{arg}
7413 @var{arg} is either an expression (in the source language), or a type
7414 visible in the current context of the program being debugged.
7417 The working of the @code{explore} command can be illustrated with an
7418 example. If a data type @code{struct ComplexStruct} is defined in your
7428 struct ComplexStruct
7430 struct SimpleStruct *ss_p;
7436 followed by variable declarations as
7439 struct SimpleStruct ss = @{ 10, 1.11 @};
7440 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
7444 then, the value of the variable @code{cs} can be explored using the
7445 @code{explore} command as follows.
7449 The value of `cs' is a struct/class of type `struct ComplexStruct' with
7450 the following fields:
7452 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
7453 arr = <Enter 1 to explore this field of type `int [10]'>
7455 Enter the field number of choice:
7459 Since the fields of @code{cs} are not scalar values, you are being
7460 prompted to chose the field you want to explore. Let's say you choose
7461 the field @code{ss_p} by entering @code{0}. Then, since this field is a
7462 pointer, you will be asked if it is pointing to a single value. From
7463 the declaration of @code{cs} above, it is indeed pointing to a single
7464 value, hence you enter @code{y}. If you enter @code{n}, then you will
7465 be asked if it were pointing to an array of values, in which case this
7466 field will be explored as if it were an array.
7469 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
7470 Continue exploring it as a pointer to a single value [y/n]: y
7471 The value of `*(cs.ss_p)' is a struct/class of type `struct
7472 SimpleStruct' with the following fields:
7474 i = 10 .. (Value of type `int')
7475 d = 1.1100000000000001 .. (Value of type `double')
7477 Press enter to return to parent value:
7481 If the field @code{arr} of @code{cs} was chosen for exploration by
7482 entering @code{1} earlier, then since it is as array, you will be
7483 prompted to enter the index of the element in the array that you want
7487 `cs.arr' is an array of `int'.
7488 Enter the index of the element you want to explore in `cs.arr': 5
7490 `(cs.arr)[5]' is a scalar value of type `int'.
7494 Press enter to return to parent value:
7497 In general, at any stage of exploration, you can go deeper towards the
7498 leaf values by responding to the prompts appropriately, or hit the
7499 return key to return to the enclosing data structure (the @i{higher}
7500 level data structure).
7502 Similar to exploring values, you can use the @code{explore} command to
7503 explore types. Instead of specifying a value (which is typically a
7504 variable name or an expression valid in the current context of the
7505 program being debugged), you specify a type name. If you consider the
7506 same example as above, your can explore the type
7507 @code{struct ComplexStruct} by passing the argument
7508 @code{struct ComplexStruct} to the @code{explore} command.
7511 (gdb) explore struct ComplexStruct
7515 By responding to the prompts appropriately in the subsequent interactive
7516 session, you can explore the type @code{struct ComplexStruct} in a
7517 manner similar to how the value @code{cs} was explored in the above
7520 The @code{explore} command also has two sub-commands,
7521 @code{explore value} and @code{explore type}. The former sub-command is
7522 a way to explicitly specify that value exploration of the argument is
7523 being invoked, while the latter is a way to explicitly specify that type
7524 exploration of the argument is being invoked.
7527 @item explore value @var{expr}
7528 @cindex explore value
7529 This sub-command of @code{explore} explores the value of the
7530 expression @var{expr} (if @var{expr} is an expression valid in the
7531 current context of the program being debugged). The behavior of this
7532 command is identical to that of the behavior of the @code{explore}
7533 command being passed the argument @var{expr}.
7535 @item explore type @var{arg}
7536 @cindex explore type
7537 This sub-command of @code{explore} explores the type of @var{arg} (if
7538 @var{arg} is a type visible in the current context of program being
7539 debugged), or the type of the value/expression @var{arg} (if @var{arg}
7540 is an expression valid in the current context of the program being
7541 debugged). If @var{arg} is a type, then the behavior of this command is
7542 identical to that of the @code{explore} command being passed the
7543 argument @var{arg}. If @var{arg} is an expression, then the behavior of
7544 this command will be identical to that of the @code{explore} command
7545 being passed the type of @var{arg} as the argument.
7549 * Expressions:: Expressions
7550 * Ambiguous Expressions:: Ambiguous Expressions
7551 * Variables:: Program variables
7552 * Arrays:: Artificial arrays
7553 * Output Formats:: Output formats
7554 * Memory:: Examining memory
7555 * Auto Display:: Automatic display
7556 * Print Settings:: Print settings
7557 * Pretty Printing:: Python pretty printing
7558 * Value History:: Value history
7559 * Convenience Vars:: Convenience variables
7560 * Registers:: Registers
7561 * Floating Point Hardware:: Floating point hardware
7562 * Vector Unit:: Vector Unit
7563 * OS Information:: Auxiliary data provided by operating system
7564 * Memory Region Attributes:: Memory region attributes
7565 * Dump/Restore Files:: Copy between memory and a file
7566 * Core File Generation:: Cause a program dump its core
7567 * Character Sets:: Debugging programs that use a different
7568 character set than GDB does
7569 * Caching Remote Data:: Data caching for remote targets
7570 * Searching Memory:: Searching memory for a sequence of bytes
7574 @section Expressions
7577 @code{print} and many other @value{GDBN} commands accept an expression and
7578 compute its value. Any kind of constant, variable or operator defined
7579 by the programming language you are using is valid in an expression in
7580 @value{GDBN}. This includes conditional expressions, function calls,
7581 casts, and string constants. It also includes preprocessor macros, if
7582 you compiled your program to include this information; see
7585 @cindex arrays in expressions
7586 @value{GDBN} supports array constants in expressions input by
7587 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
7588 you can use the command @code{print @{1, 2, 3@}} to create an array
7589 of three integers. If you pass an array to a function or assign it
7590 to a program variable, @value{GDBN} copies the array to memory that
7591 is @code{malloc}ed in the target program.
7593 Because C is so widespread, most of the expressions shown in examples in
7594 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
7595 Languages}, for information on how to use expressions in other
7598 In this section, we discuss operators that you can use in @value{GDBN}
7599 expressions regardless of your programming language.
7601 @cindex casts, in expressions
7602 Casts are supported in all languages, not just in C, because it is so
7603 useful to cast a number into a pointer in order to examine a structure
7604 at that address in memory.
7605 @c FIXME: casts supported---Mod2 true?
7607 @value{GDBN} supports these operators, in addition to those common
7608 to programming languages:
7612 @samp{@@} is a binary operator for treating parts of memory as arrays.
7613 @xref{Arrays, ,Artificial Arrays}, for more information.
7616 @samp{::} allows you to specify a variable in terms of the file or
7617 function where it is defined. @xref{Variables, ,Program Variables}.
7619 @cindex @{@var{type}@}
7620 @cindex type casting memory
7621 @cindex memory, viewing as typed object
7622 @cindex casts, to view memory
7623 @item @{@var{type}@} @var{addr}
7624 Refers to an object of type @var{type} stored at address @var{addr} in
7625 memory. @var{addr} may be any expression whose value is an integer or
7626 pointer (but parentheses are required around binary operators, just as in
7627 a cast). This construct is allowed regardless of what kind of data is
7628 normally supposed to reside at @var{addr}.
7631 @node Ambiguous Expressions
7632 @section Ambiguous Expressions
7633 @cindex ambiguous expressions
7635 Expressions can sometimes contain some ambiguous elements. For instance,
7636 some programming languages (notably Ada, C@t{++} and Objective-C) permit
7637 a single function name to be defined several times, for application in
7638 different contexts. This is called @dfn{overloading}. Another example
7639 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
7640 templates and is typically instantiated several times, resulting in
7641 the same function name being defined in different contexts.
7643 In some cases and depending on the language, it is possible to adjust
7644 the expression to remove the ambiguity. For instance in C@t{++}, you
7645 can specify the signature of the function you want to break on, as in
7646 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
7647 qualified name of your function often makes the expression unambiguous
7650 When an ambiguity that needs to be resolved is detected, the debugger
7651 has the capability to display a menu of numbered choices for each
7652 possibility, and then waits for the selection with the prompt @samp{>}.
7653 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
7654 aborts the current command. If the command in which the expression was
7655 used allows more than one choice to be selected, the next option in the
7656 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
7659 For example, the following session excerpt shows an attempt to set a
7660 breakpoint at the overloaded symbol @code{String::after}.
7661 We choose three particular definitions of that function name:
7663 @c FIXME! This is likely to change to show arg type lists, at least
7666 (@value{GDBP}) b String::after
7669 [2] file:String.cc; line number:867
7670 [3] file:String.cc; line number:860
7671 [4] file:String.cc; line number:875
7672 [5] file:String.cc; line number:853
7673 [6] file:String.cc; line number:846
7674 [7] file:String.cc; line number:735
7676 Breakpoint 1 at 0xb26c: file String.cc, line 867.
7677 Breakpoint 2 at 0xb344: file String.cc, line 875.
7678 Breakpoint 3 at 0xafcc: file String.cc, line 846.
7679 Multiple breakpoints were set.
7680 Use the "delete" command to delete unwanted
7687 @kindex set multiple-symbols
7688 @item set multiple-symbols @var{mode}
7689 @cindex multiple-symbols menu
7691 This option allows you to adjust the debugger behavior when an expression
7694 By default, @var{mode} is set to @code{all}. If the command with which
7695 the expression is used allows more than one choice, then @value{GDBN}
7696 automatically selects all possible choices. For instance, inserting
7697 a breakpoint on a function using an ambiguous name results in a breakpoint
7698 inserted on each possible match. However, if a unique choice must be made,
7699 then @value{GDBN} uses the menu to help you disambiguate the expression.
7700 For instance, printing the address of an overloaded function will result
7701 in the use of the menu.
7703 When @var{mode} is set to @code{ask}, the debugger always uses the menu
7704 when an ambiguity is detected.
7706 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
7707 an error due to the ambiguity and the command is aborted.
7709 @kindex show multiple-symbols
7710 @item show multiple-symbols
7711 Show the current value of the @code{multiple-symbols} setting.
7715 @section Program Variables
7717 The most common kind of expression to use is the name of a variable
7720 Variables in expressions are understood in the selected stack frame
7721 (@pxref{Selection, ,Selecting a Frame}); they must be either:
7725 global (or file-static)
7732 visible according to the scope rules of the
7733 programming language from the point of execution in that frame
7736 @noindent This means that in the function
7751 you can examine and use the variable @code{a} whenever your program is
7752 executing within the function @code{foo}, but you can only use or
7753 examine the variable @code{b} while your program is executing inside
7754 the block where @code{b} is declared.
7756 @cindex variable name conflict
7757 There is an exception: you can refer to a variable or function whose
7758 scope is a single source file even if the current execution point is not
7759 in this file. But it is possible to have more than one such variable or
7760 function with the same name (in different source files). If that
7761 happens, referring to that name has unpredictable effects. If you wish,
7762 you can specify a static variable in a particular function or file by
7763 using the colon-colon (@code{::}) notation:
7765 @cindex colon-colon, context for variables/functions
7767 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
7768 @cindex @code{::}, context for variables/functions
7771 @var{file}::@var{variable}
7772 @var{function}::@var{variable}
7776 Here @var{file} or @var{function} is the name of the context for the
7777 static @var{variable}. In the case of file names, you can use quotes to
7778 make sure @value{GDBN} parses the file name as a single word---for example,
7779 to print a global value of @code{x} defined in @file{f2.c}:
7782 (@value{GDBP}) p 'f2.c'::x
7785 The @code{::} notation is normally used for referring to
7786 static variables, since you typically disambiguate uses of local variables
7787 in functions by selecting the appropriate frame and using the
7788 simple name of the variable. However, you may also use this notation
7789 to refer to local variables in frames enclosing the selected frame:
7798 process (a); /* Stop here */
7809 For example, if there is a breakpoint at the commented line,
7810 here is what you might see
7811 when the program stops after executing the call @code{bar(0)}:
7816 (@value{GDBP}) p bar::a
7819 #2 0x080483d0 in foo (a=5) at foobar.c:12
7822 (@value{GDBP}) p bar::a
7826 @cindex C@t{++} scope resolution
7827 These uses of @samp{::} are very rarely in conflict with the very similar
7828 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
7829 scope resolution operator in @value{GDBN} expressions.
7830 @c FIXME: Um, so what happens in one of those rare cases where it's in
7833 @cindex wrong values
7834 @cindex variable values, wrong
7835 @cindex function entry/exit, wrong values of variables
7836 @cindex optimized code, wrong values of variables
7838 @emph{Warning:} Occasionally, a local variable may appear to have the
7839 wrong value at certain points in a function---just after entry to a new
7840 scope, and just before exit.
7842 You may see this problem when you are stepping by machine instructions.
7843 This is because, on most machines, it takes more than one instruction to
7844 set up a stack frame (including local variable definitions); if you are
7845 stepping by machine instructions, variables may appear to have the wrong
7846 values until the stack frame is completely built. On exit, it usually
7847 also takes more than one machine instruction to destroy a stack frame;
7848 after you begin stepping through that group of instructions, local
7849 variable definitions may be gone.
7851 This may also happen when the compiler does significant optimizations.
7852 To be sure of always seeing accurate values, turn off all optimization
7855 @cindex ``No symbol "foo" in current context''
7856 Another possible effect of compiler optimizations is to optimize
7857 unused variables out of existence, or assign variables to registers (as
7858 opposed to memory addresses). Depending on the support for such cases
7859 offered by the debug info format used by the compiler, @value{GDBN}
7860 might not be able to display values for such local variables. If that
7861 happens, @value{GDBN} will print a message like this:
7864 No symbol "foo" in current context.
7867 To solve such problems, either recompile without optimizations, or use a
7868 different debug info format, if the compiler supports several such
7869 formats. @xref{Compilation}, for more information on choosing compiler
7870 options. @xref{C, ,C and C@t{++}}, for more information about debug
7871 info formats that are best suited to C@t{++} programs.
7873 If you ask to print an object whose contents are unknown to
7874 @value{GDBN}, e.g., because its data type is not completely specified
7875 by the debug information, @value{GDBN} will say @samp{<incomplete
7876 type>}. @xref{Symbols, incomplete type}, for more about this.
7878 If you append @kbd{@@entry} string to a function parameter name you get its
7879 value at the time the function got called. If the value is not available an
7880 error message is printed. Entry values are available only with some compilers.
7881 Entry values are normally also printed at the function parameter list according
7882 to @ref{set print entry-values}.
7885 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
7891 (gdb) print i@@entry
7895 Strings are identified as arrays of @code{char} values without specified
7896 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
7897 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
7898 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
7899 defines literal string type @code{"char"} as @code{char} without a sign.
7904 signed char var1[] = "A";
7907 You get during debugging
7912 $2 = @{65 'A', 0 '\0'@}
7916 @section Artificial Arrays
7918 @cindex artificial array
7920 @kindex @@@r{, referencing memory as an array}
7921 It is often useful to print out several successive objects of the
7922 same type in memory; a section of an array, or an array of
7923 dynamically determined size for which only a pointer exists in the
7926 You can do this by referring to a contiguous span of memory as an
7927 @dfn{artificial array}, using the binary operator @samp{@@}. The left
7928 operand of @samp{@@} should be the first element of the desired array
7929 and be an individual object. The right operand should be the desired length
7930 of the array. The result is an array value whose elements are all of
7931 the type of the left argument. The first element is actually the left
7932 argument; the second element comes from bytes of memory immediately
7933 following those that hold the first element, and so on. Here is an
7934 example. If a program says
7937 int *array = (int *) malloc (len * sizeof (int));
7941 you can print the contents of @code{array} with
7947 The left operand of @samp{@@} must reside in memory. Array values made
7948 with @samp{@@} in this way behave just like other arrays in terms of
7949 subscripting, and are coerced to pointers when used in expressions.
7950 Artificial arrays most often appear in expressions via the value history
7951 (@pxref{Value History, ,Value History}), after printing one out.
7953 Another way to create an artificial array is to use a cast.
7954 This re-interprets a value as if it were an array.
7955 The value need not be in memory:
7957 (@value{GDBP}) p/x (short[2])0x12345678
7958 $1 = @{0x1234, 0x5678@}
7961 As a convenience, if you leave the array length out (as in
7962 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
7963 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
7965 (@value{GDBP}) p/x (short[])0x12345678
7966 $2 = @{0x1234, 0x5678@}
7969 Sometimes the artificial array mechanism is not quite enough; in
7970 moderately complex data structures, the elements of interest may not
7971 actually be adjacent---for example, if you are interested in the values
7972 of pointers in an array. One useful work-around in this situation is
7973 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
7974 Variables}) as a counter in an expression that prints the first
7975 interesting value, and then repeat that expression via @key{RET}. For
7976 instance, suppose you have an array @code{dtab} of pointers to
7977 structures, and you are interested in the values of a field @code{fv}
7978 in each structure. Here is an example of what you might type:
7988 @node Output Formats
7989 @section Output Formats
7991 @cindex formatted output
7992 @cindex output formats
7993 By default, @value{GDBN} prints a value according to its data type. Sometimes
7994 this is not what you want. For example, you might want to print a number
7995 in hex, or a pointer in decimal. Or you might want to view data in memory
7996 at a certain address as a character string or as an instruction. To do
7997 these things, specify an @dfn{output format} when you print a value.
7999 The simplest use of output formats is to say how to print a value
8000 already computed. This is done by starting the arguments of the
8001 @code{print} command with a slash and a format letter. The format
8002 letters supported are:
8006 Regard the bits of the value as an integer, and print the integer in
8010 Print as integer in signed decimal.
8013 Print as integer in unsigned decimal.
8016 Print as integer in octal.
8019 Print as integer in binary. The letter @samp{t} stands for ``two''.
8020 @footnote{@samp{b} cannot be used because these format letters are also
8021 used with the @code{x} command, where @samp{b} stands for ``byte'';
8022 see @ref{Memory,,Examining Memory}.}
8025 @cindex unknown address, locating
8026 @cindex locate address
8027 Print as an address, both absolute in hexadecimal and as an offset from
8028 the nearest preceding symbol. You can use this format used to discover
8029 where (in what function) an unknown address is located:
8032 (@value{GDBP}) p/a 0x54320
8033 $3 = 0x54320 <_initialize_vx+396>
8037 The command @code{info symbol 0x54320} yields similar results.
8038 @xref{Symbols, info symbol}.
8041 Regard as an integer and print it as a character constant. This
8042 prints both the numerical value and its character representation. The
8043 character representation is replaced with the octal escape @samp{\nnn}
8044 for characters outside the 7-bit @sc{ascii} range.
8046 Without this format, @value{GDBN} displays @code{char},
8047 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
8048 constants. Single-byte members of vectors are displayed as integer
8052 Regard the bits of the value as a floating point number and print
8053 using typical floating point syntax.
8056 @cindex printing strings
8057 @cindex printing byte arrays
8058 Regard as a string, if possible. With this format, pointers to single-byte
8059 data are displayed as null-terminated strings and arrays of single-byte data
8060 are displayed as fixed-length strings. Other values are displayed in their
8063 Without this format, @value{GDBN} displays pointers to and arrays of
8064 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
8065 strings. Single-byte members of a vector are displayed as an integer
8069 @cindex raw printing
8070 Print using the @samp{raw} formatting. By default, @value{GDBN} will
8071 use a Python-based pretty-printer, if one is available (@pxref{Pretty
8072 Printing}). This typically results in a higher-level display of the
8073 value's contents. The @samp{r} format bypasses any Python
8074 pretty-printer which might exist.
8077 For example, to print the program counter in hex (@pxref{Registers}), type
8084 Note that no space is required before the slash; this is because command
8085 names in @value{GDBN} cannot contain a slash.
8087 To reprint the last value in the value history with a different format,
8088 you can use the @code{print} command with just a format and no
8089 expression. For example, @samp{p/x} reprints the last value in hex.
8092 @section Examining Memory
8094 You can use the command @code{x} (for ``examine'') to examine memory in
8095 any of several formats, independently of your program's data types.
8097 @cindex examining memory
8099 @kindex x @r{(examine memory)}
8100 @item x/@var{nfu} @var{addr}
8103 Use the @code{x} command to examine memory.
8106 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
8107 much memory to display and how to format it; @var{addr} is an
8108 expression giving the address where you want to start displaying memory.
8109 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
8110 Several commands set convenient defaults for @var{addr}.
8113 @item @var{n}, the repeat count
8114 The repeat count is a decimal integer; the default is 1. It specifies
8115 how much memory (counting by units @var{u}) to display.
8116 @c This really is **decimal**; unaffected by 'set radix' as of GDB
8119 @item @var{f}, the display format
8120 The display format is one of the formats used by @code{print}
8121 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
8122 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
8123 The default is @samp{x} (hexadecimal) initially. The default changes
8124 each time you use either @code{x} or @code{print}.
8126 @item @var{u}, the unit size
8127 The unit size is any of
8133 Halfwords (two bytes).
8135 Words (four bytes). This is the initial default.
8137 Giant words (eight bytes).
8140 Each time you specify a unit size with @code{x}, that size becomes the
8141 default unit the next time you use @code{x}. For the @samp{i} format,
8142 the unit size is ignored and is normally not written. For the @samp{s} format,
8143 the unit size defaults to @samp{b}, unless it is explicitly given.
8144 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
8145 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
8146 Note that the results depend on the programming language of the
8147 current compilation unit. If the language is C, the @samp{s}
8148 modifier will use the UTF-16 encoding while @samp{w} will use
8149 UTF-32. The encoding is set by the programming language and cannot
8152 @item @var{addr}, starting display address
8153 @var{addr} is the address where you want @value{GDBN} to begin displaying
8154 memory. The expression need not have a pointer value (though it may);
8155 it is always interpreted as an integer address of a byte of memory.
8156 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
8157 @var{addr} is usually just after the last address examined---but several
8158 other commands also set the default address: @code{info breakpoints} (to
8159 the address of the last breakpoint listed), @code{info line} (to the
8160 starting address of a line), and @code{print} (if you use it to display
8161 a value from memory).
8164 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
8165 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
8166 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
8167 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
8168 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
8170 Since the letters indicating unit sizes are all distinct from the
8171 letters specifying output formats, you do not have to remember whether
8172 unit size or format comes first; either order works. The output
8173 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
8174 (However, the count @var{n} must come first; @samp{wx4} does not work.)
8176 Even though the unit size @var{u} is ignored for the formats @samp{s}
8177 and @samp{i}, you might still want to use a count @var{n}; for example,
8178 @samp{3i} specifies that you want to see three machine instructions,
8179 including any operands. For convenience, especially when used with
8180 the @code{display} command, the @samp{i} format also prints branch delay
8181 slot instructions, if any, beyond the count specified, which immediately
8182 follow the last instruction that is within the count. The command
8183 @code{disassemble} gives an alternative way of inspecting machine
8184 instructions; see @ref{Machine Code,,Source and Machine Code}.
8186 All the defaults for the arguments to @code{x} are designed to make it
8187 easy to continue scanning memory with minimal specifications each time
8188 you use @code{x}. For example, after you have inspected three machine
8189 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
8190 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
8191 the repeat count @var{n} is used again; the other arguments default as
8192 for successive uses of @code{x}.
8194 When examining machine instructions, the instruction at current program
8195 counter is shown with a @code{=>} marker. For example:
8198 (@value{GDBP}) x/5i $pc-6
8199 0x804837f <main+11>: mov %esp,%ebp
8200 0x8048381 <main+13>: push %ecx
8201 0x8048382 <main+14>: sub $0x4,%esp
8202 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
8203 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
8206 @cindex @code{$_}, @code{$__}, and value history
8207 The addresses and contents printed by the @code{x} command are not saved
8208 in the value history because there is often too much of them and they
8209 would get in the way. Instead, @value{GDBN} makes these values available for
8210 subsequent use in expressions as values of the convenience variables
8211 @code{$_} and @code{$__}. After an @code{x} command, the last address
8212 examined is available for use in expressions in the convenience variable
8213 @code{$_}. The contents of that address, as examined, are available in
8214 the convenience variable @code{$__}.
8216 If the @code{x} command has a repeat count, the address and contents saved
8217 are from the last memory unit printed; this is not the same as the last
8218 address printed if several units were printed on the last line of output.
8220 @cindex remote memory comparison
8221 @cindex verify remote memory image
8222 When you are debugging a program running on a remote target machine
8223 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
8224 remote machine's memory against the executable file you downloaded to
8225 the target. The @code{compare-sections} command is provided for such
8229 @kindex compare-sections
8230 @item compare-sections @r{[}@var{section-name}@r{]}
8231 Compare the data of a loadable section @var{section-name} in the
8232 executable file of the program being debugged with the same section in
8233 the remote machine's memory, and report any mismatches. With no
8234 arguments, compares all loadable sections. This command's
8235 availability depends on the target's support for the @code{"qCRC"}
8240 @section Automatic Display
8241 @cindex automatic display
8242 @cindex display of expressions
8244 If you find that you want to print the value of an expression frequently
8245 (to see how it changes), you might want to add it to the @dfn{automatic
8246 display list} so that @value{GDBN} prints its value each time your program stops.
8247 Each expression added to the list is given a number to identify it;
8248 to remove an expression from the list, you specify that number.
8249 The automatic display looks like this:
8253 3: bar[5] = (struct hack *) 0x3804
8257 This display shows item numbers, expressions and their current values. As with
8258 displays you request manually using @code{x} or @code{print}, you can
8259 specify the output format you prefer; in fact, @code{display} decides
8260 whether to use @code{print} or @code{x} depending your format
8261 specification---it uses @code{x} if you specify either the @samp{i}
8262 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
8266 @item display @var{expr}
8267 Add the expression @var{expr} to the list of expressions to display
8268 each time your program stops. @xref{Expressions, ,Expressions}.
8270 @code{display} does not repeat if you press @key{RET} again after using it.
8272 @item display/@var{fmt} @var{expr}
8273 For @var{fmt} specifying only a display format and not a size or
8274 count, add the expression @var{expr} to the auto-display list but
8275 arrange to display it each time in the specified format @var{fmt}.
8276 @xref{Output Formats,,Output Formats}.
8278 @item display/@var{fmt} @var{addr}
8279 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
8280 number of units, add the expression @var{addr} as a memory address to
8281 be examined each time your program stops. Examining means in effect
8282 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
8285 For example, @samp{display/i $pc} can be helpful, to see the machine
8286 instruction about to be executed each time execution stops (@samp{$pc}
8287 is a common name for the program counter; @pxref{Registers, ,Registers}).
8290 @kindex delete display
8292 @item undisplay @var{dnums}@dots{}
8293 @itemx delete display @var{dnums}@dots{}
8294 Remove items from the list of expressions to display. Specify the
8295 numbers of the displays that you want affected with the command
8296 argument @var{dnums}. It can be a single display number, one of the
8297 numbers shown in the first field of the @samp{info display} display;
8298 or it could be a range of display numbers, as in @code{2-4}.
8300 @code{undisplay} does not repeat if you press @key{RET} after using it.
8301 (Otherwise you would just get the error @samp{No display number @dots{}}.)
8303 @kindex disable display
8304 @item disable display @var{dnums}@dots{}
8305 Disable the display of item numbers @var{dnums}. A disabled display
8306 item is not printed automatically, but is not forgotten. It may be
8307 enabled again later. Specify the numbers of the displays that you
8308 want affected with the command argument @var{dnums}. It can be a
8309 single display number, one of the numbers shown in the first field of
8310 the @samp{info display} display; or it could be a range of display
8311 numbers, as in @code{2-4}.
8313 @kindex enable display
8314 @item enable display @var{dnums}@dots{}
8315 Enable display of item numbers @var{dnums}. It becomes effective once
8316 again in auto display of its expression, until you specify otherwise.
8317 Specify the numbers of the displays that you want affected with the
8318 command argument @var{dnums}. It can be a single display number, one
8319 of the numbers shown in the first field of the @samp{info display}
8320 display; or it could be a range of display numbers, as in @code{2-4}.
8323 Display the current values of the expressions on the list, just as is
8324 done when your program stops.
8326 @kindex info display
8328 Print the list of expressions previously set up to display
8329 automatically, each one with its item number, but without showing the
8330 values. This includes disabled expressions, which are marked as such.
8331 It also includes expressions which would not be displayed right now
8332 because they refer to automatic variables not currently available.
8335 @cindex display disabled out of scope
8336 If a display expression refers to local variables, then it does not make
8337 sense outside the lexical context for which it was set up. Such an
8338 expression is disabled when execution enters a context where one of its
8339 variables is not defined. For example, if you give the command
8340 @code{display last_char} while inside a function with an argument
8341 @code{last_char}, @value{GDBN} displays this argument while your program
8342 continues to stop inside that function. When it stops elsewhere---where
8343 there is no variable @code{last_char}---the display is disabled
8344 automatically. The next time your program stops where @code{last_char}
8345 is meaningful, you can enable the display expression once again.
8347 @node Print Settings
8348 @section Print Settings
8350 @cindex format options
8351 @cindex print settings
8352 @value{GDBN} provides the following ways to control how arrays, structures,
8353 and symbols are printed.
8356 These settings are useful for debugging programs in any language:
8360 @item set print address
8361 @itemx set print address on
8362 @cindex print/don't print memory addresses
8363 @value{GDBN} prints memory addresses showing the location of stack
8364 traces, structure values, pointer values, breakpoints, and so forth,
8365 even when it also displays the contents of those addresses. The default
8366 is @code{on}. For example, this is what a stack frame display looks like with
8367 @code{set print address on}:
8372 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
8374 530 if (lquote != def_lquote)
8378 @item set print address off
8379 Do not print addresses when displaying their contents. For example,
8380 this is the same stack frame displayed with @code{set print address off}:
8384 (@value{GDBP}) set print addr off
8386 #0 set_quotes (lq="<<", rq=">>") at input.c:530
8387 530 if (lquote != def_lquote)
8391 You can use @samp{set print address off} to eliminate all machine
8392 dependent displays from the @value{GDBN} interface. For example, with
8393 @code{print address off}, you should get the same text for backtraces on
8394 all machines---whether or not they involve pointer arguments.
8397 @item show print address
8398 Show whether or not addresses are to be printed.
8401 When @value{GDBN} prints a symbolic address, it normally prints the
8402 closest earlier symbol plus an offset. If that symbol does not uniquely
8403 identify the address (for example, it is a name whose scope is a single
8404 source file), you may need to clarify. One way to do this is with
8405 @code{info line}, for example @samp{info line *0x4537}. Alternately,
8406 you can set @value{GDBN} to print the source file and line number when
8407 it prints a symbolic address:
8410 @item set print symbol-filename on
8411 @cindex source file and line of a symbol
8412 @cindex symbol, source file and line
8413 Tell @value{GDBN} to print the source file name and line number of a
8414 symbol in the symbolic form of an address.
8416 @item set print symbol-filename off
8417 Do not print source file name and line number of a symbol. This is the
8420 @item show print symbol-filename
8421 Show whether or not @value{GDBN} will print the source file name and
8422 line number of a symbol in the symbolic form of an address.
8425 Another situation where it is helpful to show symbol filenames and line
8426 numbers is when disassembling code; @value{GDBN} shows you the line
8427 number and source file that corresponds to each instruction.
8429 Also, you may wish to see the symbolic form only if the address being
8430 printed is reasonably close to the closest earlier symbol:
8433 @item set print max-symbolic-offset @var{max-offset}
8434 @cindex maximum value for offset of closest symbol
8435 Tell @value{GDBN} to only display the symbolic form of an address if the
8436 offset between the closest earlier symbol and the address is less than
8437 @var{max-offset}. The default is 0, which tells @value{GDBN}
8438 to always print the symbolic form of an address if any symbol precedes it.
8440 @item show print max-symbolic-offset
8441 Ask how large the maximum offset is that @value{GDBN} prints in a
8445 @cindex wild pointer, interpreting
8446 @cindex pointer, finding referent
8447 If you have a pointer and you are not sure where it points, try
8448 @samp{set print symbol-filename on}. Then you can determine the name
8449 and source file location of the variable where it points, using
8450 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
8451 For example, here @value{GDBN} shows that a variable @code{ptt} points
8452 at another variable @code{t}, defined in @file{hi2.c}:
8455 (@value{GDBP}) set print symbol-filename on
8456 (@value{GDBP}) p/a ptt
8457 $4 = 0xe008 <t in hi2.c>
8461 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
8462 does not show the symbol name and filename of the referent, even with
8463 the appropriate @code{set print} options turned on.
8466 You can also enable @samp{/a}-like formatting all the time using
8467 @samp{set print symbol on}:
8470 @item set print symbol on
8471 Tell @value{GDBN} to print the symbol corresponding to an address, if
8474 @item set print symbol off
8475 Tell @value{GDBN} not to print the symbol corresponding to an
8476 address. In this mode, @value{GDBN} will still print the symbol
8477 corresponding to pointers to functions. This is the default.
8479 @item show print symbol
8480 Show whether @value{GDBN} will display the symbol corresponding to an
8484 Other settings control how different kinds of objects are printed:
8487 @item set print array
8488 @itemx set print array on
8489 @cindex pretty print arrays
8490 Pretty print arrays. This format is more convenient to read,
8491 but uses more space. The default is off.
8493 @item set print array off
8494 Return to compressed format for arrays.
8496 @item show print array
8497 Show whether compressed or pretty format is selected for displaying
8500 @cindex print array indexes
8501 @item set print array-indexes
8502 @itemx set print array-indexes on
8503 Print the index of each element when displaying arrays. May be more
8504 convenient to locate a given element in the array or quickly find the
8505 index of a given element in that printed array. The default is off.
8507 @item set print array-indexes off
8508 Stop printing element indexes when displaying arrays.
8510 @item show print array-indexes
8511 Show whether the index of each element is printed when displaying
8514 @item set print elements @var{number-of-elements}
8515 @cindex number of array elements to print
8516 @cindex limit on number of printed array elements
8517 Set a limit on how many elements of an array @value{GDBN} will print.
8518 If @value{GDBN} is printing a large array, it stops printing after it has
8519 printed the number of elements set by the @code{set print elements} command.
8520 This limit also applies to the display of strings.
8521 When @value{GDBN} starts, this limit is set to 200.
8522 Setting @var{number-of-elements} to zero means that the printing is unlimited.
8524 @item show print elements
8525 Display the number of elements of a large array that @value{GDBN} will print.
8526 If the number is 0, then the printing is unlimited.
8528 @item set print frame-arguments @var{value}
8529 @kindex set print frame-arguments
8530 @cindex printing frame argument values
8531 @cindex print all frame argument values
8532 @cindex print frame argument values for scalars only
8533 @cindex do not print frame argument values
8534 This command allows to control how the values of arguments are printed
8535 when the debugger prints a frame (@pxref{Frames}). The possible
8540 The values of all arguments are printed.
8543 Print the value of an argument only if it is a scalar. The value of more
8544 complex arguments such as arrays, structures, unions, etc, is replaced
8545 by @code{@dots{}}. This is the default. Here is an example where
8546 only scalar arguments are shown:
8549 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
8554 None of the argument values are printed. Instead, the value of each argument
8555 is replaced by @code{@dots{}}. In this case, the example above now becomes:
8558 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
8563 By default, only scalar arguments are printed. This command can be used
8564 to configure the debugger to print the value of all arguments, regardless
8565 of their type. However, it is often advantageous to not print the value
8566 of more complex parameters. For instance, it reduces the amount of
8567 information printed in each frame, making the backtrace more readable.
8568 Also, it improves performance when displaying Ada frames, because
8569 the computation of large arguments can sometimes be CPU-intensive,
8570 especially in large applications. Setting @code{print frame-arguments}
8571 to @code{scalars} (the default) or @code{none} avoids this computation,
8572 thus speeding up the display of each Ada frame.
8574 @item show print frame-arguments
8575 Show how the value of arguments should be displayed when printing a frame.
8577 @anchor{set print entry-values}
8578 @item set print entry-values @var{value}
8579 @kindex set print entry-values
8580 Set printing of frame argument values at function entry. In some cases
8581 @value{GDBN} can determine the value of function argument which was passed by
8582 the function caller, even if the value was modified inside the called function
8583 and therefore is different. With optimized code, the current value could be
8584 unavailable, but the entry value may still be known.
8586 The default value is @code{default} (see below for its description). Older
8587 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
8588 this feature will behave in the @code{default} setting the same way as with the
8591 This functionality is currently supported only by DWARF 2 debugging format and
8592 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
8593 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
8596 The @var{value} parameter can be one of the following:
8600 Print only actual parameter values, never print values from function entry
8604 #0 different (val=6)
8605 #0 lost (val=<optimized out>)
8607 #0 invalid (val=<optimized out>)
8611 Print only parameter values from function entry point. The actual parameter
8612 values are never printed.
8614 #0 equal (val@@entry=5)
8615 #0 different (val@@entry=5)
8616 #0 lost (val@@entry=5)
8617 #0 born (val@@entry=<optimized out>)
8618 #0 invalid (val@@entry=<optimized out>)
8622 Print only parameter values from function entry point. If value from function
8623 entry point is not known while the actual value is known, print the actual
8624 value for such parameter.
8626 #0 equal (val@@entry=5)
8627 #0 different (val@@entry=5)
8628 #0 lost (val@@entry=5)
8630 #0 invalid (val@@entry=<optimized out>)
8634 Print actual parameter values. If actual parameter value is not known while
8635 value from function entry point is known, print the entry point value for such
8639 #0 different (val=6)
8640 #0 lost (val@@entry=5)
8642 #0 invalid (val=<optimized out>)
8646 Always print both the actual parameter value and its value from function entry
8647 point, even if values of one or both are not available due to compiler
8650 #0 equal (val=5, val@@entry=5)
8651 #0 different (val=6, val@@entry=5)
8652 #0 lost (val=<optimized out>, val@@entry=5)
8653 #0 born (val=10, val@@entry=<optimized out>)
8654 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
8658 Print the actual parameter value if it is known and also its value from
8659 function entry point if it is known. If neither is known, print for the actual
8660 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
8661 values are known and identical, print the shortened
8662 @code{param=param@@entry=VALUE} notation.
8664 #0 equal (val=val@@entry=5)
8665 #0 different (val=6, val@@entry=5)
8666 #0 lost (val@@entry=5)
8668 #0 invalid (val=<optimized out>)
8672 Always print the actual parameter value. Print also its value from function
8673 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
8674 if both values are known and identical, print the shortened
8675 @code{param=param@@entry=VALUE} notation.
8677 #0 equal (val=val@@entry=5)
8678 #0 different (val=6, val@@entry=5)
8679 #0 lost (val=<optimized out>, val@@entry=5)
8681 #0 invalid (val=<optimized out>)
8685 For analysis messages on possible failures of frame argument values at function
8686 entry resolution see @ref{set debug entry-values}.
8688 @item show print entry-values
8689 Show the method being used for printing of frame argument values at function
8692 @item set print repeats
8693 @cindex repeated array elements
8694 Set the threshold for suppressing display of repeated array
8695 elements. When the number of consecutive identical elements of an
8696 array exceeds the threshold, @value{GDBN} prints the string
8697 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
8698 identical repetitions, instead of displaying the identical elements
8699 themselves. Setting the threshold to zero will cause all elements to
8700 be individually printed. The default threshold is 10.
8702 @item show print repeats
8703 Display the current threshold for printing repeated identical
8706 @item set print null-stop
8707 @cindex @sc{null} elements in arrays
8708 Cause @value{GDBN} to stop printing the characters of an array when the first
8709 @sc{null} is encountered. This is useful when large arrays actually
8710 contain only short strings.
8713 @item show print null-stop
8714 Show whether @value{GDBN} stops printing an array on the first
8715 @sc{null} character.
8717 @item set print pretty on
8718 @cindex print structures in indented form
8719 @cindex indentation in structure display
8720 Cause @value{GDBN} to print structures in an indented format with one member
8721 per line, like this:
8736 @item set print pretty off
8737 Cause @value{GDBN} to print structures in a compact format, like this:
8741 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
8742 meat = 0x54 "Pork"@}
8747 This is the default format.
8749 @item show print pretty
8750 Show which format @value{GDBN} is using to print structures.
8752 @item set print sevenbit-strings on
8753 @cindex eight-bit characters in strings
8754 @cindex octal escapes in strings
8755 Print using only seven-bit characters; if this option is set,
8756 @value{GDBN} displays any eight-bit characters (in strings or
8757 character values) using the notation @code{\}@var{nnn}. This setting is
8758 best if you are working in English (@sc{ascii}) and you use the
8759 high-order bit of characters as a marker or ``meta'' bit.
8761 @item set print sevenbit-strings off
8762 Print full eight-bit characters. This allows the use of more
8763 international character sets, and is the default.
8765 @item show print sevenbit-strings
8766 Show whether or not @value{GDBN} is printing only seven-bit characters.
8768 @item set print union on
8769 @cindex unions in structures, printing
8770 Tell @value{GDBN} to print unions which are contained in structures
8771 and other unions. This is the default setting.
8773 @item set print union off
8774 Tell @value{GDBN} not to print unions which are contained in
8775 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
8778 @item show print union
8779 Ask @value{GDBN} whether or not it will print unions which are contained in
8780 structures and other unions.
8782 For example, given the declarations
8785 typedef enum @{Tree, Bug@} Species;
8786 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
8787 typedef enum @{Caterpillar, Cocoon, Butterfly@}
8798 struct thing foo = @{Tree, @{Acorn@}@};
8802 with @code{set print union on} in effect @samp{p foo} would print
8805 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
8809 and with @code{set print union off} in effect it would print
8812 $1 = @{it = Tree, form = @{...@}@}
8816 @code{set print union} affects programs written in C-like languages
8822 These settings are of interest when debugging C@t{++} programs:
8825 @cindex demangling C@t{++} names
8826 @item set print demangle
8827 @itemx set print demangle on
8828 Print C@t{++} names in their source form rather than in the encoded
8829 (``mangled'') form passed to the assembler and linker for type-safe
8830 linkage. The default is on.
8832 @item show print demangle
8833 Show whether C@t{++} names are printed in mangled or demangled form.
8835 @item set print asm-demangle
8836 @itemx set print asm-demangle on
8837 Print C@t{++} names in their source form rather than their mangled form, even
8838 in assembler code printouts such as instruction disassemblies.
8841 @item show print asm-demangle
8842 Show whether C@t{++} names in assembly listings are printed in mangled
8845 @cindex C@t{++} symbol decoding style
8846 @cindex symbol decoding style, C@t{++}
8847 @kindex set demangle-style
8848 @item set demangle-style @var{style}
8849 Choose among several encoding schemes used by different compilers to
8850 represent C@t{++} names. The choices for @var{style} are currently:
8854 Allow @value{GDBN} to choose a decoding style by inspecting your program.
8857 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
8858 This is the default.
8861 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
8864 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
8867 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
8868 @strong{Warning:} this setting alone is not sufficient to allow
8869 debugging @code{cfront}-generated executables. @value{GDBN} would
8870 require further enhancement to permit that.
8873 If you omit @var{style}, you will see a list of possible formats.
8875 @item show demangle-style
8876 Display the encoding style currently in use for decoding C@t{++} symbols.
8878 @item set print object
8879 @itemx set print object on
8880 @cindex derived type of an object, printing
8881 @cindex display derived types
8882 When displaying a pointer to an object, identify the @emph{actual}
8883 (derived) type of the object rather than the @emph{declared} type, using
8884 the virtual function table. Note that the virtual function table is
8885 required---this feature can only work for objects that have run-time
8886 type identification; a single virtual method in the object's declared
8887 type is sufficient. Note that this setting is also taken into account when
8888 working with variable objects via MI (@pxref{GDB/MI}).
8890 @item set print object off
8891 Display only the declared type of objects, without reference to the
8892 virtual function table. This is the default setting.
8894 @item show print object
8895 Show whether actual, or declared, object types are displayed.
8897 @item set print static-members
8898 @itemx set print static-members on
8899 @cindex static members of C@t{++} objects
8900 Print static members when displaying a C@t{++} object. The default is on.
8902 @item set print static-members off
8903 Do not print static members when displaying a C@t{++} object.
8905 @item show print static-members
8906 Show whether C@t{++} static members are printed or not.
8908 @item set print pascal_static-members
8909 @itemx set print pascal_static-members on
8910 @cindex static members of Pascal objects
8911 @cindex Pascal objects, static members display
8912 Print static members when displaying a Pascal object. The default is on.
8914 @item set print pascal_static-members off
8915 Do not print static members when displaying a Pascal object.
8917 @item show print pascal_static-members
8918 Show whether Pascal static members are printed or not.
8920 @c These don't work with HP ANSI C++ yet.
8921 @item set print vtbl
8922 @itemx set print vtbl on
8923 @cindex pretty print C@t{++} virtual function tables
8924 @cindex virtual functions (C@t{++}) display
8925 @cindex VTBL display
8926 Pretty print C@t{++} virtual function tables. The default is off.
8927 (The @code{vtbl} commands do not work on programs compiled with the HP
8928 ANSI C@t{++} compiler (@code{aCC}).)
8930 @item set print vtbl off
8931 Do not pretty print C@t{++} virtual function tables.
8933 @item show print vtbl
8934 Show whether C@t{++} virtual function tables are pretty printed, or not.
8937 @node Pretty Printing
8938 @section Pretty Printing
8940 @value{GDBN} provides a mechanism to allow pretty-printing of values using
8941 Python code. It greatly simplifies the display of complex objects. This
8942 mechanism works for both MI and the CLI.
8945 * Pretty-Printer Introduction:: Introduction to pretty-printers
8946 * Pretty-Printer Example:: An example pretty-printer
8947 * Pretty-Printer Commands:: Pretty-printer commands
8950 @node Pretty-Printer Introduction
8951 @subsection Pretty-Printer Introduction
8953 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
8954 registered for the value. If there is then @value{GDBN} invokes the
8955 pretty-printer to print the value. Otherwise the value is printed normally.
8957 Pretty-printers are normally named. This makes them easy to manage.
8958 The @samp{info pretty-printer} command will list all the installed
8959 pretty-printers with their names.
8960 If a pretty-printer can handle multiple data types, then its
8961 @dfn{subprinters} are the printers for the individual data types.
8962 Each such subprinter has its own name.
8963 The format of the name is @var{printer-name};@var{subprinter-name}.
8965 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
8966 Typically they are automatically loaded and registered when the corresponding
8967 debug information is loaded, thus making them available without having to
8968 do anything special.
8970 There are three places where a pretty-printer can be registered.
8974 Pretty-printers registered globally are available when debugging
8978 Pretty-printers registered with a program space are available only
8979 when debugging that program.
8980 @xref{Progspaces In Python}, for more details on program spaces in Python.
8983 Pretty-printers registered with an objfile are loaded and unloaded
8984 with the corresponding objfile (e.g., shared library).
8985 @xref{Objfiles In Python}, for more details on objfiles in Python.
8988 @xref{Selecting Pretty-Printers}, for further information on how
8989 pretty-printers are selected,
8991 @xref{Writing a Pretty-Printer}, for implementing pretty printers
8994 @node Pretty-Printer Example
8995 @subsection Pretty-Printer Example
8997 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
9000 (@value{GDBP}) print s
9002 static npos = 4294967295,
9004 <std::allocator<char>> = @{
9005 <__gnu_cxx::new_allocator<char>> = @{
9006 <No data fields>@}, <No data fields>
9008 members of std::basic_string<char, std::char_traits<char>,
9009 std::allocator<char> >::_Alloc_hider:
9010 _M_p = 0x804a014 "abcd"
9015 With a pretty-printer for @code{std::string} only the contents are printed:
9018 (@value{GDBP}) print s
9022 @node Pretty-Printer Commands
9023 @subsection Pretty-Printer Commands
9024 @cindex pretty-printer commands
9027 @kindex info pretty-printer
9028 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9029 Print the list of installed pretty-printers.
9030 This includes disabled pretty-printers, which are marked as such.
9032 @var{object-regexp} is a regular expression matching the objects
9033 whose pretty-printers to list.
9034 Objects can be @code{global}, the program space's file
9035 (@pxref{Progspaces In Python}),
9036 and the object files within that program space (@pxref{Objfiles In Python}).
9037 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
9038 looks up a printer from these three objects.
9040 @var{name-regexp} is a regular expression matching the name of the printers
9043 @kindex disable pretty-printer
9044 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9045 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
9046 A disabled pretty-printer is not forgotten, it may be enabled again later.
9048 @kindex enable pretty-printer
9049 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9050 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
9055 Suppose we have three pretty-printers installed: one from library1.so
9056 named @code{foo} that prints objects of type @code{foo}, and
9057 another from library2.so named @code{bar} that prints two types of objects,
9058 @code{bar1} and @code{bar2}.
9061 (gdb) info pretty-printer
9068 (gdb) info pretty-printer library2
9073 (gdb) disable pretty-printer library1
9075 2 of 3 printers enabled
9076 (gdb) info pretty-printer
9083 (gdb) disable pretty-printer library2 bar:bar1
9085 1 of 3 printers enabled
9086 (gdb) info pretty-printer library2
9093 (gdb) disable pretty-printer library2 bar
9095 0 of 3 printers enabled
9096 (gdb) info pretty-printer library2
9105 Note that for @code{bar} the entire printer can be disabled,
9106 as can each individual subprinter.
9109 @section Value History
9111 @cindex value history
9112 @cindex history of values printed by @value{GDBN}
9113 Values printed by the @code{print} command are saved in the @value{GDBN}
9114 @dfn{value history}. This allows you to refer to them in other expressions.
9115 Values are kept until the symbol table is re-read or discarded
9116 (for example with the @code{file} or @code{symbol-file} commands).
9117 When the symbol table changes, the value history is discarded,
9118 since the values may contain pointers back to the types defined in the
9123 @cindex history number
9124 The values printed are given @dfn{history numbers} by which you can
9125 refer to them. These are successive integers starting with one.
9126 @code{print} shows you the history number assigned to a value by
9127 printing @samp{$@var{num} = } before the value; here @var{num} is the
9130 To refer to any previous value, use @samp{$} followed by the value's
9131 history number. The way @code{print} labels its output is designed to
9132 remind you of this. Just @code{$} refers to the most recent value in
9133 the history, and @code{$$} refers to the value before that.
9134 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
9135 is the value just prior to @code{$$}, @code{$$1} is equivalent to
9136 @code{$$}, and @code{$$0} is equivalent to @code{$}.
9138 For example, suppose you have just printed a pointer to a structure and
9139 want to see the contents of the structure. It suffices to type
9145 If you have a chain of structures where the component @code{next} points
9146 to the next one, you can print the contents of the next one with this:
9153 You can print successive links in the chain by repeating this
9154 command---which you can do by just typing @key{RET}.
9156 Note that the history records values, not expressions. If the value of
9157 @code{x} is 4 and you type these commands:
9165 then the value recorded in the value history by the @code{print} command
9166 remains 4 even though the value of @code{x} has changed.
9171 Print the last ten values in the value history, with their item numbers.
9172 This is like @samp{p@ $$9} repeated ten times, except that @code{show
9173 values} does not change the history.
9175 @item show values @var{n}
9176 Print ten history values centered on history item number @var{n}.
9179 Print ten history values just after the values last printed. If no more
9180 values are available, @code{show values +} produces no display.
9183 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
9184 same effect as @samp{show values +}.
9186 @node Convenience Vars
9187 @section Convenience Variables
9189 @cindex convenience variables
9190 @cindex user-defined variables
9191 @value{GDBN} provides @dfn{convenience variables} that you can use within
9192 @value{GDBN} to hold on to a value and refer to it later. These variables
9193 exist entirely within @value{GDBN}; they are not part of your program, and
9194 setting a convenience variable has no direct effect on further execution
9195 of your program. That is why you can use them freely.
9197 Convenience variables are prefixed with @samp{$}. Any name preceded by
9198 @samp{$} can be used for a convenience variable, unless it is one of
9199 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
9200 (Value history references, in contrast, are @emph{numbers} preceded
9201 by @samp{$}. @xref{Value History, ,Value History}.)
9203 You can save a value in a convenience variable with an assignment
9204 expression, just as you would set a variable in your program.
9208 set $foo = *object_ptr
9212 would save in @code{$foo} the value contained in the object pointed to by
9215 Using a convenience variable for the first time creates it, but its
9216 value is @code{void} until you assign a new value. You can alter the
9217 value with another assignment at any time.
9219 Convenience variables have no fixed types. You can assign a convenience
9220 variable any type of value, including structures and arrays, even if
9221 that variable already has a value of a different type. The convenience
9222 variable, when used as an expression, has the type of its current value.
9225 @kindex show convenience
9226 @cindex show all user variables
9227 @item show convenience
9228 Print a list of convenience variables used so far, and their values.
9229 Abbreviated @code{show conv}.
9231 @kindex init-if-undefined
9232 @cindex convenience variables, initializing
9233 @item init-if-undefined $@var{variable} = @var{expression}
9234 Set a convenience variable if it has not already been set. This is useful
9235 for user-defined commands that keep some state. It is similar, in concept,
9236 to using local static variables with initializers in C (except that
9237 convenience variables are global). It can also be used to allow users to
9238 override default values used in a command script.
9240 If the variable is already defined then the expression is not evaluated so
9241 any side-effects do not occur.
9244 One of the ways to use a convenience variable is as a counter to be
9245 incremented or a pointer to be advanced. For example, to print
9246 a field from successive elements of an array of structures:
9250 print bar[$i++]->contents
9254 Repeat that command by typing @key{RET}.
9256 Some convenience variables are created automatically by @value{GDBN} and given
9257 values likely to be useful.
9260 @vindex $_@r{, convenience variable}
9262 The variable @code{$_} is automatically set by the @code{x} command to
9263 the last address examined (@pxref{Memory, ,Examining Memory}). Other
9264 commands which provide a default address for @code{x} to examine also
9265 set @code{$_} to that address; these commands include @code{info line}
9266 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
9267 except when set by the @code{x} command, in which case it is a pointer
9268 to the type of @code{$__}.
9270 @vindex $__@r{, convenience variable}
9272 The variable @code{$__} is automatically set by the @code{x} command
9273 to the value found in the last address examined. Its type is chosen
9274 to match the format in which the data was printed.
9277 @vindex $_exitcode@r{, convenience variable}
9278 The variable @code{$_exitcode} is automatically set to the exit code when
9279 the program being debugged terminates.
9282 @itemx $_probe_arg0@dots{}$_probe_arg11
9283 Arguments to a static probe. @xref{Static Probe Points}.
9286 @vindex $_sdata@r{, inspect, convenience variable}
9287 The variable @code{$_sdata} contains extra collected static tracepoint
9288 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
9289 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
9290 if extra static tracepoint data has not been collected.
9293 @vindex $_siginfo@r{, convenience variable}
9294 The variable @code{$_siginfo} contains extra signal information
9295 (@pxref{extra signal information}). Note that @code{$_siginfo}
9296 could be empty, if the application has not yet received any signals.
9297 For example, it will be empty before you execute the @code{run} command.
9300 @vindex $_tlb@r{, convenience variable}
9301 The variable @code{$_tlb} is automatically set when debugging
9302 applications running on MS-Windows in native mode or connected to
9303 gdbserver that supports the @code{qGetTIBAddr} request.
9304 @xref{General Query Packets}.
9305 This variable contains the address of the thread information block.
9309 On HP-UX systems, if you refer to a function or variable name that
9310 begins with a dollar sign, @value{GDBN} searches for a user or system
9311 name first, before it searches for a convenience variable.
9313 @cindex convenience functions
9314 @value{GDBN} also supplies some @dfn{convenience functions}. These
9315 have a syntax similar to convenience variables. A convenience
9316 function can be used in an expression just like an ordinary function;
9317 however, a convenience function is implemented internally to
9322 @kindex help function
9323 @cindex show all convenience functions
9324 Print a list of all convenience functions.
9331 You can refer to machine register contents, in expressions, as variables
9332 with names starting with @samp{$}. The names of registers are different
9333 for each machine; use @code{info registers} to see the names used on
9337 @kindex info registers
9338 @item info registers
9339 Print the names and values of all registers except floating-point
9340 and vector registers (in the selected stack frame).
9342 @kindex info all-registers
9343 @cindex floating point registers
9344 @item info all-registers
9345 Print the names and values of all registers, including floating-point
9346 and vector registers (in the selected stack frame).
9348 @item info registers @var{regname} @dots{}
9349 Print the @dfn{relativized} value of each specified register @var{regname}.
9350 As discussed in detail below, register values are normally relative to
9351 the selected stack frame. @var{regname} may be any register name valid on
9352 the machine you are using, with or without the initial @samp{$}.
9355 @cindex stack pointer register
9356 @cindex program counter register
9357 @cindex process status register
9358 @cindex frame pointer register
9359 @cindex standard registers
9360 @value{GDBN} has four ``standard'' register names that are available (in
9361 expressions) on most machines---whenever they do not conflict with an
9362 architecture's canonical mnemonics for registers. The register names
9363 @code{$pc} and @code{$sp} are used for the program counter register and
9364 the stack pointer. @code{$fp} is used for a register that contains a
9365 pointer to the current stack frame, and @code{$ps} is used for a
9366 register that contains the processor status. For example,
9367 you could print the program counter in hex with
9374 or print the instruction to be executed next with
9381 or add four to the stack pointer@footnote{This is a way of removing
9382 one word from the stack, on machines where stacks grow downward in
9383 memory (most machines, nowadays). This assumes that the innermost
9384 stack frame is selected; setting @code{$sp} is not allowed when other
9385 stack frames are selected. To pop entire frames off the stack,
9386 regardless of machine architecture, use @code{return};
9387 see @ref{Returning, ,Returning from a Function}.} with
9393 Whenever possible, these four standard register names are available on
9394 your machine even though the machine has different canonical mnemonics,
9395 so long as there is no conflict. The @code{info registers} command
9396 shows the canonical names. For example, on the SPARC, @code{info
9397 registers} displays the processor status register as @code{$psr} but you
9398 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
9399 is an alias for the @sc{eflags} register.
9401 @value{GDBN} always considers the contents of an ordinary register as an
9402 integer when the register is examined in this way. Some machines have
9403 special registers which can hold nothing but floating point; these
9404 registers are considered to have floating point values. There is no way
9405 to refer to the contents of an ordinary register as floating point value
9406 (although you can @emph{print} it as a floating point value with
9407 @samp{print/f $@var{regname}}).
9409 Some registers have distinct ``raw'' and ``virtual'' data formats. This
9410 means that the data format in which the register contents are saved by
9411 the operating system is not the same one that your program normally
9412 sees. For example, the registers of the 68881 floating point
9413 coprocessor are always saved in ``extended'' (raw) format, but all C
9414 programs expect to work with ``double'' (virtual) format. In such
9415 cases, @value{GDBN} normally works with the virtual format only (the format
9416 that makes sense for your program), but the @code{info registers} command
9417 prints the data in both formats.
9419 @cindex SSE registers (x86)
9420 @cindex MMX registers (x86)
9421 Some machines have special registers whose contents can be interpreted
9422 in several different ways. For example, modern x86-based machines
9423 have SSE and MMX registers that can hold several values packed
9424 together in several different formats. @value{GDBN} refers to such
9425 registers in @code{struct} notation:
9428 (@value{GDBP}) print $xmm1
9430 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
9431 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
9432 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
9433 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
9434 v4_int32 = @{0, 20657912, 11, 13@},
9435 v2_int64 = @{88725056443645952, 55834574859@},
9436 uint128 = 0x0000000d0000000b013b36f800000000
9441 To set values of such registers, you need to tell @value{GDBN} which
9442 view of the register you wish to change, as if you were assigning
9443 value to a @code{struct} member:
9446 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
9449 Normally, register values are relative to the selected stack frame
9450 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
9451 value that the register would contain if all stack frames farther in
9452 were exited and their saved registers restored. In order to see the
9453 true contents of hardware registers, you must select the innermost
9454 frame (with @samp{frame 0}).
9456 However, @value{GDBN} must deduce where registers are saved, from the machine
9457 code generated by your compiler. If some registers are not saved, or if
9458 @value{GDBN} is unable to locate the saved registers, the selected stack
9459 frame makes no difference.
9461 @node Floating Point Hardware
9462 @section Floating Point Hardware
9463 @cindex floating point
9465 Depending on the configuration, @value{GDBN} may be able to give
9466 you more information about the status of the floating point hardware.
9471 Display hardware-dependent information about the floating
9472 point unit. The exact contents and layout vary depending on the
9473 floating point chip. Currently, @samp{info float} is supported on
9474 the ARM and x86 machines.
9478 @section Vector Unit
9481 Depending on the configuration, @value{GDBN} may be able to give you
9482 more information about the status of the vector unit.
9487 Display information about the vector unit. The exact contents and
9488 layout vary depending on the hardware.
9491 @node OS Information
9492 @section Operating System Auxiliary Information
9493 @cindex OS information
9495 @value{GDBN} provides interfaces to useful OS facilities that can help
9496 you debug your program.
9498 @cindex @code{ptrace} system call
9499 @cindex @code{struct user} contents
9500 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
9501 machines), it interfaces with the inferior via the @code{ptrace}
9502 system call. The operating system creates a special sata structure,
9503 called @code{struct user}, for this interface. You can use the
9504 command @code{info udot} to display the contents of this data
9510 Display the contents of the @code{struct user} maintained by the OS
9511 kernel for the program being debugged. @value{GDBN} displays the
9512 contents of @code{struct user} as a list of hex numbers, similar to
9513 the @code{examine} command.
9516 @cindex auxiliary vector
9517 @cindex vector, auxiliary
9518 Some operating systems supply an @dfn{auxiliary vector} to programs at
9519 startup. This is akin to the arguments and environment that you
9520 specify for a program, but contains a system-dependent variety of
9521 binary values that tell system libraries important details about the
9522 hardware, operating system, and process. Each value's purpose is
9523 identified by an integer tag; the meanings are well-known but system-specific.
9524 Depending on the configuration and operating system facilities,
9525 @value{GDBN} may be able to show you this information. For remote
9526 targets, this functionality may further depend on the remote stub's
9527 support of the @samp{qXfer:auxv:read} packet, see
9528 @ref{qXfer auxiliary vector read}.
9533 Display the auxiliary vector of the inferior, which can be either a
9534 live process or a core dump file. @value{GDBN} prints each tag value
9535 numerically, and also shows names and text descriptions for recognized
9536 tags. Some values in the vector are numbers, some bit masks, and some
9537 pointers to strings or other data. @value{GDBN} displays each value in the
9538 most appropriate form for a recognized tag, and in hexadecimal for
9539 an unrecognized tag.
9542 On some targets, @value{GDBN} can access operating system-specific
9543 information and show it to you. The types of information available
9544 will differ depending on the type of operating system running on the
9545 target. The mechanism used to fetch the data is described in
9546 @ref{Operating System Information}. For remote targets, this
9547 functionality depends on the remote stub's support of the
9548 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
9552 @item info os @var{infotype}
9554 Display OS information of the requested type.
9556 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
9558 @anchor{linux info os infotypes}
9560 @kindex info os processes
9562 Display the list of processes on the target. For each process,
9563 @value{GDBN} prints the process identifier, the name of the user, the
9564 command corresponding to the process, and the list of processor cores
9565 that the process is currently running on. (To understand what these
9566 properties mean, for this and the following info types, please consult
9567 the general @sc{gnu}/Linux documentation.)
9569 @kindex info os procgroups
9571 Display the list of process groups on the target. For each process,
9572 @value{GDBN} prints the identifier of the process group that it belongs
9573 to, the command corresponding to the process group leader, the process
9574 identifier, and the command line of the process. The list is sorted
9575 first by the process group identifier, then by the process identifier,
9576 so that processes belonging to the same process group are grouped together
9577 and the process group leader is listed first.
9579 @kindex info os threads
9581 Display the list of threads running on the target. For each thread,
9582 @value{GDBN} prints the identifier of the process that the thread
9583 belongs to, the command of the process, the thread identifier, and the
9584 processor core that it is currently running on. The main thread of a
9585 process is not listed.
9587 @kindex info os files
9589 Display the list of open file descriptors on the target. For each
9590 file descriptor, @value{GDBN} prints the identifier of the process
9591 owning the descriptor, the command of the owning process, the value
9592 of the descriptor, and the target of the descriptor.
9594 @kindex info os sockets
9596 Display the list of Internet-domain sockets on the target. For each
9597 socket, @value{GDBN} prints the address and port of the local and
9598 remote endpoints, the current state of the connection, the creator of
9599 the socket, the IP address family of the socket, and the type of the
9604 Display the list of all System V shared-memory regions on the target.
9605 For each shared-memory region, @value{GDBN} prints the region key,
9606 the shared-memory identifier, the access permissions, the size of the
9607 region, the process that created the region, the process that last
9608 attached to or detached from the region, the current number of live
9609 attaches to the region, and the times at which the region was last
9610 attached to, detach from, and changed.
9612 @kindex info os semaphores
9614 Display the list of all System V semaphore sets on the target. For each
9615 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
9616 set identifier, the access permissions, the number of semaphores in the
9617 set, the user and group of the owner and creator of the semaphore set,
9618 and the times at which the semaphore set was operated upon and changed.
9622 Display the list of all System V message queues on the target. For each
9623 message queue, @value{GDBN} prints the message queue key, the message
9624 queue identifier, the access permissions, the current number of bytes
9625 on the queue, the current number of messages on the queue, the processes
9626 that last sent and received a message on the queue, the user and group
9627 of the owner and creator of the message queue, the times at which a
9628 message was last sent and received on the queue, and the time at which
9629 the message queue was last changed.
9631 @kindex info os modules
9633 Display the list of all loaded kernel modules on the target. For each
9634 module, @value{GDBN} prints the module name, the size of the module in
9635 bytes, the number of times the module is used, the dependencies of the
9636 module, the status of the module, and the address of the loaded module
9641 If @var{infotype} is omitted, then list the possible values for
9642 @var{infotype} and the kind of OS information available for each
9643 @var{infotype}. If the target does not return a list of possible
9644 types, this command will report an error.
9647 @node Memory Region Attributes
9648 @section Memory Region Attributes
9649 @cindex memory region attributes
9651 @dfn{Memory region attributes} allow you to describe special handling
9652 required by regions of your target's memory. @value{GDBN} uses
9653 attributes to determine whether to allow certain types of memory
9654 accesses; whether to use specific width accesses; and whether to cache
9655 target memory. By default the description of memory regions is
9656 fetched from the target (if the current target supports this), but the
9657 user can override the fetched regions.
9659 Defined memory regions can be individually enabled and disabled. When a
9660 memory region is disabled, @value{GDBN} uses the default attributes when
9661 accessing memory in that region. Similarly, if no memory regions have
9662 been defined, @value{GDBN} uses the default attributes when accessing
9665 When a memory region is defined, it is given a number to identify it;
9666 to enable, disable, or remove a memory region, you specify that number.
9670 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
9671 Define a memory region bounded by @var{lower} and @var{upper} with
9672 attributes @var{attributes}@dots{}, and add it to the list of regions
9673 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
9674 case: it is treated as the target's maximum memory address.
9675 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
9678 Discard any user changes to the memory regions and use target-supplied
9679 regions, if available, or no regions if the target does not support.
9682 @item delete mem @var{nums}@dots{}
9683 Remove memory regions @var{nums}@dots{} from the list of regions
9684 monitored by @value{GDBN}.
9687 @item disable mem @var{nums}@dots{}
9688 Disable monitoring of memory regions @var{nums}@dots{}.
9689 A disabled memory region is not forgotten.
9690 It may be enabled again later.
9693 @item enable mem @var{nums}@dots{}
9694 Enable monitoring of memory regions @var{nums}@dots{}.
9698 Print a table of all defined memory regions, with the following columns
9702 @item Memory Region Number
9703 @item Enabled or Disabled.
9704 Enabled memory regions are marked with @samp{y}.
9705 Disabled memory regions are marked with @samp{n}.
9708 The address defining the inclusive lower bound of the memory region.
9711 The address defining the exclusive upper bound of the memory region.
9714 The list of attributes set for this memory region.
9719 @subsection Attributes
9721 @subsubsection Memory Access Mode
9722 The access mode attributes set whether @value{GDBN} may make read or
9723 write accesses to a memory region.
9725 While these attributes prevent @value{GDBN} from performing invalid
9726 memory accesses, they do nothing to prevent the target system, I/O DMA,
9727 etc.@: from accessing memory.
9731 Memory is read only.
9733 Memory is write only.
9735 Memory is read/write. This is the default.
9738 @subsubsection Memory Access Size
9739 The access size attribute tells @value{GDBN} to use specific sized
9740 accesses in the memory region. Often memory mapped device registers
9741 require specific sized accesses. If no access size attribute is
9742 specified, @value{GDBN} may use accesses of any size.
9746 Use 8 bit memory accesses.
9748 Use 16 bit memory accesses.
9750 Use 32 bit memory accesses.
9752 Use 64 bit memory accesses.
9755 @c @subsubsection Hardware/Software Breakpoints
9756 @c The hardware/software breakpoint attributes set whether @value{GDBN}
9757 @c will use hardware or software breakpoints for the internal breakpoints
9758 @c used by the step, next, finish, until, etc. commands.
9762 @c Always use hardware breakpoints
9763 @c @item swbreak (default)
9766 @subsubsection Data Cache
9767 The data cache attributes set whether @value{GDBN} will cache target
9768 memory. While this generally improves performance by reducing debug
9769 protocol overhead, it can lead to incorrect results because @value{GDBN}
9770 does not know about volatile variables or memory mapped device
9775 Enable @value{GDBN} to cache target memory.
9777 Disable @value{GDBN} from caching target memory. This is the default.
9780 @subsection Memory Access Checking
9781 @value{GDBN} can be instructed to refuse accesses to memory that is
9782 not explicitly described. This can be useful if accessing such
9783 regions has undesired effects for a specific target, or to provide
9784 better error checking. The following commands control this behaviour.
9787 @kindex set mem inaccessible-by-default
9788 @item set mem inaccessible-by-default [on|off]
9789 If @code{on} is specified, make @value{GDBN} treat memory not
9790 explicitly described by the memory ranges as non-existent and refuse accesses
9791 to such memory. The checks are only performed if there's at least one
9792 memory range defined. If @code{off} is specified, make @value{GDBN}
9793 treat the memory not explicitly described by the memory ranges as RAM.
9794 The default value is @code{on}.
9795 @kindex show mem inaccessible-by-default
9796 @item show mem inaccessible-by-default
9797 Show the current handling of accesses to unknown memory.
9801 @c @subsubsection Memory Write Verification
9802 @c The memory write verification attributes set whether @value{GDBN}
9803 @c will re-reads data after each write to verify the write was successful.
9807 @c @item noverify (default)
9810 @node Dump/Restore Files
9811 @section Copy Between Memory and a File
9812 @cindex dump/restore files
9813 @cindex append data to a file
9814 @cindex dump data to a file
9815 @cindex restore data from a file
9817 You can use the commands @code{dump}, @code{append}, and
9818 @code{restore} to copy data between target memory and a file. The
9819 @code{dump} and @code{append} commands write data to a file, and the
9820 @code{restore} command reads data from a file back into the inferior's
9821 memory. Files may be in binary, Motorola S-record, Intel hex, or
9822 Tektronix Hex format; however, @value{GDBN} can only append to binary
9828 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
9829 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
9830 Dump the contents of memory from @var{start_addr} to @var{end_addr},
9831 or the value of @var{expr}, to @var{filename} in the given format.
9833 The @var{format} parameter may be any one of:
9840 Motorola S-record format.
9842 Tektronix Hex format.
9845 @value{GDBN} uses the same definitions of these formats as the
9846 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
9847 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
9851 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
9852 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
9853 Append the contents of memory from @var{start_addr} to @var{end_addr},
9854 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
9855 (@value{GDBN} can only append data to files in raw binary form.)
9858 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
9859 Restore the contents of file @var{filename} into memory. The
9860 @code{restore} command can automatically recognize any known @sc{bfd}
9861 file format, except for raw binary. To restore a raw binary file you
9862 must specify the optional keyword @code{binary} after the filename.
9864 If @var{bias} is non-zero, its value will be added to the addresses
9865 contained in the file. Binary files always start at address zero, so
9866 they will be restored at address @var{bias}. Other bfd files have
9867 a built-in location; they will be restored at offset @var{bias}
9870 If @var{start} and/or @var{end} are non-zero, then only data between
9871 file offset @var{start} and file offset @var{end} will be restored.
9872 These offsets are relative to the addresses in the file, before
9873 the @var{bias} argument is applied.
9877 @node Core File Generation
9878 @section How to Produce a Core File from Your Program
9879 @cindex dump core from inferior
9881 A @dfn{core file} or @dfn{core dump} is a file that records the memory
9882 image of a running process and its process status (register values
9883 etc.). Its primary use is post-mortem debugging of a program that
9884 crashed while it ran outside a debugger. A program that crashes
9885 automatically produces a core file, unless this feature is disabled by
9886 the user. @xref{Files}, for information on invoking @value{GDBN} in
9887 the post-mortem debugging mode.
9889 Occasionally, you may wish to produce a core file of the program you
9890 are debugging in order to preserve a snapshot of its state.
9891 @value{GDBN} has a special command for that.
9895 @kindex generate-core-file
9896 @item generate-core-file [@var{file}]
9897 @itemx gcore [@var{file}]
9898 Produce a core dump of the inferior process. The optional argument
9899 @var{file} specifies the file name where to put the core dump. If not
9900 specified, the file name defaults to @file{core.@var{pid}}, where
9901 @var{pid} is the inferior process ID.
9903 Note that this command is implemented only for some systems (as of
9904 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
9907 @node Character Sets
9908 @section Character Sets
9909 @cindex character sets
9911 @cindex translating between character sets
9912 @cindex host character set
9913 @cindex target character set
9915 If the program you are debugging uses a different character set to
9916 represent characters and strings than the one @value{GDBN} uses itself,
9917 @value{GDBN} can automatically translate between the character sets for
9918 you. The character set @value{GDBN} uses we call the @dfn{host
9919 character set}; the one the inferior program uses we call the
9920 @dfn{target character set}.
9922 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
9923 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
9924 remote protocol (@pxref{Remote Debugging}) to debug a program
9925 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
9926 then the host character set is Latin-1, and the target character set is
9927 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
9928 target-charset EBCDIC-US}, then @value{GDBN} translates between
9929 @sc{ebcdic} and Latin 1 as you print character or string values, or use
9930 character and string literals in expressions.
9932 @value{GDBN} has no way to automatically recognize which character set
9933 the inferior program uses; you must tell it, using the @code{set
9934 target-charset} command, described below.
9936 Here are the commands for controlling @value{GDBN}'s character set
9940 @item set target-charset @var{charset}
9941 @kindex set target-charset
9942 Set the current target character set to @var{charset}. To display the
9943 list of supported target character sets, type
9944 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
9946 @item set host-charset @var{charset}
9947 @kindex set host-charset
9948 Set the current host character set to @var{charset}.
9950 By default, @value{GDBN} uses a host character set appropriate to the
9951 system it is running on; you can override that default using the
9952 @code{set host-charset} command. On some systems, @value{GDBN} cannot
9953 automatically determine the appropriate host character set. In this
9954 case, @value{GDBN} uses @samp{UTF-8}.
9956 @value{GDBN} can only use certain character sets as its host character
9957 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
9958 @value{GDBN} will list the host character sets it supports.
9960 @item set charset @var{charset}
9962 Set the current host and target character sets to @var{charset}. As
9963 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
9964 @value{GDBN} will list the names of the character sets that can be used
9965 for both host and target.
9968 @kindex show charset
9969 Show the names of the current host and target character sets.
9971 @item show host-charset
9972 @kindex show host-charset
9973 Show the name of the current host character set.
9975 @item show target-charset
9976 @kindex show target-charset
9977 Show the name of the current target character set.
9979 @item set target-wide-charset @var{charset}
9980 @kindex set target-wide-charset
9981 Set the current target's wide character set to @var{charset}. This is
9982 the character set used by the target's @code{wchar_t} type. To
9983 display the list of supported wide character sets, type
9984 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
9986 @item show target-wide-charset
9987 @kindex show target-wide-charset
9988 Show the name of the current target's wide character set.
9991 Here is an example of @value{GDBN}'s character set support in action.
9992 Assume that the following source code has been placed in the file
9993 @file{charset-test.c}:
9999 = @{72, 101, 108, 108, 111, 44, 32, 119,
10000 111, 114, 108, 100, 33, 10, 0@};
10001 char ibm1047_hello[]
10002 = @{200, 133, 147, 147, 150, 107, 64, 166,
10003 150, 153, 147, 132, 90, 37, 0@};
10007 printf ("Hello, world!\n");
10011 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
10012 containing the string @samp{Hello, world!} followed by a newline,
10013 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
10015 We compile the program, and invoke the debugger on it:
10018 $ gcc -g charset-test.c -o charset-test
10019 $ gdb -nw charset-test
10020 GNU gdb 2001-12-19-cvs
10021 Copyright 2001 Free Software Foundation, Inc.
10026 We can use the @code{show charset} command to see what character sets
10027 @value{GDBN} is currently using to interpret and display characters and
10031 (@value{GDBP}) show charset
10032 The current host and target character set is `ISO-8859-1'.
10036 For the sake of printing this manual, let's use @sc{ascii} as our
10037 initial character set:
10039 (@value{GDBP}) set charset ASCII
10040 (@value{GDBP}) show charset
10041 The current host and target character set is `ASCII'.
10045 Let's assume that @sc{ascii} is indeed the correct character set for our
10046 host system --- in other words, let's assume that if @value{GDBN} prints
10047 characters using the @sc{ascii} character set, our terminal will display
10048 them properly. Since our current target character set is also
10049 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
10052 (@value{GDBP}) print ascii_hello
10053 $1 = 0x401698 "Hello, world!\n"
10054 (@value{GDBP}) print ascii_hello[0]
10059 @value{GDBN} uses the target character set for character and string
10060 literals you use in expressions:
10063 (@value{GDBP}) print '+'
10068 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
10071 @value{GDBN} relies on the user to tell it which character set the
10072 target program uses. If we print @code{ibm1047_hello} while our target
10073 character set is still @sc{ascii}, we get jibberish:
10076 (@value{GDBP}) print ibm1047_hello
10077 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
10078 (@value{GDBP}) print ibm1047_hello[0]
10083 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
10084 @value{GDBN} tells us the character sets it supports:
10087 (@value{GDBP}) set target-charset
10088 ASCII EBCDIC-US IBM1047 ISO-8859-1
10089 (@value{GDBP}) set target-charset
10092 We can select @sc{ibm1047} as our target character set, and examine the
10093 program's strings again. Now the @sc{ascii} string is wrong, but
10094 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
10095 target character set, @sc{ibm1047}, to the host character set,
10096 @sc{ascii}, and they display correctly:
10099 (@value{GDBP}) set target-charset IBM1047
10100 (@value{GDBP}) show charset
10101 The current host character set is `ASCII'.
10102 The current target character set is `IBM1047'.
10103 (@value{GDBP}) print ascii_hello
10104 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
10105 (@value{GDBP}) print ascii_hello[0]
10107 (@value{GDBP}) print ibm1047_hello
10108 $8 = 0x4016a8 "Hello, world!\n"
10109 (@value{GDBP}) print ibm1047_hello[0]
10114 As above, @value{GDBN} uses the target character set for character and
10115 string literals you use in expressions:
10118 (@value{GDBP}) print '+'
10123 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
10126 @node Caching Remote Data
10127 @section Caching Data of Remote Targets
10128 @cindex caching data of remote targets
10130 @value{GDBN} caches data exchanged between the debugger and a
10131 remote target (@pxref{Remote Debugging}). Such caching generally improves
10132 performance, because it reduces the overhead of the remote protocol by
10133 bundling memory reads and writes into large chunks. Unfortunately, simply
10134 caching everything would lead to incorrect results, since @value{GDBN}
10135 does not necessarily know anything about volatile values, memory-mapped I/O
10136 addresses, etc. Furthermore, in non-stop mode (@pxref{Non-Stop Mode})
10137 memory can be changed @emph{while} a gdb command is executing.
10138 Therefore, by default, @value{GDBN} only caches data
10139 known to be on the stack@footnote{In non-stop mode, it is moderately
10140 rare for a running thread to modify the stack of a stopped thread
10141 in a way that would interfere with a backtrace, and caching of
10142 stack reads provides a significant speed up of remote backtraces.}.
10143 Other regions of memory can be explicitly marked as
10144 cacheable; see @pxref{Memory Region Attributes}.
10147 @kindex set remotecache
10148 @item set remotecache on
10149 @itemx set remotecache off
10150 This option no longer does anything; it exists for compatibility
10153 @kindex show remotecache
10154 @item show remotecache
10155 Show the current state of the obsolete remotecache flag.
10157 @kindex set stack-cache
10158 @item set stack-cache on
10159 @itemx set stack-cache off
10160 Enable or disable caching of stack accesses. When @code{ON}, use
10161 caching. By default, this option is @code{ON}.
10163 @kindex show stack-cache
10164 @item show stack-cache
10165 Show the current state of data caching for memory accesses.
10167 @kindex info dcache
10168 @item info dcache @r{[}line@r{]}
10169 Print the information about the data cache performance. The
10170 information displayed includes the dcache width and depth, and for
10171 each cache line, its number, address, and how many times it was
10172 referenced. This command is useful for debugging the data cache
10175 If a line number is specified, the contents of that line will be
10178 @item set dcache size @var{size}
10179 @cindex dcache size
10180 @kindex set dcache size
10181 Set maximum number of entries in dcache (dcache depth above).
10183 @item set dcache line-size @var{line-size}
10184 @cindex dcache line-size
10185 @kindex set dcache line-size
10186 Set number of bytes each dcache entry caches (dcache width above).
10187 Must be a power of 2.
10189 @item show dcache size
10190 @kindex show dcache size
10191 Show maximum number of dcache entries. See also @ref{Caching Remote Data, info dcache}.
10193 @item show dcache line-size
10194 @kindex show dcache line-size
10195 Show default size of dcache lines. See also @ref{Caching Remote Data, info dcache}.
10199 @node Searching Memory
10200 @section Search Memory
10201 @cindex searching memory
10203 Memory can be searched for a particular sequence of bytes with the
10204 @code{find} command.
10208 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
10209 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
10210 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
10211 etc. The search begins at address @var{start_addr} and continues for either
10212 @var{len} bytes or through to @var{end_addr} inclusive.
10215 @var{s} and @var{n} are optional parameters.
10216 They may be specified in either order, apart or together.
10219 @item @var{s}, search query size
10220 The size of each search query value.
10226 halfwords (two bytes)
10230 giant words (eight bytes)
10233 All values are interpreted in the current language.
10234 This means, for example, that if the current source language is C/C@t{++}
10235 then searching for the string ``hello'' includes the trailing '\0'.
10237 If the value size is not specified, it is taken from the
10238 value's type in the current language.
10239 This is useful when one wants to specify the search
10240 pattern as a mixture of types.
10241 Note that this means, for example, that in the case of C-like languages
10242 a search for an untyped 0x42 will search for @samp{(int) 0x42}
10243 which is typically four bytes.
10245 @item @var{n}, maximum number of finds
10246 The maximum number of matches to print. The default is to print all finds.
10249 You can use strings as search values. Quote them with double-quotes
10251 The string value is copied into the search pattern byte by byte,
10252 regardless of the endianness of the target and the size specification.
10254 The address of each match found is printed as well as a count of the
10255 number of matches found.
10257 The address of the last value found is stored in convenience variable
10259 A count of the number of matches is stored in @samp{$numfound}.
10261 For example, if stopped at the @code{printf} in this function:
10267 static char hello[] = "hello-hello";
10268 static struct @{ char c; short s; int i; @}
10269 __attribute__ ((packed)) mixed
10270 = @{ 'c', 0x1234, 0x87654321 @};
10271 printf ("%s\n", hello);
10276 you get during debugging:
10279 (gdb) find &hello[0], +sizeof(hello), "hello"
10280 0x804956d <hello.1620+6>
10282 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
10283 0x8049567 <hello.1620>
10284 0x804956d <hello.1620+6>
10286 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
10287 0x8049567 <hello.1620>
10289 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
10290 0x8049560 <mixed.1625>
10292 (gdb) print $numfound
10295 $2 = (void *) 0x8049560
10298 @node Optimized Code
10299 @chapter Debugging Optimized Code
10300 @cindex optimized code, debugging
10301 @cindex debugging optimized code
10303 Almost all compilers support optimization. With optimization
10304 disabled, the compiler generates assembly code that corresponds
10305 directly to your source code, in a simplistic way. As the compiler
10306 applies more powerful optimizations, the generated assembly code
10307 diverges from your original source code. With help from debugging
10308 information generated by the compiler, @value{GDBN} can map from
10309 the running program back to constructs from your original source.
10311 @value{GDBN} is more accurate with optimization disabled. If you
10312 can recompile without optimization, it is easier to follow the
10313 progress of your program during debugging. But, there are many cases
10314 where you may need to debug an optimized version.
10316 When you debug a program compiled with @samp{-g -O}, remember that the
10317 optimizer has rearranged your code; the debugger shows you what is
10318 really there. Do not be too surprised when the execution path does not
10319 exactly match your source file! An extreme example: if you define a
10320 variable, but never use it, @value{GDBN} never sees that
10321 variable---because the compiler optimizes it out of existence.
10323 Some things do not work as well with @samp{-g -O} as with just
10324 @samp{-g}, particularly on machines with instruction scheduling. If in
10325 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
10326 please report it to us as a bug (including a test case!).
10327 @xref{Variables}, for more information about debugging optimized code.
10330 * Inline Functions:: How @value{GDBN} presents inlining
10331 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
10334 @node Inline Functions
10335 @section Inline Functions
10336 @cindex inline functions, debugging
10338 @dfn{Inlining} is an optimization that inserts a copy of the function
10339 body directly at each call site, instead of jumping to a shared
10340 routine. @value{GDBN} displays inlined functions just like
10341 non-inlined functions. They appear in backtraces. You can view their
10342 arguments and local variables, step into them with @code{step}, skip
10343 them with @code{next}, and escape from them with @code{finish}.
10344 You can check whether a function was inlined by using the
10345 @code{info frame} command.
10347 For @value{GDBN} to support inlined functions, the compiler must
10348 record information about inlining in the debug information ---
10349 @value{NGCC} using the @sc{dwarf 2} format does this, and several
10350 other compilers do also. @value{GDBN} only supports inlined functions
10351 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
10352 do not emit two required attributes (@samp{DW_AT_call_file} and
10353 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
10354 function calls with earlier versions of @value{NGCC}. It instead
10355 displays the arguments and local variables of inlined functions as
10356 local variables in the caller.
10358 The body of an inlined function is directly included at its call site;
10359 unlike a non-inlined function, there are no instructions devoted to
10360 the call. @value{GDBN} still pretends that the call site and the
10361 start of the inlined function are different instructions. Stepping to
10362 the call site shows the call site, and then stepping again shows
10363 the first line of the inlined function, even though no additional
10364 instructions are executed.
10366 This makes source-level debugging much clearer; you can see both the
10367 context of the call and then the effect of the call. Only stepping by
10368 a single instruction using @code{stepi} or @code{nexti} does not do
10369 this; single instruction steps always show the inlined body.
10371 There are some ways that @value{GDBN} does not pretend that inlined
10372 function calls are the same as normal calls:
10376 Setting breakpoints at the call site of an inlined function may not
10377 work, because the call site does not contain any code. @value{GDBN}
10378 may incorrectly move the breakpoint to the next line of the enclosing
10379 function, after the call. This limitation will be removed in a future
10380 version of @value{GDBN}; until then, set a breakpoint on an earlier line
10381 or inside the inlined function instead.
10384 @value{GDBN} cannot locate the return value of inlined calls after
10385 using the @code{finish} command. This is a limitation of compiler-generated
10386 debugging information; after @code{finish}, you can step to the next line
10387 and print a variable where your program stored the return value.
10391 @node Tail Call Frames
10392 @section Tail Call Frames
10393 @cindex tail call frames, debugging
10395 Function @code{B} can call function @code{C} in its very last statement. In
10396 unoptimized compilation the call of @code{C} is immediately followed by return
10397 instruction at the end of @code{B} code. Optimizing compiler may replace the
10398 call and return in function @code{B} into one jump to function @code{C}
10399 instead. Such use of a jump instruction is called @dfn{tail call}.
10401 During execution of function @code{C}, there will be no indication in the
10402 function call stack frames that it was tail-called from @code{B}. If function
10403 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
10404 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
10405 some cases @value{GDBN} can determine that @code{C} was tail-called from
10406 @code{B}, and it will then create fictitious call frame for that, with the
10407 return address set up as if @code{B} called @code{C} normally.
10409 This functionality is currently supported only by DWARF 2 debugging format and
10410 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
10411 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
10414 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
10415 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
10419 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
10421 Stack level 1, frame at 0x7fffffffda30:
10422 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
10423 tail call frame, caller of frame at 0x7fffffffda30
10424 source language c++.
10425 Arglist at unknown address.
10426 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
10429 The detection of all the possible code path executions can find them ambiguous.
10430 There is no execution history stored (possible @ref{Reverse Execution} is never
10431 used for this purpose) and the last known caller could have reached the known
10432 callee by multiple different jump sequences. In such case @value{GDBN} still
10433 tries to show at least all the unambiguous top tail callers and all the
10434 unambiguous bottom tail calees, if any.
10437 @anchor{set debug entry-values}
10438 @item set debug entry-values
10439 @kindex set debug entry-values
10440 When set to on, enables printing of analysis messages for both frame argument
10441 values at function entry and tail calls. It will show all the possible valid
10442 tail calls code paths it has considered. It will also print the intersection
10443 of them with the final unambiguous (possibly partial or even empty) code path
10446 @item show debug entry-values
10447 @kindex show debug entry-values
10448 Show the current state of analysis messages printing for both frame argument
10449 values at function entry and tail calls.
10452 The analysis messages for tail calls can for example show why the virtual tail
10453 call frame for function @code{c} has not been recognized (due to the indirect
10454 reference by variable @code{x}):
10457 static void __attribute__((noinline, noclone)) c (void);
10458 void (*x) (void) = c;
10459 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
10460 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
10461 int main (void) @{ x (); return 0; @}
10463 Breakpoint 1, DW_OP_GNU_entry_value resolving cannot find
10464 DW_TAG_GNU_call_site 0x40039a in main
10466 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
10469 #1 0x000000000040039a in main () at t.c:5
10472 Another possibility is an ambiguous virtual tail call frames resolution:
10476 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
10477 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
10478 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
10479 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
10480 static void __attribute__((noinline, noclone)) b (void)
10481 @{ if (i) c (); else e (); @}
10482 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
10483 int main (void) @{ a (); return 0; @}
10485 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
10486 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
10487 tailcall: reduced: 0x4004d2(a) |
10490 #1 0x00000000004004d2 in a () at t.c:8
10491 #2 0x0000000000400395 in main () at t.c:9
10494 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
10495 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
10497 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
10498 @ifset HAVE_MAKEINFO_CLICK
10499 @set ARROW @click{}
10500 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
10501 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
10503 @ifclear HAVE_MAKEINFO_CLICK
10505 @set CALLSEQ1B @value{CALLSEQ1A}
10506 @set CALLSEQ2B @value{CALLSEQ2A}
10509 Frames #0 and #2 are real, #1 is a virtual tail call frame.
10510 The code can have possible execution paths @value{CALLSEQ1B} or
10511 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
10513 @code{initial:} state shows some random possible calling sequence @value{GDBN}
10514 has found. It then finds another possible calling sequcen - that one is
10515 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
10516 printed as the @code{reduced:} calling sequence. That one could have many
10517 futher @code{compare:} and @code{reduced:} statements as long as there remain
10518 any non-ambiguous sequence entries.
10520 For the frame of function @code{b} in both cases there are different possible
10521 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
10522 also ambigous. The only non-ambiguous frame is the one for function @code{a},
10523 therefore this one is displayed to the user while the ambiguous frames are
10526 There can be also reasons why printing of frame argument values at function
10531 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
10532 static void __attribute__((noinline, noclone)) a (int i);
10533 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
10534 static void __attribute__((noinline, noclone)) a (int i)
10535 @{ if (i) b (i - 1); else c (0); @}
10536 int main (void) @{ a (5); return 0; @}
10539 #0 c (i=i@@entry=0) at t.c:2
10540 #1 0x0000000000400428 in a (DW_OP_GNU_entry_value resolving has found
10541 function "a" at 0x400420 can call itself via tail calls
10542 i=<optimized out>) at t.c:6
10543 #2 0x000000000040036e in main () at t.c:7
10546 @value{GDBN} cannot find out from the inferior state if and how many times did
10547 function @code{a} call itself (via function @code{b}) as these calls would be
10548 tail calls. Such tail calls would modify thue @code{i} variable, therefore
10549 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
10550 prints @code{<optimized out>} instead.
10553 @chapter C Preprocessor Macros
10555 Some languages, such as C and C@t{++}, provide a way to define and invoke
10556 ``preprocessor macros'' which expand into strings of tokens.
10557 @value{GDBN} can evaluate expressions containing macro invocations, show
10558 the result of macro expansion, and show a macro's definition, including
10559 where it was defined.
10561 You may need to compile your program specially to provide @value{GDBN}
10562 with information about preprocessor macros. Most compilers do not
10563 include macros in their debugging information, even when you compile
10564 with the @option{-g} flag. @xref{Compilation}.
10566 A program may define a macro at one point, remove that definition later,
10567 and then provide a different definition after that. Thus, at different
10568 points in the program, a macro may have different definitions, or have
10569 no definition at all. If there is a current stack frame, @value{GDBN}
10570 uses the macros in scope at that frame's source code line. Otherwise,
10571 @value{GDBN} uses the macros in scope at the current listing location;
10574 Whenever @value{GDBN} evaluates an expression, it always expands any
10575 macro invocations present in the expression. @value{GDBN} also provides
10576 the following commands for working with macros explicitly.
10580 @kindex macro expand
10581 @cindex macro expansion, showing the results of preprocessor
10582 @cindex preprocessor macro expansion, showing the results of
10583 @cindex expanding preprocessor macros
10584 @item macro expand @var{expression}
10585 @itemx macro exp @var{expression}
10586 Show the results of expanding all preprocessor macro invocations in
10587 @var{expression}. Since @value{GDBN} simply expands macros, but does
10588 not parse the result, @var{expression} need not be a valid expression;
10589 it can be any string of tokens.
10592 @item macro expand-once @var{expression}
10593 @itemx macro exp1 @var{expression}
10594 @cindex expand macro once
10595 @i{(This command is not yet implemented.)} Show the results of
10596 expanding those preprocessor macro invocations that appear explicitly in
10597 @var{expression}. Macro invocations appearing in that expansion are
10598 left unchanged. This command allows you to see the effect of a
10599 particular macro more clearly, without being confused by further
10600 expansions. Since @value{GDBN} simply expands macros, but does not
10601 parse the result, @var{expression} need not be a valid expression; it
10602 can be any string of tokens.
10605 @cindex macro definition, showing
10606 @cindex definition of a macro, showing
10607 @cindex macros, from debug info
10608 @item info macro [-a|-all] [--] @var{macro}
10609 Show the current definition or all definitions of the named @var{macro},
10610 and describe the source location or compiler command-line where that
10611 definition was established. The optional double dash is to signify the end of
10612 argument processing and the beginning of @var{macro} for non C-like macros where
10613 the macro may begin with a hyphen.
10615 @kindex info macros
10616 @item info macros @var{linespec}
10617 Show all macro definitions that are in effect at the location specified
10618 by @var{linespec}, and describe the source location or compiler
10619 command-line where those definitions were established.
10621 @kindex macro define
10622 @cindex user-defined macros
10623 @cindex defining macros interactively
10624 @cindex macros, user-defined
10625 @item macro define @var{macro} @var{replacement-list}
10626 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
10627 Introduce a definition for a preprocessor macro named @var{macro},
10628 invocations of which are replaced by the tokens given in
10629 @var{replacement-list}. The first form of this command defines an
10630 ``object-like'' macro, which takes no arguments; the second form
10631 defines a ``function-like'' macro, which takes the arguments given in
10634 A definition introduced by this command is in scope in every
10635 expression evaluated in @value{GDBN}, until it is removed with the
10636 @code{macro undef} command, described below. The definition overrides
10637 all definitions for @var{macro} present in the program being debugged,
10638 as well as any previous user-supplied definition.
10640 @kindex macro undef
10641 @item macro undef @var{macro}
10642 Remove any user-supplied definition for the macro named @var{macro}.
10643 This command only affects definitions provided with the @code{macro
10644 define} command, described above; it cannot remove definitions present
10645 in the program being debugged.
10649 List all the macros defined using the @code{macro define} command.
10652 @cindex macros, example of debugging with
10653 Here is a transcript showing the above commands in action. First, we
10654 show our source files:
10659 #include "sample.h"
10662 #define ADD(x) (M + x)
10667 printf ("Hello, world!\n");
10669 printf ("We're so creative.\n");
10671 printf ("Goodbye, world!\n");
10678 Now, we compile the program using the @sc{gnu} C compiler,
10679 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
10680 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
10681 and @option{-gdwarf-4}; we recommend always choosing the most recent
10682 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
10683 includes information about preprocessor macros in the debugging
10687 $ gcc -gdwarf-2 -g3 sample.c -o sample
10691 Now, we start @value{GDBN} on our sample program:
10695 GNU gdb 2002-05-06-cvs
10696 Copyright 2002 Free Software Foundation, Inc.
10697 GDB is free software, @dots{}
10701 We can expand macros and examine their definitions, even when the
10702 program is not running. @value{GDBN} uses the current listing position
10703 to decide which macro definitions are in scope:
10706 (@value{GDBP}) list main
10709 5 #define ADD(x) (M + x)
10714 10 printf ("Hello, world!\n");
10716 12 printf ("We're so creative.\n");
10717 (@value{GDBP}) info macro ADD
10718 Defined at /home/jimb/gdb/macros/play/sample.c:5
10719 #define ADD(x) (M + x)
10720 (@value{GDBP}) info macro Q
10721 Defined at /home/jimb/gdb/macros/play/sample.h:1
10722 included at /home/jimb/gdb/macros/play/sample.c:2
10724 (@value{GDBP}) macro expand ADD(1)
10725 expands to: (42 + 1)
10726 (@value{GDBP}) macro expand-once ADD(1)
10727 expands to: once (M + 1)
10731 In the example above, note that @code{macro expand-once} expands only
10732 the macro invocation explicit in the original text --- the invocation of
10733 @code{ADD} --- but does not expand the invocation of the macro @code{M},
10734 which was introduced by @code{ADD}.
10736 Once the program is running, @value{GDBN} uses the macro definitions in
10737 force at the source line of the current stack frame:
10740 (@value{GDBP}) break main
10741 Breakpoint 1 at 0x8048370: file sample.c, line 10.
10743 Starting program: /home/jimb/gdb/macros/play/sample
10745 Breakpoint 1, main () at sample.c:10
10746 10 printf ("Hello, world!\n");
10750 At line 10, the definition of the macro @code{N} at line 9 is in force:
10753 (@value{GDBP}) info macro N
10754 Defined at /home/jimb/gdb/macros/play/sample.c:9
10756 (@value{GDBP}) macro expand N Q M
10757 expands to: 28 < 42
10758 (@value{GDBP}) print N Q M
10763 As we step over directives that remove @code{N}'s definition, and then
10764 give it a new definition, @value{GDBN} finds the definition (or lack
10765 thereof) in force at each point:
10768 (@value{GDBP}) next
10770 12 printf ("We're so creative.\n");
10771 (@value{GDBP}) info macro N
10772 The symbol `N' has no definition as a C/C++ preprocessor macro
10773 at /home/jimb/gdb/macros/play/sample.c:12
10774 (@value{GDBP}) next
10776 14 printf ("Goodbye, world!\n");
10777 (@value{GDBP}) info macro N
10778 Defined at /home/jimb/gdb/macros/play/sample.c:13
10780 (@value{GDBP}) macro expand N Q M
10781 expands to: 1729 < 42
10782 (@value{GDBP}) print N Q M
10787 In addition to source files, macros can be defined on the compilation command
10788 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
10789 such a way, @value{GDBN} displays the location of their definition as line zero
10790 of the source file submitted to the compiler.
10793 (@value{GDBP}) info macro __STDC__
10794 Defined at /home/jimb/gdb/macros/play/sample.c:0
10801 @chapter Tracepoints
10802 @c This chapter is based on the documentation written by Michael
10803 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
10805 @cindex tracepoints
10806 In some applications, it is not feasible for the debugger to interrupt
10807 the program's execution long enough for the developer to learn
10808 anything helpful about its behavior. If the program's correctness
10809 depends on its real-time behavior, delays introduced by a debugger
10810 might cause the program to change its behavior drastically, or perhaps
10811 fail, even when the code itself is correct. It is useful to be able
10812 to observe the program's behavior without interrupting it.
10814 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
10815 specify locations in the program, called @dfn{tracepoints}, and
10816 arbitrary expressions to evaluate when those tracepoints are reached.
10817 Later, using the @code{tfind} command, you can examine the values
10818 those expressions had when the program hit the tracepoints. The
10819 expressions may also denote objects in memory---structures or arrays,
10820 for example---whose values @value{GDBN} should record; while visiting
10821 a particular tracepoint, you may inspect those objects as if they were
10822 in memory at that moment. However, because @value{GDBN} records these
10823 values without interacting with you, it can do so quickly and
10824 unobtrusively, hopefully not disturbing the program's behavior.
10826 The tracepoint facility is currently available only for remote
10827 targets. @xref{Targets}. In addition, your remote target must know
10828 how to collect trace data. This functionality is implemented in the
10829 remote stub; however, none of the stubs distributed with @value{GDBN}
10830 support tracepoints as of this writing. The format of the remote
10831 packets used to implement tracepoints are described in @ref{Tracepoint
10834 It is also possible to get trace data from a file, in a manner reminiscent
10835 of corefiles; you specify the filename, and use @code{tfind} to search
10836 through the file. @xref{Trace Files}, for more details.
10838 This chapter describes the tracepoint commands and features.
10841 * Set Tracepoints::
10842 * Analyze Collected Data::
10843 * Tracepoint Variables::
10847 @node Set Tracepoints
10848 @section Commands to Set Tracepoints
10850 Before running such a @dfn{trace experiment}, an arbitrary number of
10851 tracepoints can be set. A tracepoint is actually a special type of
10852 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
10853 standard breakpoint commands. For instance, as with breakpoints,
10854 tracepoint numbers are successive integers starting from one, and many
10855 of the commands associated with tracepoints take the tracepoint number
10856 as their argument, to identify which tracepoint to work on.
10858 For each tracepoint, you can specify, in advance, some arbitrary set
10859 of data that you want the target to collect in the trace buffer when
10860 it hits that tracepoint. The collected data can include registers,
10861 local variables, or global data. Later, you can use @value{GDBN}
10862 commands to examine the values these data had at the time the
10863 tracepoint was hit.
10865 Tracepoints do not support every breakpoint feature. Ignore counts on
10866 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
10867 commands when they are hit. Tracepoints may not be thread-specific
10870 @cindex fast tracepoints
10871 Some targets may support @dfn{fast tracepoints}, which are inserted in
10872 a different way (such as with a jump instead of a trap), that is
10873 faster but possibly restricted in where they may be installed.
10875 @cindex static tracepoints
10876 @cindex markers, static tracepoints
10877 @cindex probing markers, static tracepoints
10878 Regular and fast tracepoints are dynamic tracing facilities, meaning
10879 that they can be used to insert tracepoints at (almost) any location
10880 in the target. Some targets may also support controlling @dfn{static
10881 tracepoints} from @value{GDBN}. With static tracing, a set of
10882 instrumentation points, also known as @dfn{markers}, are embedded in
10883 the target program, and can be activated or deactivated by name or
10884 address. These are usually placed at locations which facilitate
10885 investigating what the target is actually doing. @value{GDBN}'s
10886 support for static tracing includes being able to list instrumentation
10887 points, and attach them with @value{GDBN} defined high level
10888 tracepoints that expose the whole range of convenience of
10889 @value{GDBN}'s tracepoints support. Namely, support for collecting
10890 registers values and values of global or local (to the instrumentation
10891 point) variables; tracepoint conditions and trace state variables.
10892 The act of installing a @value{GDBN} static tracepoint on an
10893 instrumentation point, or marker, is referred to as @dfn{probing} a
10894 static tracepoint marker.
10896 @code{gdbserver} supports tracepoints on some target systems.
10897 @xref{Server,,Tracepoints support in @code{gdbserver}}.
10899 This section describes commands to set tracepoints and associated
10900 conditions and actions.
10903 * Create and Delete Tracepoints::
10904 * Enable and Disable Tracepoints::
10905 * Tracepoint Passcounts::
10906 * Tracepoint Conditions::
10907 * Trace State Variables::
10908 * Tracepoint Actions::
10909 * Listing Tracepoints::
10910 * Listing Static Tracepoint Markers::
10911 * Starting and Stopping Trace Experiments::
10912 * Tracepoint Restrictions::
10915 @node Create and Delete Tracepoints
10916 @subsection Create and Delete Tracepoints
10919 @cindex set tracepoint
10921 @item trace @var{location}
10922 The @code{trace} command is very similar to the @code{break} command.
10923 Its argument @var{location} can be a source line, a function name, or
10924 an address in the target program. @xref{Specify Location}. The
10925 @code{trace} command defines a tracepoint, which is a point in the
10926 target program where the debugger will briefly stop, collect some
10927 data, and then allow the program to continue. Setting a tracepoint or
10928 changing its actions takes effect immediately if the remote stub
10929 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
10931 If remote stub doesn't support the @samp{InstallInTrace} feature, all
10932 these changes don't take effect until the next @code{tstart}
10933 command, and once a trace experiment is running, further changes will
10934 not have any effect until the next trace experiment starts. In addition,
10935 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
10936 address is not yet resolved. (This is similar to pending breakpoints.)
10937 Pending tracepoints are not downloaded to the target and not installed
10938 until they are resolved. The resolution of pending tracepoints requires
10939 @value{GDBN} support---when debugging with the remote target, and
10940 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
10941 tracing}), pending tracepoints can not be resolved (and downloaded to
10942 the remote stub) while @value{GDBN} is disconnected.
10944 Here are some examples of using the @code{trace} command:
10947 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
10949 (@value{GDBP}) @b{trace +2} // 2 lines forward
10951 (@value{GDBP}) @b{trace my_function} // first source line of function
10953 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
10955 (@value{GDBP}) @b{trace *0x2117c4} // an address
10959 You can abbreviate @code{trace} as @code{tr}.
10961 @item trace @var{location} if @var{cond}
10962 Set a tracepoint with condition @var{cond}; evaluate the expression
10963 @var{cond} each time the tracepoint is reached, and collect data only
10964 if the value is nonzero---that is, if @var{cond} evaluates as true.
10965 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
10966 information on tracepoint conditions.
10968 @item ftrace @var{location} [ if @var{cond} ]
10969 @cindex set fast tracepoint
10970 @cindex fast tracepoints, setting
10972 The @code{ftrace} command sets a fast tracepoint. For targets that
10973 support them, fast tracepoints will use a more efficient but possibly
10974 less general technique to trigger data collection, such as a jump
10975 instruction instead of a trap, or some sort of hardware support. It
10976 may not be possible to create a fast tracepoint at the desired
10977 location, in which case the command will exit with an explanatory
10980 @value{GDBN} handles arguments to @code{ftrace} exactly as for
10983 On 32-bit x86-architecture systems, fast tracepoints normally need to
10984 be placed at an instruction that is 5 bytes or longer, but can be
10985 placed at 4-byte instructions if the low 64K of memory of the target
10986 program is available to install trampolines. Some Unix-type systems,
10987 such as @sc{gnu}/Linux, exclude low addresses from the program's
10988 address space; but for instance with the Linux kernel it is possible
10989 to let @value{GDBN} use this area by doing a @command{sysctl} command
10990 to set the @code{mmap_min_addr} kernel parameter, as in
10993 sudo sysctl -w vm.mmap_min_addr=32768
10997 which sets the low address to 32K, which leaves plenty of room for
10998 trampolines. The minimum address should be set to a page boundary.
11000 @item strace @var{location} [ if @var{cond} ]
11001 @cindex set static tracepoint
11002 @cindex static tracepoints, setting
11003 @cindex probe static tracepoint marker
11005 The @code{strace} command sets a static tracepoint. For targets that
11006 support it, setting a static tracepoint probes a static
11007 instrumentation point, or marker, found at @var{location}. It may not
11008 be possible to set a static tracepoint at the desired location, in
11009 which case the command will exit with an explanatory message.
11011 @value{GDBN} handles arguments to @code{strace} exactly as for
11012 @code{trace}, with the addition that the user can also specify
11013 @code{-m @var{marker}} as @var{location}. This probes the marker
11014 identified by the @var{marker} string identifier. This identifier
11015 depends on the static tracepoint backend library your program is
11016 using. You can find all the marker identifiers in the @samp{ID} field
11017 of the @code{info static-tracepoint-markers} command output.
11018 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
11019 Markers}. For example, in the following small program using the UST
11025 trace_mark(ust, bar33, "str %s", "FOOBAZ");
11030 the marker id is composed of joining the first two arguments to the
11031 @code{trace_mark} call with a slash, which translates to:
11034 (@value{GDBP}) info static-tracepoint-markers
11035 Cnt Enb ID Address What
11036 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
11042 so you may probe the marker above with:
11045 (@value{GDBP}) strace -m ust/bar33
11048 Static tracepoints accept an extra collect action --- @code{collect
11049 $_sdata}. This collects arbitrary user data passed in the probe point
11050 call to the tracing library. In the UST example above, you'll see
11051 that the third argument to @code{trace_mark} is a printf-like format
11052 string. The user data is then the result of running that formating
11053 string against the following arguments. Note that @code{info
11054 static-tracepoint-markers} command output lists that format string in
11055 the @samp{Data:} field.
11057 You can inspect this data when analyzing the trace buffer, by printing
11058 the $_sdata variable like any other variable available to
11059 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
11062 @cindex last tracepoint number
11063 @cindex recent tracepoint number
11064 @cindex tracepoint number
11065 The convenience variable @code{$tpnum} records the tracepoint number
11066 of the most recently set tracepoint.
11068 @kindex delete tracepoint
11069 @cindex tracepoint deletion
11070 @item delete tracepoint @r{[}@var{num}@r{]}
11071 Permanently delete one or more tracepoints. With no argument, the
11072 default is to delete all tracepoints. Note that the regular
11073 @code{delete} command can remove tracepoints also.
11078 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
11080 (@value{GDBP}) @b{delete trace} // remove all tracepoints
11084 You can abbreviate this command as @code{del tr}.
11087 @node Enable and Disable Tracepoints
11088 @subsection Enable and Disable Tracepoints
11090 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
11093 @kindex disable tracepoint
11094 @item disable tracepoint @r{[}@var{num}@r{]}
11095 Disable tracepoint @var{num}, or all tracepoints if no argument
11096 @var{num} is given. A disabled tracepoint will have no effect during
11097 a trace experiment, but it is not forgotten. You can re-enable
11098 a disabled tracepoint using the @code{enable tracepoint} command.
11099 If the command is issued during a trace experiment and the debug target
11100 has support for disabling tracepoints during a trace experiment, then the
11101 change will be effective immediately. Otherwise, it will be applied to the
11102 next trace experiment.
11104 @kindex enable tracepoint
11105 @item enable tracepoint @r{[}@var{num}@r{]}
11106 Enable tracepoint @var{num}, or all tracepoints. If this command is
11107 issued during a trace experiment and the debug target supports enabling
11108 tracepoints during a trace experiment, then the enabled tracepoints will
11109 become effective immediately. Otherwise, they will become effective the
11110 next time a trace experiment is run.
11113 @node Tracepoint Passcounts
11114 @subsection Tracepoint Passcounts
11118 @cindex tracepoint pass count
11119 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
11120 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
11121 automatically stop a trace experiment. If a tracepoint's passcount is
11122 @var{n}, then the trace experiment will be automatically stopped on
11123 the @var{n}'th time that tracepoint is hit. If the tracepoint number
11124 @var{num} is not specified, the @code{passcount} command sets the
11125 passcount of the most recently defined tracepoint. If no passcount is
11126 given, the trace experiment will run until stopped explicitly by the
11132 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
11133 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
11135 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
11136 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
11137 (@value{GDBP}) @b{trace foo}
11138 (@value{GDBP}) @b{pass 3}
11139 (@value{GDBP}) @b{trace bar}
11140 (@value{GDBP}) @b{pass 2}
11141 (@value{GDBP}) @b{trace baz}
11142 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
11143 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
11144 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
11145 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
11149 @node Tracepoint Conditions
11150 @subsection Tracepoint Conditions
11151 @cindex conditional tracepoints
11152 @cindex tracepoint conditions
11154 The simplest sort of tracepoint collects data every time your program
11155 reaches a specified place. You can also specify a @dfn{condition} for
11156 a tracepoint. A condition is just a Boolean expression in your
11157 programming language (@pxref{Expressions, ,Expressions}). A
11158 tracepoint with a condition evaluates the expression each time your
11159 program reaches it, and data collection happens only if the condition
11162 Tracepoint conditions can be specified when a tracepoint is set, by
11163 using @samp{if} in the arguments to the @code{trace} command.
11164 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
11165 also be set or changed at any time with the @code{condition} command,
11166 just as with breakpoints.
11168 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
11169 the conditional expression itself. Instead, @value{GDBN} encodes the
11170 expression into an agent expression (@pxref{Agent Expressions})
11171 suitable for execution on the target, independently of @value{GDBN}.
11172 Global variables become raw memory locations, locals become stack
11173 accesses, and so forth.
11175 For instance, suppose you have a function that is usually called
11176 frequently, but should not be called after an error has occurred. You
11177 could use the following tracepoint command to collect data about calls
11178 of that function that happen while the error code is propagating
11179 through the program; an unconditional tracepoint could end up
11180 collecting thousands of useless trace frames that you would have to
11184 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
11187 @node Trace State Variables
11188 @subsection Trace State Variables
11189 @cindex trace state variables
11191 A @dfn{trace state variable} is a special type of variable that is
11192 created and managed by target-side code. The syntax is the same as
11193 that for GDB's convenience variables (a string prefixed with ``$''),
11194 but they are stored on the target. They must be created explicitly,
11195 using a @code{tvariable} command. They are always 64-bit signed
11198 Trace state variables are remembered by @value{GDBN}, and downloaded
11199 to the target along with tracepoint information when the trace
11200 experiment starts. There are no intrinsic limits on the number of
11201 trace state variables, beyond memory limitations of the target.
11203 @cindex convenience variables, and trace state variables
11204 Although trace state variables are managed by the target, you can use
11205 them in print commands and expressions as if they were convenience
11206 variables; @value{GDBN} will get the current value from the target
11207 while the trace experiment is running. Trace state variables share
11208 the same namespace as other ``$'' variables, which means that you
11209 cannot have trace state variables with names like @code{$23} or
11210 @code{$pc}, nor can you have a trace state variable and a convenience
11211 variable with the same name.
11215 @item tvariable $@var{name} [ = @var{expression} ]
11217 The @code{tvariable} command creates a new trace state variable named
11218 @code{$@var{name}}, and optionally gives it an initial value of
11219 @var{expression}. @var{expression} is evaluated when this command is
11220 entered; the result will be converted to an integer if possible,
11221 otherwise @value{GDBN} will report an error. A subsequent
11222 @code{tvariable} command specifying the same name does not create a
11223 variable, but instead assigns the supplied initial value to the
11224 existing variable of that name, overwriting any previous initial
11225 value. The default initial value is 0.
11227 @item info tvariables
11228 @kindex info tvariables
11229 List all the trace state variables along with their initial values.
11230 Their current values may also be displayed, if the trace experiment is
11233 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
11234 @kindex delete tvariable
11235 Delete the given trace state variables, or all of them if no arguments
11240 @node Tracepoint Actions
11241 @subsection Tracepoint Action Lists
11245 @cindex tracepoint actions
11246 @item actions @r{[}@var{num}@r{]}
11247 This command will prompt for a list of actions to be taken when the
11248 tracepoint is hit. If the tracepoint number @var{num} is not
11249 specified, this command sets the actions for the one that was most
11250 recently defined (so that you can define a tracepoint and then say
11251 @code{actions} without bothering about its number). You specify the
11252 actions themselves on the following lines, one action at a time, and
11253 terminate the actions list with a line containing just @code{end}. So
11254 far, the only defined actions are @code{collect}, @code{teval}, and
11255 @code{while-stepping}.
11257 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
11258 Commands, ,Breakpoint Command Lists}), except that only the defined
11259 actions are allowed; any other @value{GDBN} command is rejected.
11261 @cindex remove actions from a tracepoint
11262 To remove all actions from a tracepoint, type @samp{actions @var{num}}
11263 and follow it immediately with @samp{end}.
11266 (@value{GDBP}) @b{collect @var{data}} // collect some data
11268 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
11270 (@value{GDBP}) @b{end} // signals the end of actions.
11273 In the following example, the action list begins with @code{collect}
11274 commands indicating the things to be collected when the tracepoint is
11275 hit. Then, in order to single-step and collect additional data
11276 following the tracepoint, a @code{while-stepping} command is used,
11277 followed by the list of things to be collected after each step in a
11278 sequence of single steps. The @code{while-stepping} command is
11279 terminated by its own separate @code{end} command. Lastly, the action
11280 list is terminated by an @code{end} command.
11283 (@value{GDBP}) @b{trace foo}
11284 (@value{GDBP}) @b{actions}
11285 Enter actions for tracepoint 1, one per line:
11288 > while-stepping 12
11289 > collect $pc, arr[i]
11294 @kindex collect @r{(tracepoints)}
11295 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
11296 Collect values of the given expressions when the tracepoint is hit.
11297 This command accepts a comma-separated list of any valid expressions.
11298 In addition to global, static, or local variables, the following
11299 special arguments are supported:
11303 Collect all registers.
11306 Collect all function arguments.
11309 Collect all local variables.
11312 Collect the return address. This is helpful if you want to see more
11316 Collects the number of arguments from the static probe at which the
11317 tracepoint is located.
11318 @xref{Static Probe Points}.
11320 @item $_probe_arg@var{n}
11321 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
11322 from the static probe at which the tracepoint is located.
11323 @xref{Static Probe Points}.
11326 @vindex $_sdata@r{, collect}
11327 Collect static tracepoint marker specific data. Only available for
11328 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
11329 Lists}. On the UST static tracepoints library backend, an
11330 instrumentation point resembles a @code{printf} function call. The
11331 tracing library is able to collect user specified data formatted to a
11332 character string using the format provided by the programmer that
11333 instrumented the program. Other backends have similar mechanisms.
11334 Here's an example of a UST marker call:
11337 const char master_name[] = "$your_name";
11338 trace_mark(channel1, marker1, "hello %s", master_name)
11341 In this case, collecting @code{$_sdata} collects the string
11342 @samp{hello $yourname}. When analyzing the trace buffer, you can
11343 inspect @samp{$_sdata} like any other variable available to
11347 You can give several consecutive @code{collect} commands, each one
11348 with a single argument, or one @code{collect} command with several
11349 arguments separated by commas; the effect is the same.
11351 The optional @var{mods} changes the usual handling of the arguments.
11352 @code{s} requests that pointers to chars be handled as strings, in
11353 particular collecting the contents of the memory being pointed at, up
11354 to the first zero. The upper bound is by default the value of the
11355 @code{print elements} variable; if @code{s} is followed by a decimal
11356 number, that is the upper bound instead. So for instance
11357 @samp{collect/s25 mystr} collects as many as 25 characters at
11360 The command @code{info scope} (@pxref{Symbols, info scope}) is
11361 particularly useful for figuring out what data to collect.
11363 @kindex teval @r{(tracepoints)}
11364 @item teval @var{expr1}, @var{expr2}, @dots{}
11365 Evaluate the given expressions when the tracepoint is hit. This
11366 command accepts a comma-separated list of expressions. The results
11367 are discarded, so this is mainly useful for assigning values to trace
11368 state variables (@pxref{Trace State Variables}) without adding those
11369 values to the trace buffer, as would be the case if the @code{collect}
11372 @kindex while-stepping @r{(tracepoints)}
11373 @item while-stepping @var{n}
11374 Perform @var{n} single-step instruction traces after the tracepoint,
11375 collecting new data after each step. The @code{while-stepping}
11376 command is followed by the list of what to collect while stepping
11377 (followed by its own @code{end} command):
11380 > while-stepping 12
11381 > collect $regs, myglobal
11387 Note that @code{$pc} is not automatically collected by
11388 @code{while-stepping}; you need to explicitly collect that register if
11389 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
11392 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
11393 @kindex set default-collect
11394 @cindex default collection action
11395 This variable is a list of expressions to collect at each tracepoint
11396 hit. It is effectively an additional @code{collect} action prepended
11397 to every tracepoint action list. The expressions are parsed
11398 individually for each tracepoint, so for instance a variable named
11399 @code{xyz} may be interpreted as a global for one tracepoint, and a
11400 local for another, as appropriate to the tracepoint's location.
11402 @item show default-collect
11403 @kindex show default-collect
11404 Show the list of expressions that are collected by default at each
11409 @node Listing Tracepoints
11410 @subsection Listing Tracepoints
11413 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
11414 @kindex info tp @r{[}@var{n}@dots{}@r{]}
11415 @cindex information about tracepoints
11416 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
11417 Display information about the tracepoint @var{num}. If you don't
11418 specify a tracepoint number, displays information about all the
11419 tracepoints defined so far. The format is similar to that used for
11420 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
11421 command, simply restricting itself to tracepoints.
11423 A tracepoint's listing may include additional information specific to
11428 its passcount as given by the @code{passcount @var{n}} command
11432 (@value{GDBP}) @b{info trace}
11433 Num Type Disp Enb Address What
11434 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
11436 collect globfoo, $regs
11445 This command can be abbreviated @code{info tp}.
11448 @node Listing Static Tracepoint Markers
11449 @subsection Listing Static Tracepoint Markers
11452 @kindex info static-tracepoint-markers
11453 @cindex information about static tracepoint markers
11454 @item info static-tracepoint-markers
11455 Display information about all static tracepoint markers defined in the
11458 For each marker, the following columns are printed:
11462 An incrementing counter, output to help readability. This is not a
11465 The marker ID, as reported by the target.
11466 @item Enabled or Disabled
11467 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
11468 that are not enabled.
11470 Where the marker is in your program, as a memory address.
11472 Where the marker is in the source for your program, as a file and line
11473 number. If the debug information included in the program does not
11474 allow @value{GDBN} to locate the source of the marker, this column
11475 will be left blank.
11479 In addition, the following information may be printed for each marker:
11483 User data passed to the tracing library by the marker call. In the
11484 UST backend, this is the format string passed as argument to the
11486 @item Static tracepoints probing the marker
11487 The list of static tracepoints attached to the marker.
11491 (@value{GDBP}) info static-tracepoint-markers
11492 Cnt ID Enb Address What
11493 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
11494 Data: number1 %d number2 %d
11495 Probed by static tracepoints: #2
11496 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
11502 @node Starting and Stopping Trace Experiments
11503 @subsection Starting and Stopping Trace Experiments
11506 @kindex tstart [ @var{notes} ]
11507 @cindex start a new trace experiment
11508 @cindex collected data discarded
11510 This command starts the trace experiment, and begins collecting data.
11511 It has the side effect of discarding all the data collected in the
11512 trace buffer during the previous trace experiment. If any arguments
11513 are supplied, they are taken as a note and stored with the trace
11514 experiment's state. The notes may be arbitrary text, and are
11515 especially useful with disconnected tracing in a multi-user context;
11516 the notes can explain what the trace is doing, supply user contact
11517 information, and so forth.
11519 @kindex tstop [ @var{notes} ]
11520 @cindex stop a running trace experiment
11522 This command stops the trace experiment. If any arguments are
11523 supplied, they are recorded with the experiment as a note. This is
11524 useful if you are stopping a trace started by someone else, for
11525 instance if the trace is interfering with the system's behavior and
11526 needs to be stopped quickly.
11528 @strong{Note}: a trace experiment and data collection may stop
11529 automatically if any tracepoint's passcount is reached
11530 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
11533 @cindex status of trace data collection
11534 @cindex trace experiment, status of
11536 This command displays the status of the current trace data
11540 Here is an example of the commands we described so far:
11543 (@value{GDBP}) @b{trace gdb_c_test}
11544 (@value{GDBP}) @b{actions}
11545 Enter actions for tracepoint #1, one per line.
11546 > collect $regs,$locals,$args
11547 > while-stepping 11
11551 (@value{GDBP}) @b{tstart}
11552 [time passes @dots{}]
11553 (@value{GDBP}) @b{tstop}
11556 @anchor{disconnected tracing}
11557 @cindex disconnected tracing
11558 You can choose to continue running the trace experiment even if
11559 @value{GDBN} disconnects from the target, voluntarily or
11560 involuntarily. For commands such as @code{detach}, the debugger will
11561 ask what you want to do with the trace. But for unexpected
11562 terminations (@value{GDBN} crash, network outage), it would be
11563 unfortunate to lose hard-won trace data, so the variable
11564 @code{disconnected-tracing} lets you decide whether the trace should
11565 continue running without @value{GDBN}.
11568 @item set disconnected-tracing on
11569 @itemx set disconnected-tracing off
11570 @kindex set disconnected-tracing
11571 Choose whether a tracing run should continue to run if @value{GDBN}
11572 has disconnected from the target. Note that @code{detach} or
11573 @code{quit} will ask you directly what to do about a running trace no
11574 matter what this variable's setting, so the variable is mainly useful
11575 for handling unexpected situations, such as loss of the network.
11577 @item show disconnected-tracing
11578 @kindex show disconnected-tracing
11579 Show the current choice for disconnected tracing.
11583 When you reconnect to the target, the trace experiment may or may not
11584 still be running; it might have filled the trace buffer in the
11585 meantime, or stopped for one of the other reasons. If it is running,
11586 it will continue after reconnection.
11588 Upon reconnection, the target will upload information about the
11589 tracepoints in effect. @value{GDBN} will then compare that
11590 information to the set of tracepoints currently defined, and attempt
11591 to match them up, allowing for the possibility that the numbers may
11592 have changed due to creation and deletion in the meantime. If one of
11593 the target's tracepoints does not match any in @value{GDBN}, the
11594 debugger will create a new tracepoint, so that you have a number with
11595 which to specify that tracepoint. This matching-up process is
11596 necessarily heuristic, and it may result in useless tracepoints being
11597 created; you may simply delete them if they are of no use.
11599 @cindex circular trace buffer
11600 If your target agent supports a @dfn{circular trace buffer}, then you
11601 can run a trace experiment indefinitely without filling the trace
11602 buffer; when space runs out, the agent deletes already-collected trace
11603 frames, oldest first, until there is enough room to continue
11604 collecting. This is especially useful if your tracepoints are being
11605 hit too often, and your trace gets terminated prematurely because the
11606 buffer is full. To ask for a circular trace buffer, simply set
11607 @samp{circular-trace-buffer} to on. You can set this at any time,
11608 including during tracing; if the agent can do it, it will change
11609 buffer handling on the fly, otherwise it will not take effect until
11613 @item set circular-trace-buffer on
11614 @itemx set circular-trace-buffer off
11615 @kindex set circular-trace-buffer
11616 Choose whether a tracing run should use a linear or circular buffer
11617 for trace data. A linear buffer will not lose any trace data, but may
11618 fill up prematurely, while a circular buffer will discard old trace
11619 data, but it will have always room for the latest tracepoint hits.
11621 @item show circular-trace-buffer
11622 @kindex show circular-trace-buffer
11623 Show the current choice for the trace buffer. Note that this may not
11624 match the agent's current buffer handling, nor is it guaranteed to
11625 match the setting that might have been in effect during a past run,
11626 for instance if you are looking at frames from a trace file.
11631 @item set trace-user @var{text}
11632 @kindex set trace-user
11634 @item show trace-user
11635 @kindex show trace-user
11637 @item set trace-notes @var{text}
11638 @kindex set trace-notes
11639 Set the trace run's notes.
11641 @item show trace-notes
11642 @kindex show trace-notes
11643 Show the trace run's notes.
11645 @item set trace-stop-notes @var{text}
11646 @kindex set trace-stop-notes
11647 Set the trace run's stop notes. The handling of the note is as for
11648 @code{tstop} arguments; the set command is convenient way to fix a
11649 stop note that is mistaken or incomplete.
11651 @item show trace-stop-notes
11652 @kindex show trace-stop-notes
11653 Show the trace run's stop notes.
11657 @node Tracepoint Restrictions
11658 @subsection Tracepoint Restrictions
11660 @cindex tracepoint restrictions
11661 There are a number of restrictions on the use of tracepoints. As
11662 described above, tracepoint data gathering occurs on the target
11663 without interaction from @value{GDBN}. Thus the full capabilities of
11664 the debugger are not available during data gathering, and then at data
11665 examination time, you will be limited by only having what was
11666 collected. The following items describe some common problems, but it
11667 is not exhaustive, and you may run into additional difficulties not
11673 Tracepoint expressions are intended to gather objects (lvalues). Thus
11674 the full flexibility of GDB's expression evaluator is not available.
11675 You cannot call functions, cast objects to aggregate types, access
11676 convenience variables or modify values (except by assignment to trace
11677 state variables). Some language features may implicitly call
11678 functions (for instance Objective-C fields with accessors), and therefore
11679 cannot be collected either.
11682 Collection of local variables, either individually or in bulk with
11683 @code{$locals} or @code{$args}, during @code{while-stepping} may
11684 behave erratically. The stepping action may enter a new scope (for
11685 instance by stepping into a function), or the location of the variable
11686 may change (for instance it is loaded into a register). The
11687 tracepoint data recorded uses the location information for the
11688 variables that is correct for the tracepoint location. When the
11689 tracepoint is created, it is not possible, in general, to determine
11690 where the steps of a @code{while-stepping} sequence will advance the
11691 program---particularly if a conditional branch is stepped.
11694 Collection of an incompletely-initialized or partially-destroyed object
11695 may result in something that @value{GDBN} cannot display, or displays
11696 in a misleading way.
11699 When @value{GDBN} displays a pointer to character it automatically
11700 dereferences the pointer to also display characters of the string
11701 being pointed to. However, collecting the pointer during tracing does
11702 not automatically collect the string. You need to explicitly
11703 dereference the pointer and provide size information if you want to
11704 collect not only the pointer, but the memory pointed to. For example,
11705 @code{*ptr@@50} can be used to collect the 50 element array pointed to
11709 It is not possible to collect a complete stack backtrace at a
11710 tracepoint. Instead, you may collect the registers and a few hundred
11711 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
11712 (adjust to use the name of the actual stack pointer register on your
11713 target architecture, and the amount of stack you wish to capture).
11714 Then the @code{backtrace} command will show a partial backtrace when
11715 using a trace frame. The number of stack frames that can be examined
11716 depends on the sizes of the frames in the collected stack. Note that
11717 if you ask for a block so large that it goes past the bottom of the
11718 stack, the target agent may report an error trying to read from an
11722 If you do not collect registers at a tracepoint, @value{GDBN} can
11723 infer that the value of @code{$pc} must be the same as the address of
11724 the tracepoint and use that when you are looking at a trace frame
11725 for that tracepoint. However, this cannot work if the tracepoint has
11726 multiple locations (for instance if it was set in a function that was
11727 inlined), or if it has a @code{while-stepping} loop. In those cases
11728 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
11733 @node Analyze Collected Data
11734 @section Using the Collected Data
11736 After the tracepoint experiment ends, you use @value{GDBN} commands
11737 for examining the trace data. The basic idea is that each tracepoint
11738 collects a trace @dfn{snapshot} every time it is hit and another
11739 snapshot every time it single-steps. All these snapshots are
11740 consecutively numbered from zero and go into a buffer, and you can
11741 examine them later. The way you examine them is to @dfn{focus} on a
11742 specific trace snapshot. When the remote stub is focused on a trace
11743 snapshot, it will respond to all @value{GDBN} requests for memory and
11744 registers by reading from the buffer which belongs to that snapshot,
11745 rather than from @emph{real} memory or registers of the program being
11746 debugged. This means that @strong{all} @value{GDBN} commands
11747 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
11748 behave as if we were currently debugging the program state as it was
11749 when the tracepoint occurred. Any requests for data that are not in
11750 the buffer will fail.
11753 * tfind:: How to select a trace snapshot
11754 * tdump:: How to display all data for a snapshot
11755 * save tracepoints:: How to save tracepoints for a future run
11759 @subsection @code{tfind @var{n}}
11762 @cindex select trace snapshot
11763 @cindex find trace snapshot
11764 The basic command for selecting a trace snapshot from the buffer is
11765 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
11766 counting from zero. If no argument @var{n} is given, the next
11767 snapshot is selected.
11769 Here are the various forms of using the @code{tfind} command.
11773 Find the first snapshot in the buffer. This is a synonym for
11774 @code{tfind 0} (since 0 is the number of the first snapshot).
11777 Stop debugging trace snapshots, resume @emph{live} debugging.
11780 Same as @samp{tfind none}.
11783 No argument means find the next trace snapshot.
11786 Find the previous trace snapshot before the current one. This permits
11787 retracing earlier steps.
11789 @item tfind tracepoint @var{num}
11790 Find the next snapshot associated with tracepoint @var{num}. Search
11791 proceeds forward from the last examined trace snapshot. If no
11792 argument @var{num} is given, it means find the next snapshot collected
11793 for the same tracepoint as the current snapshot.
11795 @item tfind pc @var{addr}
11796 Find the next snapshot associated with the value @var{addr} of the
11797 program counter. Search proceeds forward from the last examined trace
11798 snapshot. If no argument @var{addr} is given, it means find the next
11799 snapshot with the same value of PC as the current snapshot.
11801 @item tfind outside @var{addr1}, @var{addr2}
11802 Find the next snapshot whose PC is outside the given range of
11803 addresses (exclusive).
11805 @item tfind range @var{addr1}, @var{addr2}
11806 Find the next snapshot whose PC is between @var{addr1} and
11807 @var{addr2} (inclusive).
11809 @item tfind line @r{[}@var{file}:@r{]}@var{n}
11810 Find the next snapshot associated with the source line @var{n}. If
11811 the optional argument @var{file} is given, refer to line @var{n} in
11812 that source file. Search proceeds forward from the last examined
11813 trace snapshot. If no argument @var{n} is given, it means find the
11814 next line other than the one currently being examined; thus saying
11815 @code{tfind line} repeatedly can appear to have the same effect as
11816 stepping from line to line in a @emph{live} debugging session.
11819 The default arguments for the @code{tfind} commands are specifically
11820 designed to make it easy to scan through the trace buffer. For
11821 instance, @code{tfind} with no argument selects the next trace
11822 snapshot, and @code{tfind -} with no argument selects the previous
11823 trace snapshot. So, by giving one @code{tfind} command, and then
11824 simply hitting @key{RET} repeatedly you can examine all the trace
11825 snapshots in order. Or, by saying @code{tfind -} and then hitting
11826 @key{RET} repeatedly you can examine the snapshots in reverse order.
11827 The @code{tfind line} command with no argument selects the snapshot
11828 for the next source line executed. The @code{tfind pc} command with
11829 no argument selects the next snapshot with the same program counter
11830 (PC) as the current frame. The @code{tfind tracepoint} command with
11831 no argument selects the next trace snapshot collected by the same
11832 tracepoint as the current one.
11834 In addition to letting you scan through the trace buffer manually,
11835 these commands make it easy to construct @value{GDBN} scripts that
11836 scan through the trace buffer and print out whatever collected data
11837 you are interested in. Thus, if we want to examine the PC, FP, and SP
11838 registers from each trace frame in the buffer, we can say this:
11841 (@value{GDBP}) @b{tfind start}
11842 (@value{GDBP}) @b{while ($trace_frame != -1)}
11843 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
11844 $trace_frame, $pc, $sp, $fp
11848 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
11849 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
11850 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
11851 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
11852 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
11853 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
11854 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
11855 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
11856 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
11857 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
11858 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
11861 Or, if we want to examine the variable @code{X} at each source line in
11865 (@value{GDBP}) @b{tfind start}
11866 (@value{GDBP}) @b{while ($trace_frame != -1)}
11867 > printf "Frame %d, X == %d\n", $trace_frame, X
11877 @subsection @code{tdump}
11879 @cindex dump all data collected at tracepoint
11880 @cindex tracepoint data, display
11882 This command takes no arguments. It prints all the data collected at
11883 the current trace snapshot.
11886 (@value{GDBP}) @b{trace 444}
11887 (@value{GDBP}) @b{actions}
11888 Enter actions for tracepoint #2, one per line:
11889 > collect $regs, $locals, $args, gdb_long_test
11892 (@value{GDBP}) @b{tstart}
11894 (@value{GDBP}) @b{tfind line 444}
11895 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
11897 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
11899 (@value{GDBP}) @b{tdump}
11900 Data collected at tracepoint 2, trace frame 1:
11901 d0 0xc4aa0085 -995491707
11905 d4 0x71aea3d 119204413
11908 d7 0x380035 3670069
11909 a0 0x19e24a 1696330
11910 a1 0x3000668 50333288
11912 a3 0x322000 3284992
11913 a4 0x3000698 50333336
11914 a5 0x1ad3cc 1758156
11915 fp 0x30bf3c 0x30bf3c
11916 sp 0x30bf34 0x30bf34
11918 pc 0x20b2c8 0x20b2c8
11922 p = 0x20e5b4 "gdb-test"
11929 gdb_long_test = 17 '\021'
11934 @code{tdump} works by scanning the tracepoint's current collection
11935 actions and printing the value of each expression listed. So
11936 @code{tdump} can fail, if after a run, you change the tracepoint's
11937 actions to mention variables that were not collected during the run.
11939 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
11940 uses the collected value of @code{$pc} to distinguish between trace
11941 frames that were collected at the tracepoint hit, and frames that were
11942 collected while stepping. This allows it to correctly choose whether
11943 to display the basic list of collections, or the collections from the
11944 body of the while-stepping loop. However, if @code{$pc} was not collected,
11945 then @code{tdump} will always attempt to dump using the basic collection
11946 list, and may fail if a while-stepping frame does not include all the
11947 same data that is collected at the tracepoint hit.
11948 @c This is getting pretty arcane, example would be good.
11950 @node save tracepoints
11951 @subsection @code{save tracepoints @var{filename}}
11952 @kindex save tracepoints
11953 @kindex save-tracepoints
11954 @cindex save tracepoints for future sessions
11956 This command saves all current tracepoint definitions together with
11957 their actions and passcounts, into a file @file{@var{filename}}
11958 suitable for use in a later debugging session. To read the saved
11959 tracepoint definitions, use the @code{source} command (@pxref{Command
11960 Files}). The @w{@code{save-tracepoints}} command is a deprecated
11961 alias for @w{@code{save tracepoints}}
11963 @node Tracepoint Variables
11964 @section Convenience Variables for Tracepoints
11965 @cindex tracepoint variables
11966 @cindex convenience variables for tracepoints
11969 @vindex $trace_frame
11970 @item (int) $trace_frame
11971 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
11972 snapshot is selected.
11974 @vindex $tracepoint
11975 @item (int) $tracepoint
11976 The tracepoint for the current trace snapshot.
11978 @vindex $trace_line
11979 @item (int) $trace_line
11980 The line number for the current trace snapshot.
11982 @vindex $trace_file
11983 @item (char []) $trace_file
11984 The source file for the current trace snapshot.
11986 @vindex $trace_func
11987 @item (char []) $trace_func
11988 The name of the function containing @code{$tracepoint}.
11991 Note: @code{$trace_file} is not suitable for use in @code{printf},
11992 use @code{output} instead.
11994 Here's a simple example of using these convenience variables for
11995 stepping through all the trace snapshots and printing some of their
11996 data. Note that these are not the same as trace state variables,
11997 which are managed by the target.
12000 (@value{GDBP}) @b{tfind start}
12002 (@value{GDBP}) @b{while $trace_frame != -1}
12003 > output $trace_file
12004 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
12010 @section Using Trace Files
12011 @cindex trace files
12013 In some situations, the target running a trace experiment may no
12014 longer be available; perhaps it crashed, or the hardware was needed
12015 for a different activity. To handle these cases, you can arrange to
12016 dump the trace data into a file, and later use that file as a source
12017 of trace data, via the @code{target tfile} command.
12022 @item tsave [ -r ] @var{filename}
12023 Save the trace data to @var{filename}. By default, this command
12024 assumes that @var{filename} refers to the host filesystem, so if
12025 necessary @value{GDBN} will copy raw trace data up from the target and
12026 then save it. If the target supports it, you can also supply the
12027 optional argument @code{-r} (``remote'') to direct the target to save
12028 the data directly into @var{filename} in its own filesystem, which may be
12029 more efficient if the trace buffer is very large. (Note, however, that
12030 @code{target tfile} can only read from files accessible to the host.)
12032 @kindex target tfile
12034 @item target tfile @var{filename}
12035 Use the file named @var{filename} as a source of trace data. Commands
12036 that examine data work as they do with a live target, but it is not
12037 possible to run any new trace experiments. @code{tstatus} will report
12038 the state of the trace run at the moment the data was saved, as well
12039 as the current trace frame you are examining. @var{filename} must be
12040 on a filesystem accessible to the host.
12045 @chapter Debugging Programs That Use Overlays
12048 If your program is too large to fit completely in your target system's
12049 memory, you can sometimes use @dfn{overlays} to work around this
12050 problem. @value{GDBN} provides some support for debugging programs that
12054 * How Overlays Work:: A general explanation of overlays.
12055 * Overlay Commands:: Managing overlays in @value{GDBN}.
12056 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
12057 mapped by asking the inferior.
12058 * Overlay Sample Program:: A sample program using overlays.
12061 @node How Overlays Work
12062 @section How Overlays Work
12063 @cindex mapped overlays
12064 @cindex unmapped overlays
12065 @cindex load address, overlay's
12066 @cindex mapped address
12067 @cindex overlay area
12069 Suppose you have a computer whose instruction address space is only 64
12070 kilobytes long, but which has much more memory which can be accessed by
12071 other means: special instructions, segment registers, or memory
12072 management hardware, for example. Suppose further that you want to
12073 adapt a program which is larger than 64 kilobytes to run on this system.
12075 One solution is to identify modules of your program which are relatively
12076 independent, and need not call each other directly; call these modules
12077 @dfn{overlays}. Separate the overlays from the main program, and place
12078 their machine code in the larger memory. Place your main program in
12079 instruction memory, but leave at least enough space there to hold the
12080 largest overlay as well.
12082 Now, to call a function located in an overlay, you must first copy that
12083 overlay's machine code from the large memory into the space set aside
12084 for it in the instruction memory, and then jump to its entry point
12087 @c NB: In the below the mapped area's size is greater or equal to the
12088 @c size of all overlays. This is intentional to remind the developer
12089 @c that overlays don't necessarily need to be the same size.
12093 Data Instruction Larger
12094 Address Space Address Space Address Space
12095 +-----------+ +-----------+ +-----------+
12097 +-----------+ +-----------+ +-----------+<-- overlay 1
12098 | program | | main | .----| overlay 1 | load address
12099 | variables | | program | | +-----------+
12100 | and heap | | | | | |
12101 +-----------+ | | | +-----------+<-- overlay 2
12102 | | +-----------+ | | | load address
12103 +-----------+ | | | .-| overlay 2 |
12105 mapped --->+-----------+ | | +-----------+
12106 address | | | | | |
12107 | overlay | <-' | | |
12108 | area | <---' +-----------+<-- overlay 3
12109 | | <---. | | load address
12110 +-----------+ `--| overlay 3 |
12117 @anchor{A code overlay}A code overlay
12121 The diagram (@pxref{A code overlay}) shows a system with separate data
12122 and instruction address spaces. To map an overlay, the program copies
12123 its code from the larger address space to the instruction address space.
12124 Since the overlays shown here all use the same mapped address, only one
12125 may be mapped at a time. For a system with a single address space for
12126 data and instructions, the diagram would be similar, except that the
12127 program variables and heap would share an address space with the main
12128 program and the overlay area.
12130 An overlay loaded into instruction memory and ready for use is called a
12131 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
12132 instruction memory. An overlay not present (or only partially present)
12133 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
12134 is its address in the larger memory. The mapped address is also called
12135 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
12136 called the @dfn{load memory address}, or @dfn{LMA}.
12138 Unfortunately, overlays are not a completely transparent way to adapt a
12139 program to limited instruction memory. They introduce a new set of
12140 global constraints you must keep in mind as you design your program:
12145 Before calling or returning to a function in an overlay, your program
12146 must make sure that overlay is actually mapped. Otherwise, the call or
12147 return will transfer control to the right address, but in the wrong
12148 overlay, and your program will probably crash.
12151 If the process of mapping an overlay is expensive on your system, you
12152 will need to choose your overlays carefully to minimize their effect on
12153 your program's performance.
12156 The executable file you load onto your system must contain each
12157 overlay's instructions, appearing at the overlay's load address, not its
12158 mapped address. However, each overlay's instructions must be relocated
12159 and its symbols defined as if the overlay were at its mapped address.
12160 You can use GNU linker scripts to specify different load and relocation
12161 addresses for pieces of your program; see @ref{Overlay Description,,,
12162 ld.info, Using ld: the GNU linker}.
12165 The procedure for loading executable files onto your system must be able
12166 to load their contents into the larger address space as well as the
12167 instruction and data spaces.
12171 The overlay system described above is rather simple, and could be
12172 improved in many ways:
12177 If your system has suitable bank switch registers or memory management
12178 hardware, you could use those facilities to make an overlay's load area
12179 contents simply appear at their mapped address in instruction space.
12180 This would probably be faster than copying the overlay to its mapped
12181 area in the usual way.
12184 If your overlays are small enough, you could set aside more than one
12185 overlay area, and have more than one overlay mapped at a time.
12188 You can use overlays to manage data, as well as instructions. In
12189 general, data overlays are even less transparent to your design than
12190 code overlays: whereas code overlays only require care when you call or
12191 return to functions, data overlays require care every time you access
12192 the data. Also, if you change the contents of a data overlay, you
12193 must copy its contents back out to its load address before you can copy a
12194 different data overlay into the same mapped area.
12199 @node Overlay Commands
12200 @section Overlay Commands
12202 To use @value{GDBN}'s overlay support, each overlay in your program must
12203 correspond to a separate section of the executable file. The section's
12204 virtual memory address and load memory address must be the overlay's
12205 mapped and load addresses. Identifying overlays with sections allows
12206 @value{GDBN} to determine the appropriate address of a function or
12207 variable, depending on whether the overlay is mapped or not.
12209 @value{GDBN}'s overlay commands all start with the word @code{overlay};
12210 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
12215 Disable @value{GDBN}'s overlay support. When overlay support is
12216 disabled, @value{GDBN} assumes that all functions and variables are
12217 always present at their mapped addresses. By default, @value{GDBN}'s
12218 overlay support is disabled.
12220 @item overlay manual
12221 @cindex manual overlay debugging
12222 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
12223 relies on you to tell it which overlays are mapped, and which are not,
12224 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
12225 commands described below.
12227 @item overlay map-overlay @var{overlay}
12228 @itemx overlay map @var{overlay}
12229 @cindex map an overlay
12230 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
12231 be the name of the object file section containing the overlay. When an
12232 overlay is mapped, @value{GDBN} assumes it can find the overlay's
12233 functions and variables at their mapped addresses. @value{GDBN} assumes
12234 that any other overlays whose mapped ranges overlap that of
12235 @var{overlay} are now unmapped.
12237 @item overlay unmap-overlay @var{overlay}
12238 @itemx overlay unmap @var{overlay}
12239 @cindex unmap an overlay
12240 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
12241 must be the name of the object file section containing the overlay.
12242 When an overlay is unmapped, @value{GDBN} assumes it can find the
12243 overlay's functions and variables at their load addresses.
12246 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
12247 consults a data structure the overlay manager maintains in the inferior
12248 to see which overlays are mapped. For details, see @ref{Automatic
12249 Overlay Debugging}.
12251 @item overlay load-target
12252 @itemx overlay load
12253 @cindex reloading the overlay table
12254 Re-read the overlay table from the inferior. Normally, @value{GDBN}
12255 re-reads the table @value{GDBN} automatically each time the inferior
12256 stops, so this command should only be necessary if you have changed the
12257 overlay mapping yourself using @value{GDBN}. This command is only
12258 useful when using automatic overlay debugging.
12260 @item overlay list-overlays
12261 @itemx overlay list
12262 @cindex listing mapped overlays
12263 Display a list of the overlays currently mapped, along with their mapped
12264 addresses, load addresses, and sizes.
12268 Normally, when @value{GDBN} prints a code address, it includes the name
12269 of the function the address falls in:
12272 (@value{GDBP}) print main
12273 $3 = @{int ()@} 0x11a0 <main>
12276 When overlay debugging is enabled, @value{GDBN} recognizes code in
12277 unmapped overlays, and prints the names of unmapped functions with
12278 asterisks around them. For example, if @code{foo} is a function in an
12279 unmapped overlay, @value{GDBN} prints it this way:
12282 (@value{GDBP}) overlay list
12283 No sections are mapped.
12284 (@value{GDBP}) print foo
12285 $5 = @{int (int)@} 0x100000 <*foo*>
12288 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
12292 (@value{GDBP}) overlay list
12293 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
12294 mapped at 0x1016 - 0x104a
12295 (@value{GDBP}) print foo
12296 $6 = @{int (int)@} 0x1016 <foo>
12299 When overlay debugging is enabled, @value{GDBN} can find the correct
12300 address for functions and variables in an overlay, whether or not the
12301 overlay is mapped. This allows most @value{GDBN} commands, like
12302 @code{break} and @code{disassemble}, to work normally, even on unmapped
12303 code. However, @value{GDBN}'s breakpoint support has some limitations:
12307 @cindex breakpoints in overlays
12308 @cindex overlays, setting breakpoints in
12309 You can set breakpoints in functions in unmapped overlays, as long as
12310 @value{GDBN} can write to the overlay at its load address.
12312 @value{GDBN} can not set hardware or simulator-based breakpoints in
12313 unmapped overlays. However, if you set a breakpoint at the end of your
12314 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
12315 you are using manual overlay management), @value{GDBN} will re-set its
12316 breakpoints properly.
12320 @node Automatic Overlay Debugging
12321 @section Automatic Overlay Debugging
12322 @cindex automatic overlay debugging
12324 @value{GDBN} can automatically track which overlays are mapped and which
12325 are not, given some simple co-operation from the overlay manager in the
12326 inferior. If you enable automatic overlay debugging with the
12327 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
12328 looks in the inferior's memory for certain variables describing the
12329 current state of the overlays.
12331 Here are the variables your overlay manager must define to support
12332 @value{GDBN}'s automatic overlay debugging:
12336 @item @code{_ovly_table}:
12337 This variable must be an array of the following structures:
12342 /* The overlay's mapped address. */
12345 /* The size of the overlay, in bytes. */
12346 unsigned long size;
12348 /* The overlay's load address. */
12351 /* Non-zero if the overlay is currently mapped;
12353 unsigned long mapped;
12357 @item @code{_novlys}:
12358 This variable must be a four-byte signed integer, holding the total
12359 number of elements in @code{_ovly_table}.
12363 To decide whether a particular overlay is mapped or not, @value{GDBN}
12364 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
12365 @code{lma} members equal the VMA and LMA of the overlay's section in the
12366 executable file. When @value{GDBN} finds a matching entry, it consults
12367 the entry's @code{mapped} member to determine whether the overlay is
12370 In addition, your overlay manager may define a function called
12371 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
12372 will silently set a breakpoint there. If the overlay manager then
12373 calls this function whenever it has changed the overlay table, this
12374 will enable @value{GDBN} to accurately keep track of which overlays
12375 are in program memory, and update any breakpoints that may be set
12376 in overlays. This will allow breakpoints to work even if the
12377 overlays are kept in ROM or other non-writable memory while they
12378 are not being executed.
12380 @node Overlay Sample Program
12381 @section Overlay Sample Program
12382 @cindex overlay example program
12384 When linking a program which uses overlays, you must place the overlays
12385 at their load addresses, while relocating them to run at their mapped
12386 addresses. To do this, you must write a linker script (@pxref{Overlay
12387 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
12388 since linker scripts are specific to a particular host system, target
12389 architecture, and target memory layout, this manual cannot provide
12390 portable sample code demonstrating @value{GDBN}'s overlay support.
12392 However, the @value{GDBN} source distribution does contain an overlaid
12393 program, with linker scripts for a few systems, as part of its test
12394 suite. The program consists of the following files from
12395 @file{gdb/testsuite/gdb.base}:
12399 The main program file.
12401 A simple overlay manager, used by @file{overlays.c}.
12406 Overlay modules, loaded and used by @file{overlays.c}.
12409 Linker scripts for linking the test program on the @code{d10v-elf}
12410 and @code{m32r-elf} targets.
12413 You can build the test program using the @code{d10v-elf} GCC
12414 cross-compiler like this:
12417 $ d10v-elf-gcc -g -c overlays.c
12418 $ d10v-elf-gcc -g -c ovlymgr.c
12419 $ d10v-elf-gcc -g -c foo.c
12420 $ d10v-elf-gcc -g -c bar.c
12421 $ d10v-elf-gcc -g -c baz.c
12422 $ d10v-elf-gcc -g -c grbx.c
12423 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
12424 baz.o grbx.o -Wl,-Td10v.ld -o overlays
12427 The build process is identical for any other architecture, except that
12428 you must substitute the appropriate compiler and linker script for the
12429 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
12433 @chapter Using @value{GDBN} with Different Languages
12436 Although programming languages generally have common aspects, they are
12437 rarely expressed in the same manner. For instance, in ANSI C,
12438 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
12439 Modula-2, it is accomplished by @code{p^}. Values can also be
12440 represented (and displayed) differently. Hex numbers in C appear as
12441 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
12443 @cindex working language
12444 Language-specific information is built into @value{GDBN} for some languages,
12445 allowing you to express operations like the above in your program's
12446 native language, and allowing @value{GDBN} to output values in a manner
12447 consistent with the syntax of your program's native language. The
12448 language you use to build expressions is called the @dfn{working
12452 * Setting:: Switching between source languages
12453 * Show:: Displaying the language
12454 * Checks:: Type and range checks
12455 * Supported Languages:: Supported languages
12456 * Unsupported Languages:: Unsupported languages
12460 @section Switching Between Source Languages
12462 There are two ways to control the working language---either have @value{GDBN}
12463 set it automatically, or select it manually yourself. You can use the
12464 @code{set language} command for either purpose. On startup, @value{GDBN}
12465 defaults to setting the language automatically. The working language is
12466 used to determine how expressions you type are interpreted, how values
12469 In addition to the working language, every source file that
12470 @value{GDBN} knows about has its own working language. For some object
12471 file formats, the compiler might indicate which language a particular
12472 source file is in. However, most of the time @value{GDBN} infers the
12473 language from the name of the file. The language of a source file
12474 controls whether C@t{++} names are demangled---this way @code{backtrace} can
12475 show each frame appropriately for its own language. There is no way to
12476 set the language of a source file from within @value{GDBN}, but you can
12477 set the language associated with a filename extension. @xref{Show, ,
12478 Displaying the Language}.
12480 This is most commonly a problem when you use a program, such
12481 as @code{cfront} or @code{f2c}, that generates C but is written in
12482 another language. In that case, make the
12483 program use @code{#line} directives in its C output; that way
12484 @value{GDBN} will know the correct language of the source code of the original
12485 program, and will display that source code, not the generated C code.
12488 * Filenames:: Filename extensions and languages.
12489 * Manually:: Setting the working language manually
12490 * Automatically:: Having @value{GDBN} infer the source language
12494 @subsection List of Filename Extensions and Languages
12496 If a source file name ends in one of the following extensions, then
12497 @value{GDBN} infers that its language is the one indicated.
12515 C@t{++} source file
12521 Objective-C source file
12525 Fortran source file
12528 Modula-2 source file
12532 Assembler source file. This actually behaves almost like C, but
12533 @value{GDBN} does not skip over function prologues when stepping.
12536 In addition, you may set the language associated with a filename
12537 extension. @xref{Show, , Displaying the Language}.
12540 @subsection Setting the Working Language
12542 If you allow @value{GDBN} to set the language automatically,
12543 expressions are interpreted the same way in your debugging session and
12546 @kindex set language
12547 If you wish, you may set the language manually. To do this, issue the
12548 command @samp{set language @var{lang}}, where @var{lang} is the name of
12549 a language, such as
12550 @code{c} or @code{modula-2}.
12551 For a list of the supported languages, type @samp{set language}.
12553 Setting the language manually prevents @value{GDBN} from updating the working
12554 language automatically. This can lead to confusion if you try
12555 to debug a program when the working language is not the same as the
12556 source language, when an expression is acceptable to both
12557 languages---but means different things. For instance, if the current
12558 source file were written in C, and @value{GDBN} was parsing Modula-2, a
12566 might not have the effect you intended. In C, this means to add
12567 @code{b} and @code{c} and place the result in @code{a}. The result
12568 printed would be the value of @code{a}. In Modula-2, this means to compare
12569 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
12571 @node Automatically
12572 @subsection Having @value{GDBN} Infer the Source Language
12574 To have @value{GDBN} set the working language automatically, use
12575 @samp{set language local} or @samp{set language auto}. @value{GDBN}
12576 then infers the working language. That is, when your program stops in a
12577 frame (usually by encountering a breakpoint), @value{GDBN} sets the
12578 working language to the language recorded for the function in that
12579 frame. If the language for a frame is unknown (that is, if the function
12580 or block corresponding to the frame was defined in a source file that
12581 does not have a recognized extension), the current working language is
12582 not changed, and @value{GDBN} issues a warning.
12584 This may not seem necessary for most programs, which are written
12585 entirely in one source language. However, program modules and libraries
12586 written in one source language can be used by a main program written in
12587 a different source language. Using @samp{set language auto} in this
12588 case frees you from having to set the working language manually.
12591 @section Displaying the Language
12593 The following commands help you find out which language is the
12594 working language, and also what language source files were written in.
12597 @item show language
12598 @kindex show language
12599 Display the current working language. This is the
12600 language you can use with commands such as @code{print} to
12601 build and compute expressions that may involve variables in your program.
12604 @kindex info frame@r{, show the source language}
12605 Display the source language for this frame. This language becomes the
12606 working language if you use an identifier from this frame.
12607 @xref{Frame Info, ,Information about a Frame}, to identify the other
12608 information listed here.
12611 @kindex info source@r{, show the source language}
12612 Display the source language of this source file.
12613 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
12614 information listed here.
12617 In unusual circumstances, you may have source files with extensions
12618 not in the standard list. You can then set the extension associated
12619 with a language explicitly:
12622 @item set extension-language @var{ext} @var{language}
12623 @kindex set extension-language
12624 Tell @value{GDBN} that source files with extension @var{ext} are to be
12625 assumed as written in the source language @var{language}.
12627 @item info extensions
12628 @kindex info extensions
12629 List all the filename extensions and the associated languages.
12633 @section Type and Range Checking
12636 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
12637 checking are included, but they do not yet have any effect. This
12638 section documents the intended facilities.
12640 @c FIXME remove warning when type/range code added
12642 Some languages are designed to guard you against making seemingly common
12643 errors through a series of compile- and run-time checks. These include
12644 checking the type of arguments to functions and operators, and making
12645 sure mathematical overflows are caught at run time. Checks such as
12646 these help to ensure a program's correctness once it has been compiled
12647 by eliminating type mismatches, and providing active checks for range
12648 errors when your program is running.
12650 @value{GDBN} can check for conditions like the above if you wish.
12651 Although @value{GDBN} does not check the statements in your program,
12652 it can check expressions entered directly into @value{GDBN} for
12653 evaluation via the @code{print} command, for example. As with the
12654 working language, @value{GDBN} can also decide whether or not to check
12655 automatically based on your program's source language.
12656 @xref{Supported Languages, ,Supported Languages}, for the default
12657 settings of supported languages.
12660 * Type Checking:: An overview of type checking
12661 * Range Checking:: An overview of range checking
12664 @cindex type checking
12665 @cindex checks, type
12666 @node Type Checking
12667 @subsection An Overview of Type Checking
12669 Some languages, such as Modula-2, are strongly typed, meaning that the
12670 arguments to operators and functions have to be of the correct type,
12671 otherwise an error occurs. These checks prevent type mismatch
12672 errors from ever causing any run-time problems. For example,
12680 The second example fails because the @code{CARDINAL} 1 is not
12681 type-compatible with the @code{REAL} 2.3.
12683 For the expressions you use in @value{GDBN} commands, you can tell the
12684 @value{GDBN} type checker to skip checking;
12685 to treat any mismatches as errors and abandon the expression;
12686 or to only issue warnings when type mismatches occur,
12687 but evaluate the expression anyway. When you choose the last of
12688 these, @value{GDBN} evaluates expressions like the second example above, but
12689 also issues a warning.
12691 Even if you turn type checking off, there may be other reasons
12692 related to type that prevent @value{GDBN} from evaluating an expression.
12693 For instance, @value{GDBN} does not know how to add an @code{int} and
12694 a @code{struct foo}. These particular type errors have nothing to do
12695 with the language in use, and usually arise from expressions, such as
12696 the one described above, which make little sense to evaluate anyway.
12698 Each language defines to what degree it is strict about type. For
12699 instance, both Modula-2 and C require the arguments to arithmetical
12700 operators to be numbers. In C, enumerated types and pointers can be
12701 represented as numbers, so that they are valid arguments to mathematical
12702 operators. @xref{Supported Languages, ,Supported Languages}, for further
12703 details on specific languages.
12705 @value{GDBN} provides some additional commands for controlling the type checker:
12707 @kindex set check type
12708 @kindex show check type
12710 @item set check type auto
12711 Set type checking on or off based on the current working language.
12712 @xref{Supported Languages, ,Supported Languages}, for the default settings for
12715 @item set check type on
12716 @itemx set check type off
12717 Set type checking on or off, overriding the default setting for the
12718 current working language. Issue a warning if the setting does not
12719 match the language default. If any type mismatches occur in
12720 evaluating an expression while type checking is on, @value{GDBN} prints a
12721 message and aborts evaluation of the expression.
12723 @item set check type warn
12724 Cause the type checker to issue warnings, but to always attempt to
12725 evaluate the expression. Evaluating the expression may still
12726 be impossible for other reasons. For example, @value{GDBN} cannot add
12727 numbers and structures.
12730 Show the current setting of the type checker, and whether or not @value{GDBN}
12731 is setting it automatically.
12734 @cindex range checking
12735 @cindex checks, range
12736 @node Range Checking
12737 @subsection An Overview of Range Checking
12739 In some languages (such as Modula-2), it is an error to exceed the
12740 bounds of a type; this is enforced with run-time checks. Such range
12741 checking is meant to ensure program correctness by making sure
12742 computations do not overflow, or indices on an array element access do
12743 not exceed the bounds of the array.
12745 For expressions you use in @value{GDBN} commands, you can tell
12746 @value{GDBN} to treat range errors in one of three ways: ignore them,
12747 always treat them as errors and abandon the expression, or issue
12748 warnings but evaluate the expression anyway.
12750 A range error can result from numerical overflow, from exceeding an
12751 array index bound, or when you type a constant that is not a member
12752 of any type. Some languages, however, do not treat overflows as an
12753 error. In many implementations of C, mathematical overflow causes the
12754 result to ``wrap around'' to lower values---for example, if @var{m} is
12755 the largest integer value, and @var{s} is the smallest, then
12758 @var{m} + 1 @result{} @var{s}
12761 This, too, is specific to individual languages, and in some cases
12762 specific to individual compilers or machines. @xref{Supported Languages, ,
12763 Supported Languages}, for further details on specific languages.
12765 @value{GDBN} provides some additional commands for controlling the range checker:
12767 @kindex set check range
12768 @kindex show check range
12770 @item set check range auto
12771 Set range checking on or off based on the current working language.
12772 @xref{Supported Languages, ,Supported Languages}, for the default settings for
12775 @item set check range on
12776 @itemx set check range off
12777 Set range checking on or off, overriding the default setting for the
12778 current working language. A warning is issued if the setting does not
12779 match the language default. If a range error occurs and range checking is on,
12780 then a message is printed and evaluation of the expression is aborted.
12782 @item set check range warn
12783 Output messages when the @value{GDBN} range checker detects a range error,
12784 but attempt to evaluate the expression anyway. Evaluating the
12785 expression may still be impossible for other reasons, such as accessing
12786 memory that the process does not own (a typical example from many Unix
12790 Show the current setting of the range checker, and whether or not it is
12791 being set automatically by @value{GDBN}.
12794 @node Supported Languages
12795 @section Supported Languages
12797 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran, Java,
12798 OpenCL C, Pascal, assembly, Modula-2, and Ada.
12799 @c This is false ...
12800 Some @value{GDBN} features may be used in expressions regardless of the
12801 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
12802 and the @samp{@{type@}addr} construct (@pxref{Expressions,
12803 ,Expressions}) can be used with the constructs of any supported
12806 The following sections detail to what degree each source language is
12807 supported by @value{GDBN}. These sections are not meant to be language
12808 tutorials or references, but serve only as a reference guide to what the
12809 @value{GDBN} expression parser accepts, and what input and output
12810 formats should look like for different languages. There are many good
12811 books written on each of these languages; please look to these for a
12812 language reference or tutorial.
12815 * C:: C and C@t{++}
12818 * Objective-C:: Objective-C
12819 * OpenCL C:: OpenCL C
12820 * Fortran:: Fortran
12822 * Modula-2:: Modula-2
12827 @subsection C and C@t{++}
12829 @cindex C and C@t{++}
12830 @cindex expressions in C or C@t{++}
12832 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
12833 to both languages. Whenever this is the case, we discuss those languages
12837 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
12838 @cindex @sc{gnu} C@t{++}
12839 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
12840 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
12841 effectively, you must compile your C@t{++} programs with a supported
12842 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
12843 compiler (@code{aCC}).
12846 * C Operators:: C and C@t{++} operators
12847 * C Constants:: C and C@t{++} constants
12848 * C Plus Plus Expressions:: C@t{++} expressions
12849 * C Defaults:: Default settings for C and C@t{++}
12850 * C Checks:: C and C@t{++} type and range checks
12851 * Debugging C:: @value{GDBN} and C
12852 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
12853 * Decimal Floating Point:: Numbers in Decimal Floating Point format
12857 @subsubsection C and C@t{++} Operators
12859 @cindex C and C@t{++} operators
12861 Operators must be defined on values of specific types. For instance,
12862 @code{+} is defined on numbers, but not on structures. Operators are
12863 often defined on groups of types.
12865 For the purposes of C and C@t{++}, the following definitions hold:
12870 @emph{Integral types} include @code{int} with any of its storage-class
12871 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
12874 @emph{Floating-point types} include @code{float}, @code{double}, and
12875 @code{long double} (if supported by the target platform).
12878 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
12881 @emph{Scalar types} include all of the above.
12886 The following operators are supported. They are listed here
12887 in order of increasing precedence:
12891 The comma or sequencing operator. Expressions in a comma-separated list
12892 are evaluated from left to right, with the result of the entire
12893 expression being the last expression evaluated.
12896 Assignment. The value of an assignment expression is the value
12897 assigned. Defined on scalar types.
12900 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
12901 and translated to @w{@code{@var{a} = @var{a op b}}}.
12902 @w{@code{@var{op}=}} and @code{=} have the same precedence.
12903 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
12904 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
12907 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
12908 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
12912 Logical @sc{or}. Defined on integral types.
12915 Logical @sc{and}. Defined on integral types.
12918 Bitwise @sc{or}. Defined on integral types.
12921 Bitwise exclusive-@sc{or}. Defined on integral types.
12924 Bitwise @sc{and}. Defined on integral types.
12927 Equality and inequality. Defined on scalar types. The value of these
12928 expressions is 0 for false and non-zero for true.
12930 @item <@r{, }>@r{, }<=@r{, }>=
12931 Less than, greater than, less than or equal, greater than or equal.
12932 Defined on scalar types. The value of these expressions is 0 for false
12933 and non-zero for true.
12936 left shift, and right shift. Defined on integral types.
12939 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
12942 Addition and subtraction. Defined on integral types, floating-point types and
12945 @item *@r{, }/@r{, }%
12946 Multiplication, division, and modulus. Multiplication and division are
12947 defined on integral and floating-point types. Modulus is defined on
12951 Increment and decrement. When appearing before a variable, the
12952 operation is performed before the variable is used in an expression;
12953 when appearing after it, the variable's value is used before the
12954 operation takes place.
12957 Pointer dereferencing. Defined on pointer types. Same precedence as
12961 Address operator. Defined on variables. Same precedence as @code{++}.
12963 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
12964 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
12965 to examine the address
12966 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
12970 Negative. Defined on integral and floating-point types. Same
12971 precedence as @code{++}.
12974 Logical negation. Defined on integral types. Same precedence as
12978 Bitwise complement operator. Defined on integral types. Same precedence as
12983 Structure member, and pointer-to-structure member. For convenience,
12984 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
12985 pointer based on the stored type information.
12986 Defined on @code{struct} and @code{union} data.
12989 Dereferences of pointers to members.
12992 Array indexing. @code{@var{a}[@var{i}]} is defined as
12993 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
12996 Function parameter list. Same precedence as @code{->}.
12999 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
13000 and @code{class} types.
13003 Doubled colons also represent the @value{GDBN} scope operator
13004 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
13008 If an operator is redefined in the user code, @value{GDBN} usually
13009 attempts to invoke the redefined version instead of using the operator's
13010 predefined meaning.
13013 @subsubsection C and C@t{++} Constants
13015 @cindex C and C@t{++} constants
13017 @value{GDBN} allows you to express the constants of C and C@t{++} in the
13022 Integer constants are a sequence of digits. Octal constants are
13023 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
13024 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
13025 @samp{l}, specifying that the constant should be treated as a
13029 Floating point constants are a sequence of digits, followed by a decimal
13030 point, followed by a sequence of digits, and optionally followed by an
13031 exponent. An exponent is of the form:
13032 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
13033 sequence of digits. The @samp{+} is optional for positive exponents.
13034 A floating-point constant may also end with a letter @samp{f} or
13035 @samp{F}, specifying that the constant should be treated as being of
13036 the @code{float} (as opposed to the default @code{double}) type; or with
13037 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
13041 Enumerated constants consist of enumerated identifiers, or their
13042 integral equivalents.
13045 Character constants are a single character surrounded by single quotes
13046 (@code{'}), or a number---the ordinal value of the corresponding character
13047 (usually its @sc{ascii} value). Within quotes, the single character may
13048 be represented by a letter or by @dfn{escape sequences}, which are of
13049 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
13050 of the character's ordinal value; or of the form @samp{\@var{x}}, where
13051 @samp{@var{x}} is a predefined special character---for example,
13052 @samp{\n} for newline.
13054 Wide character constants can be written by prefixing a character
13055 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
13056 form of @samp{x}. The target wide character set is used when
13057 computing the value of this constant (@pxref{Character Sets}).
13060 String constants are a sequence of character constants surrounded by
13061 double quotes (@code{"}). Any valid character constant (as described
13062 above) may appear. Double quotes within the string must be preceded by
13063 a backslash, so for instance @samp{"a\"b'c"} is a string of five
13066 Wide string constants can be written by prefixing a string constant
13067 with @samp{L}, as in C. The target wide character set is used when
13068 computing the value of this constant (@pxref{Character Sets}).
13071 Pointer constants are an integral value. You can also write pointers
13072 to constants using the C operator @samp{&}.
13075 Array constants are comma-separated lists surrounded by braces @samp{@{}
13076 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
13077 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
13078 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
13081 @node C Plus Plus Expressions
13082 @subsubsection C@t{++} Expressions
13084 @cindex expressions in C@t{++}
13085 @value{GDBN} expression handling can interpret most C@t{++} expressions.
13087 @cindex debugging C@t{++} programs
13088 @cindex C@t{++} compilers
13089 @cindex debug formats and C@t{++}
13090 @cindex @value{NGCC} and C@t{++}
13092 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
13093 the proper compiler and the proper debug format. Currently,
13094 @value{GDBN} works best when debugging C@t{++} code that is compiled
13095 with the most recent version of @value{NGCC} possible. The DWARF
13096 debugging format is preferred; @value{NGCC} defaults to this on most
13097 popular platforms. Other compilers and/or debug formats are likely to
13098 work badly or not at all when using @value{GDBN} to debug C@t{++}
13099 code. @xref{Compilation}.
13104 @cindex member functions
13106 Member function calls are allowed; you can use expressions like
13109 count = aml->GetOriginal(x, y)
13112 @vindex this@r{, inside C@t{++} member functions}
13113 @cindex namespace in C@t{++}
13115 While a member function is active (in the selected stack frame), your
13116 expressions have the same namespace available as the member function;
13117 that is, @value{GDBN} allows implicit references to the class instance
13118 pointer @code{this} following the same rules as C@t{++}. @code{using}
13119 declarations in the current scope are also respected by @value{GDBN}.
13121 @cindex call overloaded functions
13122 @cindex overloaded functions, calling
13123 @cindex type conversions in C@t{++}
13125 You can call overloaded functions; @value{GDBN} resolves the function
13126 call to the right definition, with some restrictions. @value{GDBN} does not
13127 perform overload resolution involving user-defined type conversions,
13128 calls to constructors, or instantiations of templates that do not exist
13129 in the program. It also cannot handle ellipsis argument lists or
13132 It does perform integral conversions and promotions, floating-point
13133 promotions, arithmetic conversions, pointer conversions, conversions of
13134 class objects to base classes, and standard conversions such as those of
13135 functions or arrays to pointers; it requires an exact match on the
13136 number of function arguments.
13138 Overload resolution is always performed, unless you have specified
13139 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
13140 ,@value{GDBN} Features for C@t{++}}.
13142 You must specify @code{set overload-resolution off} in order to use an
13143 explicit function signature to call an overloaded function, as in
13145 p 'foo(char,int)'('x', 13)
13148 The @value{GDBN} command-completion facility can simplify this;
13149 see @ref{Completion, ,Command Completion}.
13151 @cindex reference declarations
13153 @value{GDBN} understands variables declared as C@t{++} references; you can use
13154 them in expressions just as you do in C@t{++} source---they are automatically
13157 In the parameter list shown when @value{GDBN} displays a frame, the values of
13158 reference variables are not displayed (unlike other variables); this
13159 avoids clutter, since references are often used for large structures.
13160 The @emph{address} of a reference variable is always shown, unless
13161 you have specified @samp{set print address off}.
13164 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
13165 expressions can use it just as expressions in your program do. Since
13166 one scope may be defined in another, you can use @code{::} repeatedly if
13167 necessary, for example in an expression like
13168 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
13169 resolving name scope by reference to source files, in both C and C@t{++}
13170 debugging (@pxref{Variables, ,Program Variables}).
13173 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
13178 @subsubsection C and C@t{++} Defaults
13180 @cindex C and C@t{++} defaults
13182 If you allow @value{GDBN} to set type and range checking automatically, they
13183 both default to @code{off} whenever the working language changes to
13184 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
13185 selects the working language.
13187 If you allow @value{GDBN} to set the language automatically, it
13188 recognizes source files whose names end with @file{.c}, @file{.C}, or
13189 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
13190 these files, it sets the working language to C or C@t{++}.
13191 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
13192 for further details.
13194 @c Type checking is (a) primarily motivated by Modula-2, and (b)
13195 @c unimplemented. If (b) changes, it might make sense to let this node
13196 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
13199 @subsubsection C and C@t{++} Type and Range Checks
13201 @cindex C and C@t{++} checks
13203 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
13204 is not used. However, if you turn type checking on, @value{GDBN}
13205 considers two variables type equivalent if:
13209 The two variables are structured and have the same structure, union, or
13213 The two variables have the same type name, or types that have been
13214 declared equivalent through @code{typedef}.
13217 @c leaving this out because neither J Gilmore nor R Pesch understand it.
13220 The two @code{struct}, @code{union}, or @code{enum} variables are
13221 declared in the same declaration. (Note: this may not be true for all C
13226 Range checking, if turned on, is done on mathematical operations. Array
13227 indices are not checked, since they are often used to index a pointer
13228 that is not itself an array.
13231 @subsubsection @value{GDBN} and C
13233 The @code{set print union} and @code{show print union} commands apply to
13234 the @code{union} type. When set to @samp{on}, any @code{union} that is
13235 inside a @code{struct} or @code{class} is also printed. Otherwise, it
13236 appears as @samp{@{...@}}.
13238 The @code{@@} operator aids in the debugging of dynamic arrays, formed
13239 with pointers and a memory allocation function. @xref{Expressions,
13242 @node Debugging C Plus Plus
13243 @subsubsection @value{GDBN} Features for C@t{++}
13245 @cindex commands for C@t{++}
13247 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
13248 designed specifically for use with C@t{++}. Here is a summary:
13251 @cindex break in overloaded functions
13252 @item @r{breakpoint menus}
13253 When you want a breakpoint in a function whose name is overloaded,
13254 @value{GDBN} has the capability to display a menu of possible breakpoint
13255 locations to help you specify which function definition you want.
13256 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
13258 @cindex overloading in C@t{++}
13259 @item rbreak @var{regex}
13260 Setting breakpoints using regular expressions is helpful for setting
13261 breakpoints on overloaded functions that are not members of any special
13263 @xref{Set Breaks, ,Setting Breakpoints}.
13265 @cindex C@t{++} exception handling
13268 Debug C@t{++} exception handling using these commands. @xref{Set
13269 Catchpoints, , Setting Catchpoints}.
13271 @cindex inheritance
13272 @item ptype @var{typename}
13273 Print inheritance relationships as well as other information for type
13275 @xref{Symbols, ,Examining the Symbol Table}.
13277 @item info vtbl @var{expression}.
13278 The @code{info vtbl} command can be used to display the virtual
13279 method tables of the object computed by @var{expression}. This shows
13280 one entry per virtual table; there may be multiple virtual tables when
13281 multiple inheritance is in use.
13283 @cindex C@t{++} symbol display
13284 @item set print demangle
13285 @itemx show print demangle
13286 @itemx set print asm-demangle
13287 @itemx show print asm-demangle
13288 Control whether C@t{++} symbols display in their source form, both when
13289 displaying code as C@t{++} source and when displaying disassemblies.
13290 @xref{Print Settings, ,Print Settings}.
13292 @item set print object
13293 @itemx show print object
13294 Choose whether to print derived (actual) or declared types of objects.
13295 @xref{Print Settings, ,Print Settings}.
13297 @item set print vtbl
13298 @itemx show print vtbl
13299 Control the format for printing virtual function tables.
13300 @xref{Print Settings, ,Print Settings}.
13301 (The @code{vtbl} commands do not work on programs compiled with the HP
13302 ANSI C@t{++} compiler (@code{aCC}).)
13304 @kindex set overload-resolution
13305 @cindex overloaded functions, overload resolution
13306 @item set overload-resolution on
13307 Enable overload resolution for C@t{++} expression evaluation. The default
13308 is on. For overloaded functions, @value{GDBN} evaluates the arguments
13309 and searches for a function whose signature matches the argument types,
13310 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
13311 Expressions, ,C@t{++} Expressions}, for details).
13312 If it cannot find a match, it emits a message.
13314 @item set overload-resolution off
13315 Disable overload resolution for C@t{++} expression evaluation. For
13316 overloaded functions that are not class member functions, @value{GDBN}
13317 chooses the first function of the specified name that it finds in the
13318 symbol table, whether or not its arguments are of the correct type. For
13319 overloaded functions that are class member functions, @value{GDBN}
13320 searches for a function whose signature @emph{exactly} matches the
13323 @kindex show overload-resolution
13324 @item show overload-resolution
13325 Show the current setting of overload resolution.
13327 @item @r{Overloaded symbol names}
13328 You can specify a particular definition of an overloaded symbol, using
13329 the same notation that is used to declare such symbols in C@t{++}: type
13330 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
13331 also use the @value{GDBN} command-line word completion facilities to list the
13332 available choices, or to finish the type list for you.
13333 @xref{Completion,, Command Completion}, for details on how to do this.
13336 @node Decimal Floating Point
13337 @subsubsection Decimal Floating Point format
13338 @cindex decimal floating point format
13340 @value{GDBN} can examine, set and perform computations with numbers in
13341 decimal floating point format, which in the C language correspond to the
13342 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
13343 specified by the extension to support decimal floating-point arithmetic.
13345 There are two encodings in use, depending on the architecture: BID (Binary
13346 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
13347 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
13350 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
13351 to manipulate decimal floating point numbers, it is not possible to convert
13352 (using a cast, for example) integers wider than 32-bit to decimal float.
13354 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
13355 point computations, error checking in decimal float operations ignores
13356 underflow, overflow and divide by zero exceptions.
13358 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
13359 to inspect @code{_Decimal128} values stored in floating point registers.
13360 See @ref{PowerPC,,PowerPC} for more details.
13366 @value{GDBN} can be used to debug programs written in D and compiled with
13367 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
13368 specific feature --- dynamic arrays.
13373 @cindex Go (programming language)
13374 @value{GDBN} can be used to debug programs written in Go and compiled with
13375 @file{gccgo} or @file{6g} compilers.
13377 Here is a summary of the Go-specific features and restrictions:
13380 @cindex current Go package
13381 @item The current Go package
13382 The name of the current package does not need to be specified when
13383 specifying global variables and functions.
13385 For example, given the program:
13389 var myglob = "Shall we?"
13395 When stopped inside @code{main} either of these work:
13399 (gdb) p main.myglob
13402 @cindex builtin Go types
13403 @item Builtin Go types
13404 The @code{string} type is recognized by @value{GDBN} and is printed
13407 @cindex builtin Go functions
13408 @item Builtin Go functions
13409 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
13410 function and handles it internally.
13412 @cindex restrictions on Go expressions
13413 @item Restrictions on Go expressions
13414 All Go operators are supported except @code{&^}.
13415 The Go @code{_} ``blank identifier'' is not supported.
13416 Automatic dereferencing of pointers is not supported.
13420 @subsection Objective-C
13422 @cindex Objective-C
13423 This section provides information about some commands and command
13424 options that are useful for debugging Objective-C code. See also
13425 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
13426 few more commands specific to Objective-C support.
13429 * Method Names in Commands::
13430 * The Print Command with Objective-C::
13433 @node Method Names in Commands
13434 @subsubsection Method Names in Commands
13436 The following commands have been extended to accept Objective-C method
13437 names as line specifications:
13439 @kindex clear@r{, and Objective-C}
13440 @kindex break@r{, and Objective-C}
13441 @kindex info line@r{, and Objective-C}
13442 @kindex jump@r{, and Objective-C}
13443 @kindex list@r{, and Objective-C}
13447 @item @code{info line}
13452 A fully qualified Objective-C method name is specified as
13455 -[@var{Class} @var{methodName}]
13458 where the minus sign is used to indicate an instance method and a
13459 plus sign (not shown) is used to indicate a class method. The class
13460 name @var{Class} and method name @var{methodName} are enclosed in
13461 brackets, similar to the way messages are specified in Objective-C
13462 source code. For example, to set a breakpoint at the @code{create}
13463 instance method of class @code{Fruit} in the program currently being
13467 break -[Fruit create]
13470 To list ten program lines around the @code{initialize} class method,
13474 list +[NSText initialize]
13477 In the current version of @value{GDBN}, the plus or minus sign is
13478 required. In future versions of @value{GDBN}, the plus or minus
13479 sign will be optional, but you can use it to narrow the search. It
13480 is also possible to specify just a method name:
13486 You must specify the complete method name, including any colons. If
13487 your program's source files contain more than one @code{create} method,
13488 you'll be presented with a numbered list of classes that implement that
13489 method. Indicate your choice by number, or type @samp{0} to exit if
13492 As another example, to clear a breakpoint established at the
13493 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
13496 clear -[NSWindow makeKeyAndOrderFront:]
13499 @node The Print Command with Objective-C
13500 @subsubsection The Print Command With Objective-C
13501 @cindex Objective-C, print objects
13502 @kindex print-object
13503 @kindex po @r{(@code{print-object})}
13505 The print command has also been extended to accept methods. For example:
13508 print -[@var{object} hash]
13511 @cindex print an Objective-C object description
13512 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
13514 will tell @value{GDBN} to send the @code{hash} message to @var{object}
13515 and print the result. Also, an additional command has been added,
13516 @code{print-object} or @code{po} for short, which is meant to print
13517 the description of an object. However, this command may only work
13518 with certain Objective-C libraries that have a particular hook
13519 function, @code{_NSPrintForDebugger}, defined.
13522 @subsection OpenCL C
13525 This section provides information about @value{GDBN}s OpenCL C support.
13528 * OpenCL C Datatypes::
13529 * OpenCL C Expressions::
13530 * OpenCL C Operators::
13533 @node OpenCL C Datatypes
13534 @subsubsection OpenCL C Datatypes
13536 @cindex OpenCL C Datatypes
13537 @value{GDBN} supports the builtin scalar and vector datatypes specified
13538 by OpenCL 1.1. In addition the half- and double-precision floating point
13539 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
13540 extensions are also known to @value{GDBN}.
13542 @node OpenCL C Expressions
13543 @subsubsection OpenCL C Expressions
13545 @cindex OpenCL C Expressions
13546 @value{GDBN} supports accesses to vector components including the access as
13547 lvalue where possible. Since OpenCL C is based on C99 most C expressions
13548 supported by @value{GDBN} can be used as well.
13550 @node OpenCL C Operators
13551 @subsubsection OpenCL C Operators
13553 @cindex OpenCL C Operators
13554 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
13558 @subsection Fortran
13559 @cindex Fortran-specific support in @value{GDBN}
13561 @value{GDBN} can be used to debug programs written in Fortran, but it
13562 currently supports only the features of Fortran 77 language.
13564 @cindex trailing underscore, in Fortran symbols
13565 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
13566 among them) append an underscore to the names of variables and
13567 functions. When you debug programs compiled by those compilers, you
13568 will need to refer to variables and functions with a trailing
13572 * Fortran Operators:: Fortran operators and expressions
13573 * Fortran Defaults:: Default settings for Fortran
13574 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
13577 @node Fortran Operators
13578 @subsubsection Fortran Operators and Expressions
13580 @cindex Fortran operators and expressions
13582 Operators must be defined on values of specific types. For instance,
13583 @code{+} is defined on numbers, but not on characters or other non-
13584 arithmetic types. Operators are often defined on groups of types.
13588 The exponentiation operator. It raises the first operand to the power
13592 The range operator. Normally used in the form of array(low:high) to
13593 represent a section of array.
13596 The access component operator. Normally used to access elements in derived
13597 types. Also suitable for unions. As unions aren't part of regular Fortran,
13598 this can only happen when accessing a register that uses a gdbarch-defined
13602 @node Fortran Defaults
13603 @subsubsection Fortran Defaults
13605 @cindex Fortran Defaults
13607 Fortran symbols are usually case-insensitive, so @value{GDBN} by
13608 default uses case-insensitive matches for Fortran symbols. You can
13609 change that with the @samp{set case-insensitive} command, see
13610 @ref{Symbols}, for the details.
13612 @node Special Fortran Commands
13613 @subsubsection Special Fortran Commands
13615 @cindex Special Fortran commands
13617 @value{GDBN} has some commands to support Fortran-specific features,
13618 such as displaying common blocks.
13621 @cindex @code{COMMON} blocks, Fortran
13622 @kindex info common
13623 @item info common @r{[}@var{common-name}@r{]}
13624 This command prints the values contained in the Fortran @code{COMMON}
13625 block whose name is @var{common-name}. With no argument, the names of
13626 all @code{COMMON} blocks visible at the current program location are
13633 @cindex Pascal support in @value{GDBN}, limitations
13634 Debugging Pascal programs which use sets, subranges, file variables, or
13635 nested functions does not currently work. @value{GDBN} does not support
13636 entering expressions, printing values, or similar features using Pascal
13639 The Pascal-specific command @code{set print pascal_static-members}
13640 controls whether static members of Pascal objects are displayed.
13641 @xref{Print Settings, pascal_static-members}.
13644 @subsection Modula-2
13646 @cindex Modula-2, @value{GDBN} support
13648 The extensions made to @value{GDBN} to support Modula-2 only support
13649 output from the @sc{gnu} Modula-2 compiler (which is currently being
13650 developed). Other Modula-2 compilers are not currently supported, and
13651 attempting to debug executables produced by them is most likely
13652 to give an error as @value{GDBN} reads in the executable's symbol
13655 @cindex expressions in Modula-2
13657 * M2 Operators:: Built-in operators
13658 * Built-In Func/Proc:: Built-in functions and procedures
13659 * M2 Constants:: Modula-2 constants
13660 * M2 Types:: Modula-2 types
13661 * M2 Defaults:: Default settings for Modula-2
13662 * Deviations:: Deviations from standard Modula-2
13663 * M2 Checks:: Modula-2 type and range checks
13664 * M2 Scope:: The scope operators @code{::} and @code{.}
13665 * GDB/M2:: @value{GDBN} and Modula-2
13669 @subsubsection Operators
13670 @cindex Modula-2 operators
13672 Operators must be defined on values of specific types. For instance,
13673 @code{+} is defined on numbers, but not on structures. Operators are
13674 often defined on groups of types. For the purposes of Modula-2, the
13675 following definitions hold:
13680 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
13684 @emph{Character types} consist of @code{CHAR} and its subranges.
13687 @emph{Floating-point types} consist of @code{REAL}.
13690 @emph{Pointer types} consist of anything declared as @code{POINTER TO
13694 @emph{Scalar types} consist of all of the above.
13697 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
13700 @emph{Boolean types} consist of @code{BOOLEAN}.
13704 The following operators are supported, and appear in order of
13705 increasing precedence:
13709 Function argument or array index separator.
13712 Assignment. The value of @var{var} @code{:=} @var{value} is
13716 Less than, greater than on integral, floating-point, or enumerated
13720 Less than or equal to, greater than or equal to
13721 on integral, floating-point and enumerated types, or set inclusion on
13722 set types. Same precedence as @code{<}.
13724 @item =@r{, }<>@r{, }#
13725 Equality and two ways of expressing inequality, valid on scalar types.
13726 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
13727 available for inequality, since @code{#} conflicts with the script
13731 Set membership. Defined on set types and the types of their members.
13732 Same precedence as @code{<}.
13735 Boolean disjunction. Defined on boolean types.
13738 Boolean conjunction. Defined on boolean types.
13741 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
13744 Addition and subtraction on integral and floating-point types, or union
13745 and difference on set types.
13748 Multiplication on integral and floating-point types, or set intersection
13752 Division on floating-point types, or symmetric set difference on set
13753 types. Same precedence as @code{*}.
13756 Integer division and remainder. Defined on integral types. Same
13757 precedence as @code{*}.
13760 Negative. Defined on @code{INTEGER} and @code{REAL} data.
13763 Pointer dereferencing. Defined on pointer types.
13766 Boolean negation. Defined on boolean types. Same precedence as
13770 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
13771 precedence as @code{^}.
13774 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
13777 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
13781 @value{GDBN} and Modula-2 scope operators.
13785 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
13786 treats the use of the operator @code{IN}, or the use of operators
13787 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
13788 @code{<=}, and @code{>=} on sets as an error.
13792 @node Built-In Func/Proc
13793 @subsubsection Built-in Functions and Procedures
13794 @cindex Modula-2 built-ins
13796 Modula-2 also makes available several built-in procedures and functions.
13797 In describing these, the following metavariables are used:
13802 represents an @code{ARRAY} variable.
13805 represents a @code{CHAR} constant or variable.
13808 represents a variable or constant of integral type.
13811 represents an identifier that belongs to a set. Generally used in the
13812 same function with the metavariable @var{s}. The type of @var{s} should
13813 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
13816 represents a variable or constant of integral or floating-point type.
13819 represents a variable or constant of floating-point type.
13825 represents a variable.
13828 represents a variable or constant of one of many types. See the
13829 explanation of the function for details.
13832 All Modula-2 built-in procedures also return a result, described below.
13836 Returns the absolute value of @var{n}.
13839 If @var{c} is a lower case letter, it returns its upper case
13840 equivalent, otherwise it returns its argument.
13843 Returns the character whose ordinal value is @var{i}.
13846 Decrements the value in the variable @var{v} by one. Returns the new value.
13848 @item DEC(@var{v},@var{i})
13849 Decrements the value in the variable @var{v} by @var{i}. Returns the
13852 @item EXCL(@var{m},@var{s})
13853 Removes the element @var{m} from the set @var{s}. Returns the new
13856 @item FLOAT(@var{i})
13857 Returns the floating point equivalent of the integer @var{i}.
13859 @item HIGH(@var{a})
13860 Returns the index of the last member of @var{a}.
13863 Increments the value in the variable @var{v} by one. Returns the new value.
13865 @item INC(@var{v},@var{i})
13866 Increments the value in the variable @var{v} by @var{i}. Returns the
13869 @item INCL(@var{m},@var{s})
13870 Adds the element @var{m} to the set @var{s} if it is not already
13871 there. Returns the new set.
13874 Returns the maximum value of the type @var{t}.
13877 Returns the minimum value of the type @var{t}.
13880 Returns boolean TRUE if @var{i} is an odd number.
13883 Returns the ordinal value of its argument. For example, the ordinal
13884 value of a character is its @sc{ascii} value (on machines supporting the
13885 @sc{ascii} character set). @var{x} must be of an ordered type, which include
13886 integral, character and enumerated types.
13888 @item SIZE(@var{x})
13889 Returns the size of its argument. @var{x} can be a variable or a type.
13891 @item TRUNC(@var{r})
13892 Returns the integral part of @var{r}.
13894 @item TSIZE(@var{x})
13895 Returns the size of its argument. @var{x} can be a variable or a type.
13897 @item VAL(@var{t},@var{i})
13898 Returns the member of the type @var{t} whose ordinal value is @var{i}.
13902 @emph{Warning:} Sets and their operations are not yet supported, so
13903 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
13907 @cindex Modula-2 constants
13909 @subsubsection Constants
13911 @value{GDBN} allows you to express the constants of Modula-2 in the following
13917 Integer constants are simply a sequence of digits. When used in an
13918 expression, a constant is interpreted to be type-compatible with the
13919 rest of the expression. Hexadecimal integers are specified by a
13920 trailing @samp{H}, and octal integers by a trailing @samp{B}.
13923 Floating point constants appear as a sequence of digits, followed by a
13924 decimal point and another sequence of digits. An optional exponent can
13925 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
13926 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
13927 digits of the floating point constant must be valid decimal (base 10)
13931 Character constants consist of a single character enclosed by a pair of
13932 like quotes, either single (@code{'}) or double (@code{"}). They may
13933 also be expressed by their ordinal value (their @sc{ascii} value, usually)
13934 followed by a @samp{C}.
13937 String constants consist of a sequence of characters enclosed by a
13938 pair of like quotes, either single (@code{'}) or double (@code{"}).
13939 Escape sequences in the style of C are also allowed. @xref{C
13940 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
13944 Enumerated constants consist of an enumerated identifier.
13947 Boolean constants consist of the identifiers @code{TRUE} and
13951 Pointer constants consist of integral values only.
13954 Set constants are not yet supported.
13958 @subsubsection Modula-2 Types
13959 @cindex Modula-2 types
13961 Currently @value{GDBN} can print the following data types in Modula-2
13962 syntax: array types, record types, set types, pointer types, procedure
13963 types, enumerated types, subrange types and base types. You can also
13964 print the contents of variables declared using these type.
13965 This section gives a number of simple source code examples together with
13966 sample @value{GDBN} sessions.
13968 The first example contains the following section of code:
13977 and you can request @value{GDBN} to interrogate the type and value of
13978 @code{r} and @code{s}.
13981 (@value{GDBP}) print s
13983 (@value{GDBP}) ptype s
13985 (@value{GDBP}) print r
13987 (@value{GDBP}) ptype r
13992 Likewise if your source code declares @code{s} as:
13996 s: SET ['A'..'Z'] ;
14000 then you may query the type of @code{s} by:
14003 (@value{GDBP}) ptype s
14004 type = SET ['A'..'Z']
14008 Note that at present you cannot interactively manipulate set
14009 expressions using the debugger.
14011 The following example shows how you might declare an array in Modula-2
14012 and how you can interact with @value{GDBN} to print its type and contents:
14016 s: ARRAY [-10..10] OF CHAR ;
14020 (@value{GDBP}) ptype s
14021 ARRAY [-10..10] OF CHAR
14024 Note that the array handling is not yet complete and although the type
14025 is printed correctly, expression handling still assumes that all
14026 arrays have a lower bound of zero and not @code{-10} as in the example
14029 Here are some more type related Modula-2 examples:
14033 colour = (blue, red, yellow, green) ;
14034 t = [blue..yellow] ;
14042 The @value{GDBN} interaction shows how you can query the data type
14043 and value of a variable.
14046 (@value{GDBP}) print s
14048 (@value{GDBP}) ptype t
14049 type = [blue..yellow]
14053 In this example a Modula-2 array is declared and its contents
14054 displayed. Observe that the contents are written in the same way as
14055 their @code{C} counterparts.
14059 s: ARRAY [1..5] OF CARDINAL ;
14065 (@value{GDBP}) print s
14066 $1 = @{1, 0, 0, 0, 0@}
14067 (@value{GDBP}) ptype s
14068 type = ARRAY [1..5] OF CARDINAL
14071 The Modula-2 language interface to @value{GDBN} also understands
14072 pointer types as shown in this example:
14076 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
14083 and you can request that @value{GDBN} describes the type of @code{s}.
14086 (@value{GDBP}) ptype s
14087 type = POINTER TO ARRAY [1..5] OF CARDINAL
14090 @value{GDBN} handles compound types as we can see in this example.
14091 Here we combine array types, record types, pointer types and subrange
14102 myarray = ARRAY myrange OF CARDINAL ;
14103 myrange = [-2..2] ;
14105 s: POINTER TO ARRAY myrange OF foo ;
14109 and you can ask @value{GDBN} to describe the type of @code{s} as shown
14113 (@value{GDBP}) ptype s
14114 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
14117 f3 : ARRAY [-2..2] OF CARDINAL;
14122 @subsubsection Modula-2 Defaults
14123 @cindex Modula-2 defaults
14125 If type and range checking are set automatically by @value{GDBN}, they
14126 both default to @code{on} whenever the working language changes to
14127 Modula-2. This happens regardless of whether you or @value{GDBN}
14128 selected the working language.
14130 If you allow @value{GDBN} to set the language automatically, then entering
14131 code compiled from a file whose name ends with @file{.mod} sets the
14132 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
14133 Infer the Source Language}, for further details.
14136 @subsubsection Deviations from Standard Modula-2
14137 @cindex Modula-2, deviations from
14139 A few changes have been made to make Modula-2 programs easier to debug.
14140 This is done primarily via loosening its type strictness:
14144 Unlike in standard Modula-2, pointer constants can be formed by
14145 integers. This allows you to modify pointer variables during
14146 debugging. (In standard Modula-2, the actual address contained in a
14147 pointer variable is hidden from you; it can only be modified
14148 through direct assignment to another pointer variable or expression that
14149 returned a pointer.)
14152 C escape sequences can be used in strings and characters to represent
14153 non-printable characters. @value{GDBN} prints out strings with these
14154 escape sequences embedded. Single non-printable characters are
14155 printed using the @samp{CHR(@var{nnn})} format.
14158 The assignment operator (@code{:=}) returns the value of its right-hand
14162 All built-in procedures both modify @emph{and} return their argument.
14166 @subsubsection Modula-2 Type and Range Checks
14167 @cindex Modula-2 checks
14170 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
14173 @c FIXME remove warning when type/range checks added
14175 @value{GDBN} considers two Modula-2 variables type equivalent if:
14179 They are of types that have been declared equivalent via a @code{TYPE
14180 @var{t1} = @var{t2}} statement
14183 They have been declared on the same line. (Note: This is true of the
14184 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
14187 As long as type checking is enabled, any attempt to combine variables
14188 whose types are not equivalent is an error.
14190 Range checking is done on all mathematical operations, assignment, array
14191 index bounds, and all built-in functions and procedures.
14194 @subsubsection The Scope Operators @code{::} and @code{.}
14196 @cindex @code{.}, Modula-2 scope operator
14197 @cindex colon, doubled as scope operator
14199 @vindex colon-colon@r{, in Modula-2}
14200 @c Info cannot handle :: but TeX can.
14203 @vindex ::@r{, in Modula-2}
14206 There are a few subtle differences between the Modula-2 scope operator
14207 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
14212 @var{module} . @var{id}
14213 @var{scope} :: @var{id}
14217 where @var{scope} is the name of a module or a procedure,
14218 @var{module} the name of a module, and @var{id} is any declared
14219 identifier within your program, except another module.
14221 Using the @code{::} operator makes @value{GDBN} search the scope
14222 specified by @var{scope} for the identifier @var{id}. If it is not
14223 found in the specified scope, then @value{GDBN} searches all scopes
14224 enclosing the one specified by @var{scope}.
14226 Using the @code{.} operator makes @value{GDBN} search the current scope for
14227 the identifier specified by @var{id} that was imported from the
14228 definition module specified by @var{module}. With this operator, it is
14229 an error if the identifier @var{id} was not imported from definition
14230 module @var{module}, or if @var{id} is not an identifier in
14234 @subsubsection @value{GDBN} and Modula-2
14236 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
14237 Five subcommands of @code{set print} and @code{show print} apply
14238 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
14239 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
14240 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
14241 analogue in Modula-2.
14243 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
14244 with any language, is not useful with Modula-2. Its
14245 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
14246 created in Modula-2 as they can in C or C@t{++}. However, because an
14247 address can be specified by an integral constant, the construct
14248 @samp{@{@var{type}@}@var{adrexp}} is still useful.
14250 @cindex @code{#} in Modula-2
14251 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
14252 interpreted as the beginning of a comment. Use @code{<>} instead.
14258 The extensions made to @value{GDBN} for Ada only support
14259 output from the @sc{gnu} Ada (GNAT) compiler.
14260 Other Ada compilers are not currently supported, and
14261 attempting to debug executables produced by them is most likely
14265 @cindex expressions in Ada
14267 * Ada Mode Intro:: General remarks on the Ada syntax
14268 and semantics supported by Ada mode
14270 * Omissions from Ada:: Restrictions on the Ada expression syntax.
14271 * Additions to Ada:: Extensions of the Ada expression syntax.
14272 * Stopping Before Main Program:: Debugging the program during elaboration.
14273 * Ada Tasks:: Listing and setting breakpoints in tasks.
14274 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
14275 * Ravenscar Profile:: Tasking Support when using the Ravenscar
14277 * Ada Glitches:: Known peculiarities of Ada mode.
14280 @node Ada Mode Intro
14281 @subsubsection Introduction
14282 @cindex Ada mode, general
14284 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
14285 syntax, with some extensions.
14286 The philosophy behind the design of this subset is
14290 That @value{GDBN} should provide basic literals and access to operations for
14291 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
14292 leaving more sophisticated computations to subprograms written into the
14293 program (which therefore may be called from @value{GDBN}).
14296 That type safety and strict adherence to Ada language restrictions
14297 are not particularly important to the @value{GDBN} user.
14300 That brevity is important to the @value{GDBN} user.
14303 Thus, for brevity, the debugger acts as if all names declared in
14304 user-written packages are directly visible, even if they are not visible
14305 according to Ada rules, thus making it unnecessary to fully qualify most
14306 names with their packages, regardless of context. Where this causes
14307 ambiguity, @value{GDBN} asks the user's intent.
14309 The debugger will start in Ada mode if it detects an Ada main program.
14310 As for other languages, it will enter Ada mode when stopped in a program that
14311 was translated from an Ada source file.
14313 While in Ada mode, you may use `@t{--}' for comments. This is useful
14314 mostly for documenting command files. The standard @value{GDBN} comment
14315 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
14316 middle (to allow based literals).
14318 The debugger supports limited overloading. Given a subprogram call in which
14319 the function symbol has multiple definitions, it will use the number of
14320 actual parameters and some information about their types to attempt to narrow
14321 the set of definitions. It also makes very limited use of context, preferring
14322 procedures to functions in the context of the @code{call} command, and
14323 functions to procedures elsewhere.
14325 @node Omissions from Ada
14326 @subsubsection Omissions from Ada
14327 @cindex Ada, omissions from
14329 Here are the notable omissions from the subset:
14333 Only a subset of the attributes are supported:
14337 @t{'First}, @t{'Last}, and @t{'Length}
14338 on array objects (not on types and subtypes).
14341 @t{'Min} and @t{'Max}.
14344 @t{'Pos} and @t{'Val}.
14350 @t{'Range} on array objects (not subtypes), but only as the right
14351 operand of the membership (@code{in}) operator.
14354 @t{'Access}, @t{'Unchecked_Access}, and
14355 @t{'Unrestricted_Access} (a GNAT extension).
14363 @code{Characters.Latin_1} are not available and
14364 concatenation is not implemented. Thus, escape characters in strings are
14365 not currently available.
14368 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
14369 equality of representations. They will generally work correctly
14370 for strings and arrays whose elements have integer or enumeration types.
14371 They may not work correctly for arrays whose element
14372 types have user-defined equality, for arrays of real values
14373 (in particular, IEEE-conformant floating point, because of negative
14374 zeroes and NaNs), and for arrays whose elements contain unused bits with
14375 indeterminate values.
14378 The other component-by-component array operations (@code{and}, @code{or},
14379 @code{xor}, @code{not}, and relational tests other than equality)
14380 are not implemented.
14383 @cindex array aggregates (Ada)
14384 @cindex record aggregates (Ada)
14385 @cindex aggregates (Ada)
14386 There is limited support for array and record aggregates. They are
14387 permitted only on the right sides of assignments, as in these examples:
14390 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
14391 (@value{GDBP}) set An_Array := (1, others => 0)
14392 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
14393 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
14394 (@value{GDBP}) set A_Record := (1, "Peter", True);
14395 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
14399 discriminant's value by assigning an aggregate has an
14400 undefined effect if that discriminant is used within the record.
14401 However, you can first modify discriminants by directly assigning to
14402 them (which normally would not be allowed in Ada), and then performing an
14403 aggregate assignment. For example, given a variable @code{A_Rec}
14404 declared to have a type such as:
14407 type Rec (Len : Small_Integer := 0) is record
14409 Vals : IntArray (1 .. Len);
14413 you can assign a value with a different size of @code{Vals} with two
14417 (@value{GDBP}) set A_Rec.Len := 4
14418 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
14421 As this example also illustrates, @value{GDBN} is very loose about the usual
14422 rules concerning aggregates. You may leave out some of the
14423 components of an array or record aggregate (such as the @code{Len}
14424 component in the assignment to @code{A_Rec} above); they will retain their
14425 original values upon assignment. You may freely use dynamic values as
14426 indices in component associations. You may even use overlapping or
14427 redundant component associations, although which component values are
14428 assigned in such cases is not defined.
14431 Calls to dispatching subprograms are not implemented.
14434 The overloading algorithm is much more limited (i.e., less selective)
14435 than that of real Ada. It makes only limited use of the context in
14436 which a subexpression appears to resolve its meaning, and it is much
14437 looser in its rules for allowing type matches. As a result, some
14438 function calls will be ambiguous, and the user will be asked to choose
14439 the proper resolution.
14442 The @code{new} operator is not implemented.
14445 Entry calls are not implemented.
14448 Aside from printing, arithmetic operations on the native VAX floating-point
14449 formats are not supported.
14452 It is not possible to slice a packed array.
14455 The names @code{True} and @code{False}, when not part of a qualified name,
14456 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
14458 Should your program
14459 redefine these names in a package or procedure (at best a dubious practice),
14460 you will have to use fully qualified names to access their new definitions.
14463 @node Additions to Ada
14464 @subsubsection Additions to Ada
14465 @cindex Ada, deviations from
14467 As it does for other languages, @value{GDBN} makes certain generic
14468 extensions to Ada (@pxref{Expressions}):
14472 If the expression @var{E} is a variable residing in memory (typically
14473 a local variable or array element) and @var{N} is a positive integer,
14474 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
14475 @var{N}-1 adjacent variables following it in memory as an array. In
14476 Ada, this operator is generally not necessary, since its prime use is
14477 in displaying parts of an array, and slicing will usually do this in
14478 Ada. However, there are occasional uses when debugging programs in
14479 which certain debugging information has been optimized away.
14482 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
14483 appears in function or file @var{B}.'' When @var{B} is a file name,
14484 you must typically surround it in single quotes.
14487 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
14488 @var{type} that appears at address @var{addr}.''
14491 A name starting with @samp{$} is a convenience variable
14492 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
14495 In addition, @value{GDBN} provides a few other shortcuts and outright
14496 additions specific to Ada:
14500 The assignment statement is allowed as an expression, returning
14501 its right-hand operand as its value. Thus, you may enter
14504 (@value{GDBP}) set x := y + 3
14505 (@value{GDBP}) print A(tmp := y + 1)
14509 The semicolon is allowed as an ``operator,'' returning as its value
14510 the value of its right-hand operand.
14511 This allows, for example,
14512 complex conditional breaks:
14515 (@value{GDBP}) break f
14516 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
14520 Rather than use catenation and symbolic character names to introduce special
14521 characters into strings, one may instead use a special bracket notation,
14522 which is also used to print strings. A sequence of characters of the form
14523 @samp{["@var{XX}"]} within a string or character literal denotes the
14524 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
14525 sequence of characters @samp{["""]} also denotes a single quotation mark
14526 in strings. For example,
14528 "One line.["0a"]Next line.["0a"]"
14531 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
14535 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
14536 @t{'Max} is optional (and is ignored in any case). For example, it is valid
14540 (@value{GDBP}) print 'max(x, y)
14544 When printing arrays, @value{GDBN} uses positional notation when the
14545 array has a lower bound of 1, and uses a modified named notation otherwise.
14546 For example, a one-dimensional array of three integers with a lower bound
14547 of 3 might print as
14554 That is, in contrast to valid Ada, only the first component has a @code{=>}
14558 You may abbreviate attributes in expressions with any unique,
14559 multi-character subsequence of
14560 their names (an exact match gets preference).
14561 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
14562 in place of @t{a'length}.
14565 @cindex quoting Ada internal identifiers
14566 Since Ada is case-insensitive, the debugger normally maps identifiers you type
14567 to lower case. The GNAT compiler uses upper-case characters for
14568 some of its internal identifiers, which are normally of no interest to users.
14569 For the rare occasions when you actually have to look at them,
14570 enclose them in angle brackets to avoid the lower-case mapping.
14573 (@value{GDBP}) print <JMPBUF_SAVE>[0]
14577 Printing an object of class-wide type or dereferencing an
14578 access-to-class-wide value will display all the components of the object's
14579 specific type (as indicated by its run-time tag). Likewise, component
14580 selection on such a value will operate on the specific type of the
14585 @node Stopping Before Main Program
14586 @subsubsection Stopping at the Very Beginning
14588 @cindex breakpointing Ada elaboration code
14589 It is sometimes necessary to debug the program during elaboration, and
14590 before reaching the main procedure.
14591 As defined in the Ada Reference
14592 Manual, the elaboration code is invoked from a procedure called
14593 @code{adainit}. To run your program up to the beginning of
14594 elaboration, simply use the following two commands:
14595 @code{tbreak adainit} and @code{run}.
14598 @subsubsection Extensions for Ada Tasks
14599 @cindex Ada, tasking
14601 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
14602 @value{GDBN} provides the following task-related commands:
14607 This command shows a list of current Ada tasks, as in the following example:
14614 (@value{GDBP}) info tasks
14615 ID TID P-ID Pri State Name
14616 1 8088000 0 15 Child Activation Wait main_task
14617 2 80a4000 1 15 Accept Statement b
14618 3 809a800 1 15 Child Activation Wait a
14619 * 4 80ae800 3 15 Runnable c
14624 In this listing, the asterisk before the last task indicates it to be the
14625 task currently being inspected.
14629 Represents @value{GDBN}'s internal task number.
14635 The parent's task ID (@value{GDBN}'s internal task number).
14638 The base priority of the task.
14641 Current state of the task.
14645 The task has been created but has not been activated. It cannot be
14649 The task is not blocked for any reason known to Ada. (It may be waiting
14650 for a mutex, though.) It is conceptually "executing" in normal mode.
14653 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
14654 that were waiting on terminate alternatives have been awakened and have
14655 terminated themselves.
14657 @item Child Activation Wait
14658 The task is waiting for created tasks to complete activation.
14660 @item Accept Statement
14661 The task is waiting on an accept or selective wait statement.
14663 @item Waiting on entry call
14664 The task is waiting on an entry call.
14666 @item Async Select Wait
14667 The task is waiting to start the abortable part of an asynchronous
14671 The task is waiting on a select statement with only a delay
14674 @item Child Termination Wait
14675 The task is sleeping having completed a master within itself, and is
14676 waiting for the tasks dependent on that master to become terminated or
14677 waiting on a terminate Phase.
14679 @item Wait Child in Term Alt
14680 The task is sleeping waiting for tasks on terminate alternatives to
14681 finish terminating.
14683 @item Accepting RV with @var{taskno}
14684 The task is accepting a rendez-vous with the task @var{taskno}.
14688 Name of the task in the program.
14692 @kindex info task @var{taskno}
14693 @item info task @var{taskno}
14694 This command shows detailled informations on the specified task, as in
14695 the following example:
14700 (@value{GDBP}) info tasks
14701 ID TID P-ID Pri State Name
14702 1 8077880 0 15 Child Activation Wait main_task
14703 * 2 807c468 1 15 Runnable task_1
14704 (@value{GDBP}) info task 2
14705 Ada Task: 0x807c468
14708 Parent: 1 (main_task)
14714 @kindex task@r{ (Ada)}
14715 @cindex current Ada task ID
14716 This command prints the ID of the current task.
14722 (@value{GDBP}) info tasks
14723 ID TID P-ID Pri State Name
14724 1 8077870 0 15 Child Activation Wait main_task
14725 * 2 807c458 1 15 Runnable t
14726 (@value{GDBP}) task
14727 [Current task is 2]
14730 @item task @var{taskno}
14731 @cindex Ada task switching
14732 This command is like the @code{thread @var{threadno}}
14733 command (@pxref{Threads}). It switches the context of debugging
14734 from the current task to the given task.
14740 (@value{GDBP}) info tasks
14741 ID TID P-ID Pri State Name
14742 1 8077870 0 15 Child Activation Wait main_task
14743 * 2 807c458 1 15 Runnable t
14744 (@value{GDBP}) task 1
14745 [Switching to task 1]
14746 #0 0x8067726 in pthread_cond_wait ()
14748 #0 0x8067726 in pthread_cond_wait ()
14749 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
14750 #2 0x805cb63 in system.task_primitives.operations.sleep ()
14751 #3 0x806153e in system.tasking.stages.activate_tasks ()
14752 #4 0x804aacc in un () at un.adb:5
14755 @item break @var{linespec} task @var{taskno}
14756 @itemx break @var{linespec} task @var{taskno} if @dots{}
14757 @cindex breakpoints and tasks, in Ada
14758 @cindex task breakpoints, in Ada
14759 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
14760 These commands are like the @code{break @dots{} thread @dots{}}
14761 command (@pxref{Thread Stops}).
14762 @var{linespec} specifies source lines, as described
14763 in @ref{Specify Location}.
14765 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
14766 to specify that you only want @value{GDBN} to stop the program when a
14767 particular Ada task reaches this breakpoint. @var{taskno} is one of the
14768 numeric task identifiers assigned by @value{GDBN}, shown in the first
14769 column of the @samp{info tasks} display.
14771 If you do not specify @samp{task @var{taskno}} when you set a
14772 breakpoint, the breakpoint applies to @emph{all} tasks of your
14775 You can use the @code{task} qualifier on conditional breakpoints as
14776 well; in this case, place @samp{task @var{taskno}} before the
14777 breakpoint condition (before the @code{if}).
14785 (@value{GDBP}) info tasks
14786 ID TID P-ID Pri State Name
14787 1 140022020 0 15 Child Activation Wait main_task
14788 2 140045060 1 15 Accept/Select Wait t2
14789 3 140044840 1 15 Runnable t1
14790 * 4 140056040 1 15 Runnable t3
14791 (@value{GDBP}) b 15 task 2
14792 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
14793 (@value{GDBP}) cont
14798 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
14800 (@value{GDBP}) info tasks
14801 ID TID P-ID Pri State Name
14802 1 140022020 0 15 Child Activation Wait main_task
14803 * 2 140045060 1 15 Runnable t2
14804 3 140044840 1 15 Runnable t1
14805 4 140056040 1 15 Delay Sleep t3
14809 @node Ada Tasks and Core Files
14810 @subsubsection Tasking Support when Debugging Core Files
14811 @cindex Ada tasking and core file debugging
14813 When inspecting a core file, as opposed to debugging a live program,
14814 tasking support may be limited or even unavailable, depending on
14815 the platform being used.
14816 For instance, on x86-linux, the list of tasks is available, but task
14817 switching is not supported. On Tru64, however, task switching will work
14820 On certain platforms, including Tru64, the debugger needs to perform some
14821 memory writes in order to provide Ada tasking support. When inspecting
14822 a core file, this means that the core file must be opened with read-write
14823 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
14824 Under these circumstances, you should make a backup copy of the core
14825 file before inspecting it with @value{GDBN}.
14827 @node Ravenscar Profile
14828 @subsubsection Tasking Support when using the Ravenscar Profile
14829 @cindex Ravenscar Profile
14831 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
14832 specifically designed for systems with safety-critical real-time
14836 @kindex set ravenscar task-switching on
14837 @cindex task switching with program using Ravenscar Profile
14838 @item set ravenscar task-switching on
14839 Allows task switching when debugging a program that uses the Ravenscar
14840 Profile. This is the default.
14842 @kindex set ravenscar task-switching off
14843 @item set ravenscar task-switching off
14844 Turn off task switching when debugging a program that uses the Ravenscar
14845 Profile. This is mostly intended to disable the code that adds support
14846 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
14847 the Ravenscar runtime is preventing @value{GDBN} from working properly.
14848 To be effective, this command should be run before the program is started.
14850 @kindex show ravenscar task-switching
14851 @item show ravenscar task-switching
14852 Show whether it is possible to switch from task to task in a program
14853 using the Ravenscar Profile.
14858 @subsubsection Known Peculiarities of Ada Mode
14859 @cindex Ada, problems
14861 Besides the omissions listed previously (@pxref{Omissions from Ada}),
14862 we know of several problems with and limitations of Ada mode in
14864 some of which will be fixed with planned future releases of the debugger
14865 and the GNU Ada compiler.
14869 Static constants that the compiler chooses not to materialize as objects in
14870 storage are invisible to the debugger.
14873 Named parameter associations in function argument lists are ignored (the
14874 argument lists are treated as positional).
14877 Many useful library packages are currently invisible to the debugger.
14880 Fixed-point arithmetic, conversions, input, and output is carried out using
14881 floating-point arithmetic, and may give results that only approximate those on
14885 The GNAT compiler never generates the prefix @code{Standard} for any of
14886 the standard symbols defined by the Ada language. @value{GDBN} knows about
14887 this: it will strip the prefix from names when you use it, and will never
14888 look for a name you have so qualified among local symbols, nor match against
14889 symbols in other packages or subprograms. If you have
14890 defined entities anywhere in your program other than parameters and
14891 local variables whose simple names match names in @code{Standard},
14892 GNAT's lack of qualification here can cause confusion. When this happens,
14893 you can usually resolve the confusion
14894 by qualifying the problematic names with package
14895 @code{Standard} explicitly.
14898 Older versions of the compiler sometimes generate erroneous debugging
14899 information, resulting in the debugger incorrectly printing the value
14900 of affected entities. In some cases, the debugger is able to work
14901 around an issue automatically. In other cases, the debugger is able
14902 to work around the issue, but the work-around has to be specifically
14905 @kindex set ada trust-PAD-over-XVS
14906 @kindex show ada trust-PAD-over-XVS
14909 @item set ada trust-PAD-over-XVS on
14910 Configure GDB to strictly follow the GNAT encoding when computing the
14911 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
14912 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
14913 a complete description of the encoding used by the GNAT compiler).
14914 This is the default.
14916 @item set ada trust-PAD-over-XVS off
14917 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
14918 sometimes prints the wrong value for certain entities, changing @code{ada
14919 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
14920 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
14921 @code{off}, but this incurs a slight performance penalty, so it is
14922 recommended to leave this setting to @code{on} unless necessary.
14926 @node Unsupported Languages
14927 @section Unsupported Languages
14929 @cindex unsupported languages
14930 @cindex minimal language
14931 In addition to the other fully-supported programming languages,
14932 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
14933 It does not represent a real programming language, but provides a set
14934 of capabilities close to what the C or assembly languages provide.
14935 This should allow most simple operations to be performed while debugging
14936 an application that uses a language currently not supported by @value{GDBN}.
14938 If the language is set to @code{auto}, @value{GDBN} will automatically
14939 select this language if the current frame corresponds to an unsupported
14943 @chapter Examining the Symbol Table
14945 The commands described in this chapter allow you to inquire about the
14946 symbols (names of variables, functions and types) defined in your
14947 program. This information is inherent in the text of your program and
14948 does not change as your program executes. @value{GDBN} finds it in your
14949 program's symbol table, in the file indicated when you started @value{GDBN}
14950 (@pxref{File Options, ,Choosing Files}), or by one of the
14951 file-management commands (@pxref{Files, ,Commands to Specify Files}).
14953 @cindex symbol names
14954 @cindex names of symbols
14955 @cindex quoting names
14956 Occasionally, you may need to refer to symbols that contain unusual
14957 characters, which @value{GDBN} ordinarily treats as word delimiters. The
14958 most frequent case is in referring to static variables in other
14959 source files (@pxref{Variables,,Program Variables}). File names
14960 are recorded in object files as debugging symbols, but @value{GDBN} would
14961 ordinarily parse a typical file name, like @file{foo.c}, as the three words
14962 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
14963 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
14970 looks up the value of @code{x} in the scope of the file @file{foo.c}.
14973 @cindex case-insensitive symbol names
14974 @cindex case sensitivity in symbol names
14975 @kindex set case-sensitive
14976 @item set case-sensitive on
14977 @itemx set case-sensitive off
14978 @itemx set case-sensitive auto
14979 Normally, when @value{GDBN} looks up symbols, it matches their names
14980 with case sensitivity determined by the current source language.
14981 Occasionally, you may wish to control that. The command @code{set
14982 case-sensitive} lets you do that by specifying @code{on} for
14983 case-sensitive matches or @code{off} for case-insensitive ones. If
14984 you specify @code{auto}, case sensitivity is reset to the default
14985 suitable for the source language. The default is case-sensitive
14986 matches for all languages except for Fortran, for which the default is
14987 case-insensitive matches.
14989 @kindex show case-sensitive
14990 @item show case-sensitive
14991 This command shows the current setting of case sensitivity for symbols
14994 @kindex info address
14995 @cindex address of a symbol
14996 @item info address @var{symbol}
14997 Describe where the data for @var{symbol} is stored. For a register
14998 variable, this says which register it is kept in. For a non-register
14999 local variable, this prints the stack-frame offset at which the variable
15002 Note the contrast with @samp{print &@var{symbol}}, which does not work
15003 at all for a register variable, and for a stack local variable prints
15004 the exact address of the current instantiation of the variable.
15006 @kindex info symbol
15007 @cindex symbol from address
15008 @cindex closest symbol and offset for an address
15009 @item info symbol @var{addr}
15010 Print the name of a symbol which is stored at the address @var{addr}.
15011 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
15012 nearest symbol and an offset from it:
15015 (@value{GDBP}) info symbol 0x54320
15016 _initialize_vx + 396 in section .text
15020 This is the opposite of the @code{info address} command. You can use
15021 it to find out the name of a variable or a function given its address.
15023 For dynamically linked executables, the name of executable or shared
15024 library containing the symbol is also printed:
15027 (@value{GDBP}) info symbol 0x400225
15028 _start + 5 in section .text of /tmp/a.out
15029 (@value{GDBP}) info symbol 0x2aaaac2811cf
15030 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
15034 @item whatis [@var{arg}]
15035 Print the data type of @var{arg}, which can be either an expression
15036 or a name of a data type. With no argument, print the data type of
15037 @code{$}, the last value in the value history.
15039 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
15040 is not actually evaluated, and any side-effecting operations (such as
15041 assignments or function calls) inside it do not take place.
15043 If @var{arg} is a variable or an expression, @code{whatis} prints its
15044 literal type as it is used in the source code. If the type was
15045 defined using a @code{typedef}, @code{whatis} will @emph{not} print
15046 the data type underlying the @code{typedef}. If the type of the
15047 variable or the expression is a compound data type, such as
15048 @code{struct} or @code{class}, @code{whatis} never prints their
15049 fields or methods. It just prints the @code{struct}/@code{class}
15050 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
15051 such a compound data type, use @code{ptype}.
15053 If @var{arg} is a type name that was defined using @code{typedef},
15054 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
15055 Unrolling means that @code{whatis} will show the underlying type used
15056 in the @code{typedef} declaration of @var{arg}. However, if that
15057 underlying type is also a @code{typedef}, @code{whatis} will not
15060 For C code, the type names may also have the form @samp{class
15061 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
15062 @var{union-tag}} or @samp{enum @var{enum-tag}}.
15065 @item ptype [@var{arg}]
15066 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
15067 detailed description of the type, instead of just the name of the type.
15068 @xref{Expressions, ,Expressions}.
15070 Contrary to @code{whatis}, @code{ptype} always unrolls any
15071 @code{typedef}s in its argument declaration, whether the argument is
15072 a variable, expression, or a data type. This means that @code{ptype}
15073 of a variable or an expression will not print literally its type as
15074 present in the source code---use @code{whatis} for that. @code{typedef}s at
15075 the pointer or reference targets are also unrolled. Only @code{typedef}s of
15076 fields, methods and inner @code{class typedef}s of @code{struct}s,
15077 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
15079 For example, for this variable declaration:
15082 typedef double real_t;
15083 struct complex @{ real_t real; double imag; @};
15084 typedef struct complex complex_t;
15086 real_t *real_pointer_var;
15090 the two commands give this output:
15094 (@value{GDBP}) whatis var
15096 (@value{GDBP}) ptype var
15097 type = struct complex @{
15101 (@value{GDBP}) whatis complex_t
15102 type = struct complex
15103 (@value{GDBP}) whatis struct complex
15104 type = struct complex
15105 (@value{GDBP}) ptype struct complex
15106 type = struct complex @{
15110 (@value{GDBP}) whatis real_pointer_var
15112 (@value{GDBP}) ptype real_pointer_var
15118 As with @code{whatis}, using @code{ptype} without an argument refers to
15119 the type of @code{$}, the last value in the value history.
15121 @cindex incomplete type
15122 Sometimes, programs use opaque data types or incomplete specifications
15123 of complex data structure. If the debug information included in the
15124 program does not allow @value{GDBN} to display a full declaration of
15125 the data type, it will say @samp{<incomplete type>}. For example,
15126 given these declarations:
15130 struct foo *fooptr;
15134 but no definition for @code{struct foo} itself, @value{GDBN} will say:
15137 (@value{GDBP}) ptype foo
15138 $1 = <incomplete type>
15142 ``Incomplete type'' is C terminology for data types that are not
15143 completely specified.
15146 @item info types @var{regexp}
15148 Print a brief description of all types whose names match the regular
15149 expression @var{regexp} (or all types in your program, if you supply
15150 no argument). Each complete typename is matched as though it were a
15151 complete line; thus, @samp{i type value} gives information on all
15152 types in your program whose names include the string @code{value}, but
15153 @samp{i type ^value$} gives information only on types whose complete
15154 name is @code{value}.
15156 This command differs from @code{ptype} in two ways: first, like
15157 @code{whatis}, it does not print a detailed description; second, it
15158 lists all source files where a type is defined.
15161 @cindex local variables
15162 @item info scope @var{location}
15163 List all the variables local to a particular scope. This command
15164 accepts a @var{location} argument---a function name, a source line, or
15165 an address preceded by a @samp{*}, and prints all the variables local
15166 to the scope defined by that location. (@xref{Specify Location}, for
15167 details about supported forms of @var{location}.) For example:
15170 (@value{GDBP}) @b{info scope command_line_handler}
15171 Scope for command_line_handler:
15172 Symbol rl is an argument at stack/frame offset 8, length 4.
15173 Symbol linebuffer is in static storage at address 0x150a18, length 4.
15174 Symbol linelength is in static storage at address 0x150a1c, length 4.
15175 Symbol p is a local variable in register $esi, length 4.
15176 Symbol p1 is a local variable in register $ebx, length 4.
15177 Symbol nline is a local variable in register $edx, length 4.
15178 Symbol repeat is a local variable at frame offset -8, length 4.
15182 This command is especially useful for determining what data to collect
15183 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
15186 @kindex info source
15188 Show information about the current source file---that is, the source file for
15189 the function containing the current point of execution:
15192 the name of the source file, and the directory containing it,
15194 the directory it was compiled in,
15196 its length, in lines,
15198 which programming language it is written in,
15200 whether the executable includes debugging information for that file, and
15201 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
15203 whether the debugging information includes information about
15204 preprocessor macros.
15208 @kindex info sources
15210 Print the names of all source files in your program for which there is
15211 debugging information, organized into two lists: files whose symbols
15212 have already been read, and files whose symbols will be read when needed.
15214 @kindex info functions
15215 @item info functions
15216 Print the names and data types of all defined functions.
15218 @item info functions @var{regexp}
15219 Print the names and data types of all defined functions
15220 whose names contain a match for regular expression @var{regexp}.
15221 Thus, @samp{info fun step} finds all functions whose names
15222 include @code{step}; @samp{info fun ^step} finds those whose names
15223 start with @code{step}. If a function name contains characters
15224 that conflict with the regular expression language (e.g.@:
15225 @samp{operator*()}), they may be quoted with a backslash.
15227 @kindex info variables
15228 @item info variables
15229 Print the names and data types of all variables that are defined
15230 outside of functions (i.e.@: excluding local variables).
15232 @item info variables @var{regexp}
15233 Print the names and data types of all variables (except for local
15234 variables) whose names contain a match for regular expression
15237 @kindex info classes
15238 @cindex Objective-C, classes and selectors
15240 @itemx info classes @var{regexp}
15241 Display all Objective-C classes in your program, or
15242 (with the @var{regexp} argument) all those matching a particular regular
15245 @kindex info selectors
15246 @item info selectors
15247 @itemx info selectors @var{regexp}
15248 Display all Objective-C selectors in your program, or
15249 (with the @var{regexp} argument) all those matching a particular regular
15253 This was never implemented.
15254 @kindex info methods
15256 @itemx info methods @var{regexp}
15257 The @code{info methods} command permits the user to examine all defined
15258 methods within C@t{++} program, or (with the @var{regexp} argument) a
15259 specific set of methods found in the various C@t{++} classes. Many
15260 C@t{++} classes provide a large number of methods. Thus, the output
15261 from the @code{ptype} command can be overwhelming and hard to use. The
15262 @code{info-methods} command filters the methods, printing only those
15263 which match the regular-expression @var{regexp}.
15266 @cindex opaque data types
15267 @kindex set opaque-type-resolution
15268 @item set opaque-type-resolution on
15269 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
15270 declared as a pointer to a @code{struct}, @code{class}, or
15271 @code{union}---for example, @code{struct MyType *}---that is used in one
15272 source file although the full declaration of @code{struct MyType} is in
15273 another source file. The default is on.
15275 A change in the setting of this subcommand will not take effect until
15276 the next time symbols for a file are loaded.
15278 @item set opaque-type-resolution off
15279 Tell @value{GDBN} not to resolve opaque types. In this case, the type
15280 is printed as follows:
15282 @{<no data fields>@}
15285 @kindex show opaque-type-resolution
15286 @item show opaque-type-resolution
15287 Show whether opaque types are resolved or not.
15289 @kindex maint print symbols
15290 @cindex symbol dump
15291 @kindex maint print psymbols
15292 @cindex partial symbol dump
15293 @item maint print symbols @var{filename}
15294 @itemx maint print psymbols @var{filename}
15295 @itemx maint print msymbols @var{filename}
15296 Write a dump of debugging symbol data into the file @var{filename}.
15297 These commands are used to debug the @value{GDBN} symbol-reading code. Only
15298 symbols with debugging data are included. If you use @samp{maint print
15299 symbols}, @value{GDBN} includes all the symbols for which it has already
15300 collected full details: that is, @var{filename} reflects symbols for
15301 only those files whose symbols @value{GDBN} has read. You can use the
15302 command @code{info sources} to find out which files these are. If you
15303 use @samp{maint print psymbols} instead, the dump shows information about
15304 symbols that @value{GDBN} only knows partially---that is, symbols defined in
15305 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
15306 @samp{maint print msymbols} dumps just the minimal symbol information
15307 required for each object file from which @value{GDBN} has read some symbols.
15308 @xref{Files, ,Commands to Specify Files}, for a discussion of how
15309 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
15311 @kindex maint info symtabs
15312 @kindex maint info psymtabs
15313 @cindex listing @value{GDBN}'s internal symbol tables
15314 @cindex symbol tables, listing @value{GDBN}'s internal
15315 @cindex full symbol tables, listing @value{GDBN}'s internal
15316 @cindex partial symbol tables, listing @value{GDBN}'s internal
15317 @item maint info symtabs @r{[} @var{regexp} @r{]}
15318 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
15320 List the @code{struct symtab} or @code{struct partial_symtab}
15321 structures whose names match @var{regexp}. If @var{regexp} is not
15322 given, list them all. The output includes expressions which you can
15323 copy into a @value{GDBN} debugging this one to examine a particular
15324 structure in more detail. For example:
15327 (@value{GDBP}) maint info psymtabs dwarf2read
15328 @{ objfile /home/gnu/build/gdb/gdb
15329 ((struct objfile *) 0x82e69d0)
15330 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
15331 ((struct partial_symtab *) 0x8474b10)
15334 text addresses 0x814d3c8 -- 0x8158074
15335 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
15336 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
15337 dependencies (none)
15340 (@value{GDBP}) maint info symtabs
15344 We see that there is one partial symbol table whose filename contains
15345 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
15346 and we see that @value{GDBN} has not read in any symtabs yet at all.
15347 If we set a breakpoint on a function, that will cause @value{GDBN} to
15348 read the symtab for the compilation unit containing that function:
15351 (@value{GDBP}) break dwarf2_psymtab_to_symtab
15352 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
15354 (@value{GDBP}) maint info symtabs
15355 @{ objfile /home/gnu/build/gdb/gdb
15356 ((struct objfile *) 0x82e69d0)
15357 @{ symtab /home/gnu/src/gdb/dwarf2read.c
15358 ((struct symtab *) 0x86c1f38)
15361 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
15362 linetable ((struct linetable *) 0x8370fa0)
15363 debugformat DWARF 2
15372 @chapter Altering Execution
15374 Once you think you have found an error in your program, you might want to
15375 find out for certain whether correcting the apparent error would lead to
15376 correct results in the rest of the run. You can find the answer by
15377 experiment, using the @value{GDBN} features for altering execution of the
15380 For example, you can store new values into variables or memory
15381 locations, give your program a signal, restart it at a different
15382 address, or even return prematurely from a function.
15385 * Assignment:: Assignment to variables
15386 * Jumping:: Continuing at a different address
15387 * Signaling:: Giving your program a signal
15388 * Returning:: Returning from a function
15389 * Calling:: Calling your program's functions
15390 * Patching:: Patching your program
15394 @section Assignment to Variables
15397 @cindex setting variables
15398 To alter the value of a variable, evaluate an assignment expression.
15399 @xref{Expressions, ,Expressions}. For example,
15406 stores the value 4 into the variable @code{x}, and then prints the
15407 value of the assignment expression (which is 4).
15408 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
15409 information on operators in supported languages.
15411 @kindex set variable
15412 @cindex variables, setting
15413 If you are not interested in seeing the value of the assignment, use the
15414 @code{set} command instead of the @code{print} command. @code{set} is
15415 really the same as @code{print} except that the expression's value is
15416 not printed and is not put in the value history (@pxref{Value History,
15417 ,Value History}). The expression is evaluated only for its effects.
15419 If the beginning of the argument string of the @code{set} command
15420 appears identical to a @code{set} subcommand, use the @code{set
15421 variable} command instead of just @code{set}. This command is identical
15422 to @code{set} except for its lack of subcommands. For example, if your
15423 program has a variable @code{width}, you get an error if you try to set
15424 a new value with just @samp{set width=13}, because @value{GDBN} has the
15425 command @code{set width}:
15428 (@value{GDBP}) whatis width
15430 (@value{GDBP}) p width
15432 (@value{GDBP}) set width=47
15433 Invalid syntax in expression.
15437 The invalid expression, of course, is @samp{=47}. In
15438 order to actually set the program's variable @code{width}, use
15441 (@value{GDBP}) set var width=47
15444 Because the @code{set} command has many subcommands that can conflict
15445 with the names of program variables, it is a good idea to use the
15446 @code{set variable} command instead of just @code{set}. For example, if
15447 your program has a variable @code{g}, you run into problems if you try
15448 to set a new value with just @samp{set g=4}, because @value{GDBN} has
15449 the command @code{set gnutarget}, abbreviated @code{set g}:
15453 (@value{GDBP}) whatis g
15457 (@value{GDBP}) set g=4
15461 The program being debugged has been started already.
15462 Start it from the beginning? (y or n) y
15463 Starting program: /home/smith/cc_progs/a.out
15464 "/home/smith/cc_progs/a.out": can't open to read symbols:
15465 Invalid bfd target.
15466 (@value{GDBP}) show g
15467 The current BFD target is "=4".
15472 The program variable @code{g} did not change, and you silently set the
15473 @code{gnutarget} to an invalid value. In order to set the variable
15477 (@value{GDBP}) set var g=4
15480 @value{GDBN} allows more implicit conversions in assignments than C; you can
15481 freely store an integer value into a pointer variable or vice versa,
15482 and you can convert any structure to any other structure that is the
15483 same length or shorter.
15484 @comment FIXME: how do structs align/pad in these conversions?
15485 @comment /doc@cygnus.com 18dec1990
15487 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
15488 construct to generate a value of specified type at a specified address
15489 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
15490 to memory location @code{0x83040} as an integer (which implies a certain size
15491 and representation in memory), and
15494 set @{int@}0x83040 = 4
15498 stores the value 4 into that memory location.
15501 @section Continuing at a Different Address
15503 Ordinarily, when you continue your program, you do so at the place where
15504 it stopped, with the @code{continue} command. You can instead continue at
15505 an address of your own choosing, with the following commands:
15509 @item jump @var{linespec}
15510 @itemx jump @var{location}
15511 Resume execution at line @var{linespec} or at address given by
15512 @var{location}. Execution stops again immediately if there is a
15513 breakpoint there. @xref{Specify Location}, for a description of the
15514 different forms of @var{linespec} and @var{location}. It is common
15515 practice to use the @code{tbreak} command in conjunction with
15516 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
15518 The @code{jump} command does not change the current stack frame, or
15519 the stack pointer, or the contents of any memory location or any
15520 register other than the program counter. If line @var{linespec} is in
15521 a different function from the one currently executing, the results may
15522 be bizarre if the two functions expect different patterns of arguments or
15523 of local variables. For this reason, the @code{jump} command requests
15524 confirmation if the specified line is not in the function currently
15525 executing. However, even bizarre results are predictable if you are
15526 well acquainted with the machine-language code of your program.
15529 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
15530 On many systems, you can get much the same effect as the @code{jump}
15531 command by storing a new value into the register @code{$pc}. The
15532 difference is that this does not start your program running; it only
15533 changes the address of where it @emph{will} run when you continue. For
15541 makes the next @code{continue} command or stepping command execute at
15542 address @code{0x485}, rather than at the address where your program stopped.
15543 @xref{Continuing and Stepping, ,Continuing and Stepping}.
15545 The most common occasion to use the @code{jump} command is to back
15546 up---perhaps with more breakpoints set---over a portion of a program
15547 that has already executed, in order to examine its execution in more
15552 @section Giving your Program a Signal
15553 @cindex deliver a signal to a program
15557 @item signal @var{signal}
15558 Resume execution where your program stopped, but immediately give it the
15559 signal @var{signal}. @var{signal} can be the name or the number of a
15560 signal. For example, on many systems @code{signal 2} and @code{signal
15561 SIGINT} are both ways of sending an interrupt signal.
15563 Alternatively, if @var{signal} is zero, continue execution without
15564 giving a signal. This is useful when your program stopped on account of
15565 a signal and would ordinary see the signal when resumed with the
15566 @code{continue} command; @samp{signal 0} causes it to resume without a
15569 @code{signal} does not repeat when you press @key{RET} a second time
15570 after executing the command.
15574 Invoking the @code{signal} command is not the same as invoking the
15575 @code{kill} utility from the shell. Sending a signal with @code{kill}
15576 causes @value{GDBN} to decide what to do with the signal depending on
15577 the signal handling tables (@pxref{Signals}). The @code{signal} command
15578 passes the signal directly to your program.
15582 @section Returning from a Function
15585 @cindex returning from a function
15588 @itemx return @var{expression}
15589 You can cancel execution of a function call with the @code{return}
15590 command. If you give an
15591 @var{expression} argument, its value is used as the function's return
15595 When you use @code{return}, @value{GDBN} discards the selected stack frame
15596 (and all frames within it). You can think of this as making the
15597 discarded frame return prematurely. If you wish to specify a value to
15598 be returned, give that value as the argument to @code{return}.
15600 This pops the selected stack frame (@pxref{Selection, ,Selecting a
15601 Frame}), and any other frames inside of it, leaving its caller as the
15602 innermost remaining frame. That frame becomes selected. The
15603 specified value is stored in the registers used for returning values
15606 The @code{return} command does not resume execution; it leaves the
15607 program stopped in the state that would exist if the function had just
15608 returned. In contrast, the @code{finish} command (@pxref{Continuing
15609 and Stepping, ,Continuing and Stepping}) resumes execution until the
15610 selected stack frame returns naturally.
15612 @value{GDBN} needs to know how the @var{expression} argument should be set for
15613 the inferior. The concrete registers assignment depends on the OS ABI and the
15614 type being returned by the selected stack frame. For example it is common for
15615 OS ABI to return floating point values in FPU registers while integer values in
15616 CPU registers. Still some ABIs return even floating point values in CPU
15617 registers. Larger integer widths (such as @code{long long int}) also have
15618 specific placement rules. @value{GDBN} already knows the OS ABI from its
15619 current target so it needs to find out also the type being returned to make the
15620 assignment into the right register(s).
15622 Normally, the selected stack frame has debug info. @value{GDBN} will always
15623 use the debug info instead of the implicit type of @var{expression} when the
15624 debug info is available. For example, if you type @kbd{return -1}, and the
15625 function in the current stack frame is declared to return a @code{long long
15626 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
15627 into a @code{long long int}:
15630 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
15632 (@value{GDBP}) return -1
15633 Make func return now? (y or n) y
15634 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
15635 43 printf ("result=%lld\n", func ());
15639 However, if the selected stack frame does not have a debug info, e.g., if the
15640 function was compiled without debug info, @value{GDBN} has to find out the type
15641 to return from user. Specifying a different type by mistake may set the value
15642 in different inferior registers than the caller code expects. For example,
15643 typing @kbd{return -1} with its implicit type @code{int} would set only a part
15644 of a @code{long long int} result for a debug info less function (on 32-bit
15645 architectures). Therefore the user is required to specify the return type by
15646 an appropriate cast explicitly:
15649 Breakpoint 2, 0x0040050b in func ()
15650 (@value{GDBP}) return -1
15651 Return value type not available for selected stack frame.
15652 Please use an explicit cast of the value to return.
15653 (@value{GDBP}) return (long long int) -1
15654 Make selected stack frame return now? (y or n) y
15655 #0 0x00400526 in main ()
15660 @section Calling Program Functions
15663 @cindex calling functions
15664 @cindex inferior functions, calling
15665 @item print @var{expr}
15666 Evaluate the expression @var{expr} and display the resulting value.
15667 @var{expr} may include calls to functions in the program being
15671 @item call @var{expr}
15672 Evaluate the expression @var{expr} without displaying @code{void}
15675 You can use this variant of the @code{print} command if you want to
15676 execute a function from your program that does not return anything
15677 (a.k.a.@: @dfn{a void function}), but without cluttering the output
15678 with @code{void} returned values that @value{GDBN} will otherwise
15679 print. If the result is not void, it is printed and saved in the
15683 It is possible for the function you call via the @code{print} or
15684 @code{call} command to generate a signal (e.g., if there's a bug in
15685 the function, or if you passed it incorrect arguments). What happens
15686 in that case is controlled by the @code{set unwindonsignal} command.
15688 Similarly, with a C@t{++} program it is possible for the function you
15689 call via the @code{print} or @code{call} command to generate an
15690 exception that is not handled due to the constraints of the dummy
15691 frame. In this case, any exception that is raised in the frame, but has
15692 an out-of-frame exception handler will not be found. GDB builds a
15693 dummy-frame for the inferior function call, and the unwinder cannot
15694 seek for exception handlers outside of this dummy-frame. What happens
15695 in that case is controlled by the
15696 @code{set unwind-on-terminating-exception} command.
15699 @item set unwindonsignal
15700 @kindex set unwindonsignal
15701 @cindex unwind stack in called functions
15702 @cindex call dummy stack unwinding
15703 Set unwinding of the stack if a signal is received while in a function
15704 that @value{GDBN} called in the program being debugged. If set to on,
15705 @value{GDBN} unwinds the stack it created for the call and restores
15706 the context to what it was before the call. If set to off (the
15707 default), @value{GDBN} stops in the frame where the signal was
15710 @item show unwindonsignal
15711 @kindex show unwindonsignal
15712 Show the current setting of stack unwinding in the functions called by
15715 @item set unwind-on-terminating-exception
15716 @kindex set unwind-on-terminating-exception
15717 @cindex unwind stack in called functions with unhandled exceptions
15718 @cindex call dummy stack unwinding on unhandled exception.
15719 Set unwinding of the stack if a C@t{++} exception is raised, but left
15720 unhandled while in a function that @value{GDBN} called in the program being
15721 debugged. If set to on (the default), @value{GDBN} unwinds the stack
15722 it created for the call and restores the context to what it was before
15723 the call. If set to off, @value{GDBN} the exception is delivered to
15724 the default C@t{++} exception handler and the inferior terminated.
15726 @item show unwind-on-terminating-exception
15727 @kindex show unwind-on-terminating-exception
15728 Show the current setting of stack unwinding in the functions called by
15733 @cindex weak alias functions
15734 Sometimes, a function you wish to call is actually a @dfn{weak alias}
15735 for another function. In such case, @value{GDBN} might not pick up
15736 the type information, including the types of the function arguments,
15737 which causes @value{GDBN} to call the inferior function incorrectly.
15738 As a result, the called function will function erroneously and may
15739 even crash. A solution to that is to use the name of the aliased
15743 @section Patching Programs
15745 @cindex patching binaries
15746 @cindex writing into executables
15747 @cindex writing into corefiles
15749 By default, @value{GDBN} opens the file containing your program's
15750 executable code (or the corefile) read-only. This prevents accidental
15751 alterations to machine code; but it also prevents you from intentionally
15752 patching your program's binary.
15754 If you'd like to be able to patch the binary, you can specify that
15755 explicitly with the @code{set write} command. For example, you might
15756 want to turn on internal debugging flags, or even to make emergency
15762 @itemx set write off
15763 If you specify @samp{set write on}, @value{GDBN} opens executable and
15764 core files for both reading and writing; if you specify @kbd{set write
15765 off} (the default), @value{GDBN} opens them read-only.
15767 If you have already loaded a file, you must load it again (using the
15768 @code{exec-file} or @code{core-file} command) after changing @code{set
15769 write}, for your new setting to take effect.
15773 Display whether executable files and core files are opened for writing
15774 as well as reading.
15778 @chapter @value{GDBN} Files
15780 @value{GDBN} needs to know the file name of the program to be debugged,
15781 both in order to read its symbol table and in order to start your
15782 program. To debug a core dump of a previous run, you must also tell
15783 @value{GDBN} the name of the core dump file.
15786 * Files:: Commands to specify files
15787 * Separate Debug Files:: Debugging information in separate files
15788 * Index Files:: Index files speed up GDB
15789 * Symbol Errors:: Errors reading symbol files
15790 * Data Files:: GDB data files
15794 @section Commands to Specify Files
15796 @cindex symbol table
15797 @cindex core dump file
15799 You may want to specify executable and core dump file names. The usual
15800 way to do this is at start-up time, using the arguments to
15801 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
15802 Out of @value{GDBN}}).
15804 Occasionally it is necessary to change to a different file during a
15805 @value{GDBN} session. Or you may run @value{GDBN} and forget to
15806 specify a file you want to use. Or you are debugging a remote target
15807 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
15808 Program}). In these situations the @value{GDBN} commands to specify
15809 new files are useful.
15812 @cindex executable file
15814 @item file @var{filename}
15815 Use @var{filename} as the program to be debugged. It is read for its
15816 symbols and for the contents of pure memory. It is also the program
15817 executed when you use the @code{run} command. If you do not specify a
15818 directory and the file is not found in the @value{GDBN} working directory,
15819 @value{GDBN} uses the environment variable @code{PATH} as a list of
15820 directories to search, just as the shell does when looking for a program
15821 to run. You can change the value of this variable, for both @value{GDBN}
15822 and your program, using the @code{path} command.
15824 @cindex unlinked object files
15825 @cindex patching object files
15826 You can load unlinked object @file{.o} files into @value{GDBN} using
15827 the @code{file} command. You will not be able to ``run'' an object
15828 file, but you can disassemble functions and inspect variables. Also,
15829 if the underlying BFD functionality supports it, you could use
15830 @kbd{gdb -write} to patch object files using this technique. Note
15831 that @value{GDBN} can neither interpret nor modify relocations in this
15832 case, so branches and some initialized variables will appear to go to
15833 the wrong place. But this feature is still handy from time to time.
15836 @code{file} with no argument makes @value{GDBN} discard any information it
15837 has on both executable file and the symbol table.
15840 @item exec-file @r{[} @var{filename} @r{]}
15841 Specify that the program to be run (but not the symbol table) is found
15842 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
15843 if necessary to locate your program. Omitting @var{filename} means to
15844 discard information on the executable file.
15846 @kindex symbol-file
15847 @item symbol-file @r{[} @var{filename} @r{]}
15848 Read symbol table information from file @var{filename}. @code{PATH} is
15849 searched when necessary. Use the @code{file} command to get both symbol
15850 table and program to run from the same file.
15852 @code{symbol-file} with no argument clears out @value{GDBN} information on your
15853 program's symbol table.
15855 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
15856 some breakpoints and auto-display expressions. This is because they may
15857 contain pointers to the internal data recording symbols and data types,
15858 which are part of the old symbol table data being discarded inside
15861 @code{symbol-file} does not repeat if you press @key{RET} again after
15864 When @value{GDBN} is configured for a particular environment, it
15865 understands debugging information in whatever format is the standard
15866 generated for that environment; you may use either a @sc{gnu} compiler, or
15867 other compilers that adhere to the local conventions.
15868 Best results are usually obtained from @sc{gnu} compilers; for example,
15869 using @code{@value{NGCC}} you can generate debugging information for
15872 For most kinds of object files, with the exception of old SVR3 systems
15873 using COFF, the @code{symbol-file} command does not normally read the
15874 symbol table in full right away. Instead, it scans the symbol table
15875 quickly to find which source files and which symbols are present. The
15876 details are read later, one source file at a time, as they are needed.
15878 The purpose of this two-stage reading strategy is to make @value{GDBN}
15879 start up faster. For the most part, it is invisible except for
15880 occasional pauses while the symbol table details for a particular source
15881 file are being read. (The @code{set verbose} command can turn these
15882 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
15883 Warnings and Messages}.)
15885 We have not implemented the two-stage strategy for COFF yet. When the
15886 symbol table is stored in COFF format, @code{symbol-file} reads the
15887 symbol table data in full right away. Note that ``stabs-in-COFF''
15888 still does the two-stage strategy, since the debug info is actually
15892 @cindex reading symbols immediately
15893 @cindex symbols, reading immediately
15894 @item symbol-file @r{[} -readnow @r{]} @var{filename}
15895 @itemx file @r{[} -readnow @r{]} @var{filename}
15896 You can override the @value{GDBN} two-stage strategy for reading symbol
15897 tables by using the @samp{-readnow} option with any of the commands that
15898 load symbol table information, if you want to be sure @value{GDBN} has the
15899 entire symbol table available.
15901 @c FIXME: for now no mention of directories, since this seems to be in
15902 @c flux. 13mar1992 status is that in theory GDB would look either in
15903 @c current dir or in same dir as myprog; but issues like competing
15904 @c GDB's, or clutter in system dirs, mean that in practice right now
15905 @c only current dir is used. FFish says maybe a special GDB hierarchy
15906 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
15910 @item core-file @r{[}@var{filename}@r{]}
15912 Specify the whereabouts of a core dump file to be used as the ``contents
15913 of memory''. Traditionally, core files contain only some parts of the
15914 address space of the process that generated them; @value{GDBN} can access the
15915 executable file itself for other parts.
15917 @code{core-file} with no argument specifies that no core file is
15920 Note that the core file is ignored when your program is actually running
15921 under @value{GDBN}. So, if you have been running your program and you
15922 wish to debug a core file instead, you must kill the subprocess in which
15923 the program is running. To do this, use the @code{kill} command
15924 (@pxref{Kill Process, ,Killing the Child Process}).
15926 @kindex add-symbol-file
15927 @cindex dynamic linking
15928 @item add-symbol-file @var{filename} @var{address}
15929 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
15930 @itemx add-symbol-file @var{filename} @var{address} -s @var{section} @var{address} @dots{}
15931 The @code{add-symbol-file} command reads additional symbol table
15932 information from the file @var{filename}. You would use this command
15933 when @var{filename} has been dynamically loaded (by some other means)
15934 into the program that is running. @var{address} should be the memory
15935 address at which the file has been loaded; @value{GDBN} cannot figure
15936 this out for itself. You can additionally specify an arbitrary number
15937 of @samp{-s @var{section} @var{address}} pairs, to give an explicit
15938 section name and base address for that section. You can specify any
15939 @var{address} as an expression.
15941 The symbol table of the file @var{filename} is added to the symbol table
15942 originally read with the @code{symbol-file} command. You can use the
15943 @code{add-symbol-file} command any number of times; the new symbol data
15944 thus read keeps adding to the old. To discard all old symbol data
15945 instead, use the @code{symbol-file} command without any arguments.
15947 @cindex relocatable object files, reading symbols from
15948 @cindex object files, relocatable, reading symbols from
15949 @cindex reading symbols from relocatable object files
15950 @cindex symbols, reading from relocatable object files
15951 @cindex @file{.o} files, reading symbols from
15952 Although @var{filename} is typically a shared library file, an
15953 executable file, or some other object file which has been fully
15954 relocated for loading into a process, you can also load symbolic
15955 information from relocatable @file{.o} files, as long as:
15959 the file's symbolic information refers only to linker symbols defined in
15960 that file, not to symbols defined by other object files,
15962 every section the file's symbolic information refers to has actually
15963 been loaded into the inferior, as it appears in the file, and
15965 you can determine the address at which every section was loaded, and
15966 provide these to the @code{add-symbol-file} command.
15970 Some embedded operating systems, like Sun Chorus and VxWorks, can load
15971 relocatable files into an already running program; such systems
15972 typically make the requirements above easy to meet. However, it's
15973 important to recognize that many native systems use complex link
15974 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
15975 assembly, for example) that make the requirements difficult to meet. In
15976 general, one cannot assume that using @code{add-symbol-file} to read a
15977 relocatable object file's symbolic information will have the same effect
15978 as linking the relocatable object file into the program in the normal
15981 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
15983 @kindex add-symbol-file-from-memory
15984 @cindex @code{syscall DSO}
15985 @cindex load symbols from memory
15986 @item add-symbol-file-from-memory @var{address}
15987 Load symbols from the given @var{address} in a dynamically loaded
15988 object file whose image is mapped directly into the inferior's memory.
15989 For example, the Linux kernel maps a @code{syscall DSO} into each
15990 process's address space; this DSO provides kernel-specific code for
15991 some system calls. The argument can be any expression whose
15992 evaluation yields the address of the file's shared object file header.
15993 For this command to work, you must have used @code{symbol-file} or
15994 @code{exec-file} commands in advance.
15996 @kindex add-shared-symbol-files
15998 @item add-shared-symbol-files @var{library-file}
15999 @itemx assf @var{library-file}
16000 The @code{add-shared-symbol-files} command can currently be used only
16001 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
16002 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
16003 @value{GDBN} automatically looks for shared libraries, however if
16004 @value{GDBN} does not find yours, you can invoke
16005 @code{add-shared-symbol-files}. It takes one argument: the shared
16006 library's file name. @code{assf} is a shorthand alias for
16007 @code{add-shared-symbol-files}.
16010 @item section @var{section} @var{addr}
16011 The @code{section} command changes the base address of the named
16012 @var{section} of the exec file to @var{addr}. This can be used if the
16013 exec file does not contain section addresses, (such as in the
16014 @code{a.out} format), or when the addresses specified in the file
16015 itself are wrong. Each section must be changed separately. The
16016 @code{info files} command, described below, lists all the sections and
16020 @kindex info target
16023 @code{info files} and @code{info target} are synonymous; both print the
16024 current target (@pxref{Targets, ,Specifying a Debugging Target}),
16025 including the names of the executable and core dump files currently in
16026 use by @value{GDBN}, and the files from which symbols were loaded. The
16027 command @code{help target} lists all possible targets rather than
16030 @kindex maint info sections
16031 @item maint info sections
16032 Another command that can give you extra information about program sections
16033 is @code{maint info sections}. In addition to the section information
16034 displayed by @code{info files}, this command displays the flags and file
16035 offset of each section in the executable and core dump files. In addition,
16036 @code{maint info sections} provides the following command options (which
16037 may be arbitrarily combined):
16041 Display sections for all loaded object files, including shared libraries.
16042 @item @var{sections}
16043 Display info only for named @var{sections}.
16044 @item @var{section-flags}
16045 Display info only for sections for which @var{section-flags} are true.
16046 The section flags that @value{GDBN} currently knows about are:
16049 Section will have space allocated in the process when loaded.
16050 Set for all sections except those containing debug information.
16052 Section will be loaded from the file into the child process memory.
16053 Set for pre-initialized code and data, clear for @code{.bss} sections.
16055 Section needs to be relocated before loading.
16057 Section cannot be modified by the child process.
16059 Section contains executable code only.
16061 Section contains data only (no executable code).
16063 Section will reside in ROM.
16065 Section contains data for constructor/destructor lists.
16067 Section is not empty.
16069 An instruction to the linker to not output the section.
16070 @item COFF_SHARED_LIBRARY
16071 A notification to the linker that the section contains
16072 COFF shared library information.
16074 Section contains common symbols.
16077 @kindex set trust-readonly-sections
16078 @cindex read-only sections
16079 @item set trust-readonly-sections on
16080 Tell @value{GDBN} that readonly sections in your object file
16081 really are read-only (i.e.@: that their contents will not change).
16082 In that case, @value{GDBN} can fetch values from these sections
16083 out of the object file, rather than from the target program.
16084 For some targets (notably embedded ones), this can be a significant
16085 enhancement to debugging performance.
16087 The default is off.
16089 @item set trust-readonly-sections off
16090 Tell @value{GDBN} not to trust readonly sections. This means that
16091 the contents of the section might change while the program is running,
16092 and must therefore be fetched from the target when needed.
16094 @item show trust-readonly-sections
16095 Show the current setting of trusting readonly sections.
16098 All file-specifying commands allow both absolute and relative file names
16099 as arguments. @value{GDBN} always converts the file name to an absolute file
16100 name and remembers it that way.
16102 @cindex shared libraries
16103 @anchor{Shared Libraries}
16104 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
16105 and IBM RS/6000 AIX shared libraries.
16107 On MS-Windows @value{GDBN} must be linked with the Expat library to support
16108 shared libraries. @xref{Expat}.
16110 @value{GDBN} automatically loads symbol definitions from shared libraries
16111 when you use the @code{run} command, or when you examine a core file.
16112 (Before you issue the @code{run} command, @value{GDBN} does not understand
16113 references to a function in a shared library, however---unless you are
16114 debugging a core file).
16116 On HP-UX, if the program loads a library explicitly, @value{GDBN}
16117 automatically loads the symbols at the time of the @code{shl_load} call.
16119 @c FIXME: some @value{GDBN} release may permit some refs to undef
16120 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
16121 @c FIXME...lib; check this from time to time when updating manual
16123 There are times, however, when you may wish to not automatically load
16124 symbol definitions from shared libraries, such as when they are
16125 particularly large or there are many of them.
16127 To control the automatic loading of shared library symbols, use the
16131 @kindex set auto-solib-add
16132 @item set auto-solib-add @var{mode}
16133 If @var{mode} is @code{on}, symbols from all shared object libraries
16134 will be loaded automatically when the inferior begins execution, you
16135 attach to an independently started inferior, or when the dynamic linker
16136 informs @value{GDBN} that a new library has been loaded. If @var{mode}
16137 is @code{off}, symbols must be loaded manually, using the
16138 @code{sharedlibrary} command. The default value is @code{on}.
16140 @cindex memory used for symbol tables
16141 If your program uses lots of shared libraries with debug info that
16142 takes large amounts of memory, you can decrease the @value{GDBN}
16143 memory footprint by preventing it from automatically loading the
16144 symbols from shared libraries. To that end, type @kbd{set
16145 auto-solib-add off} before running the inferior, then load each
16146 library whose debug symbols you do need with @kbd{sharedlibrary
16147 @var{regexp}}, where @var{regexp} is a regular expression that matches
16148 the libraries whose symbols you want to be loaded.
16150 @kindex show auto-solib-add
16151 @item show auto-solib-add
16152 Display the current autoloading mode.
16155 @cindex load shared library
16156 To explicitly load shared library symbols, use the @code{sharedlibrary}
16160 @kindex info sharedlibrary
16162 @item info share @var{regex}
16163 @itemx info sharedlibrary @var{regex}
16164 Print the names of the shared libraries which are currently loaded
16165 that match @var{regex}. If @var{regex} is omitted then print
16166 all shared libraries that are loaded.
16168 @kindex sharedlibrary
16170 @item sharedlibrary @var{regex}
16171 @itemx share @var{regex}
16172 Load shared object library symbols for files matching a
16173 Unix regular expression.
16174 As with files loaded automatically, it only loads shared libraries
16175 required by your program for a core file or after typing @code{run}. If
16176 @var{regex} is omitted all shared libraries required by your program are
16179 @item nosharedlibrary
16180 @kindex nosharedlibrary
16181 @cindex unload symbols from shared libraries
16182 Unload all shared object library symbols. This discards all symbols
16183 that have been loaded from all shared libraries. Symbols from shared
16184 libraries that were loaded by explicit user requests are not
16188 Sometimes you may wish that @value{GDBN} stops and gives you control
16189 when any of shared library events happen. The best way to do this is
16190 to use @code{catch load} and @code{catch unload} (@pxref{Set
16193 @value{GDBN} also supports the the @code{set stop-on-solib-events}
16194 command for this. This command exists for historical reasons. It is
16195 less useful than setting a catchpoint, because it does not allow for
16196 conditions or commands as a catchpoint does.
16199 @item set stop-on-solib-events
16200 @kindex set stop-on-solib-events
16201 This command controls whether @value{GDBN} should give you control
16202 when the dynamic linker notifies it about some shared library event.
16203 The most common event of interest is loading or unloading of a new
16206 @item show stop-on-solib-events
16207 @kindex show stop-on-solib-events
16208 Show whether @value{GDBN} stops and gives you control when shared
16209 library events happen.
16212 Shared libraries are also supported in many cross or remote debugging
16213 configurations. @value{GDBN} needs to have access to the target's libraries;
16214 this can be accomplished either by providing copies of the libraries
16215 on the host system, or by asking @value{GDBN} to automatically retrieve the
16216 libraries from the target. If copies of the target libraries are
16217 provided, they need to be the same as the target libraries, although the
16218 copies on the target can be stripped as long as the copies on the host are
16221 @cindex where to look for shared libraries
16222 For remote debugging, you need to tell @value{GDBN} where the target
16223 libraries are, so that it can load the correct copies---otherwise, it
16224 may try to load the host's libraries. @value{GDBN} has two variables
16225 to specify the search directories for target libraries.
16228 @cindex prefix for shared library file names
16229 @cindex system root, alternate
16230 @kindex set solib-absolute-prefix
16231 @kindex set sysroot
16232 @item set sysroot @var{path}
16233 Use @var{path} as the system root for the program being debugged. Any
16234 absolute shared library paths will be prefixed with @var{path}; many
16235 runtime loaders store the absolute paths to the shared library in the
16236 target program's memory. If you use @code{set sysroot} to find shared
16237 libraries, they need to be laid out in the same way that they are on
16238 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
16241 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
16242 retrieve the target libraries from the remote system. This is only
16243 supported when using a remote target that supports the @code{remote get}
16244 command (@pxref{File Transfer,,Sending files to a remote system}).
16245 The part of @var{path} following the initial @file{remote:}
16246 (if present) is used as system root prefix on the remote file system.
16247 @footnote{If you want to specify a local system root using a directory
16248 that happens to be named @file{remote:}, you need to use some equivalent
16249 variant of the name like @file{./remote:}.}
16251 For targets with an MS-DOS based filesystem, such as MS-Windows and
16252 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
16253 absolute file name with @var{path}. But first, on Unix hosts,
16254 @value{GDBN} converts all backslash directory separators into forward
16255 slashes, because the backslash is not a directory separator on Unix:
16258 c:\foo\bar.dll @result{} c:/foo/bar.dll
16261 Then, @value{GDBN} attempts prefixing the target file name with
16262 @var{path}, and looks for the resulting file name in the host file
16266 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
16269 If that does not find the shared library, @value{GDBN} tries removing
16270 the @samp{:} character from the drive spec, both for convenience, and,
16271 for the case of the host file system not supporting file names with
16275 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
16278 This makes it possible to have a system root that mirrors a target
16279 with more than one drive. E.g., you may want to setup your local
16280 copies of the target system shared libraries like so (note @samp{c} vs
16284 @file{/path/to/sysroot/c/sys/bin/foo.dll}
16285 @file{/path/to/sysroot/c/sys/bin/bar.dll}
16286 @file{/path/to/sysroot/z/sys/bin/bar.dll}
16290 and point the system root at @file{/path/to/sysroot}, so that
16291 @value{GDBN} can find the correct copies of both
16292 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
16294 If that still does not find the shared library, @value{GDBN} tries
16295 removing the whole drive spec from the target file name:
16298 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
16301 This last lookup makes it possible to not care about the drive name,
16302 if you don't want or need to.
16304 The @code{set solib-absolute-prefix} command is an alias for @code{set
16307 @cindex default system root
16308 @cindex @samp{--with-sysroot}
16309 You can set the default system root by using the configure-time
16310 @samp{--with-sysroot} option. If the system root is inside
16311 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
16312 @samp{--exec-prefix}), then the default system root will be updated
16313 automatically if the installed @value{GDBN} is moved to a new
16316 @kindex show sysroot
16318 Display the current shared library prefix.
16320 @kindex set solib-search-path
16321 @item set solib-search-path @var{path}
16322 If this variable is set, @var{path} is a colon-separated list of
16323 directories to search for shared libraries. @samp{solib-search-path}
16324 is used after @samp{sysroot} fails to locate the library, or if the
16325 path to the library is relative instead of absolute. If you want to
16326 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
16327 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
16328 finding your host's libraries. @samp{sysroot} is preferred; setting
16329 it to a nonexistent directory may interfere with automatic loading
16330 of shared library symbols.
16332 @kindex show solib-search-path
16333 @item show solib-search-path
16334 Display the current shared library search path.
16336 @cindex DOS file-name semantics of file names.
16337 @kindex set target-file-system-kind (unix|dos-based|auto)
16338 @kindex show target-file-system-kind
16339 @item set target-file-system-kind @var{kind}
16340 Set assumed file system kind for target reported file names.
16342 Shared library file names as reported by the target system may not
16343 make sense as is on the system @value{GDBN} is running on. For
16344 example, when remote debugging a target that has MS-DOS based file
16345 system semantics, from a Unix host, the target may be reporting to
16346 @value{GDBN} a list of loaded shared libraries with file names such as
16347 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
16348 drive letters, so the @samp{c:\} prefix is not normally understood as
16349 indicating an absolute file name, and neither is the backslash
16350 normally considered a directory separator character. In that case,
16351 the native file system would interpret this whole absolute file name
16352 as a relative file name with no directory components. This would make
16353 it impossible to point @value{GDBN} at a copy of the remote target's
16354 shared libraries on the host using @code{set sysroot}, and impractical
16355 with @code{set solib-search-path}. Setting
16356 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
16357 to interpret such file names similarly to how the target would, and to
16358 map them to file names valid on @value{GDBN}'s native file system
16359 semantics. The value of @var{kind} can be @code{"auto"}, in addition
16360 to one of the supported file system kinds. In that case, @value{GDBN}
16361 tries to determine the appropriate file system variant based on the
16362 current target's operating system (@pxref{ABI, ,Configuring the
16363 Current ABI}). The supported file system settings are:
16367 Instruct @value{GDBN} to assume the target file system is of Unix
16368 kind. Only file names starting the forward slash (@samp{/}) character
16369 are considered absolute, and the directory separator character is also
16373 Instruct @value{GDBN} to assume the target file system is DOS based.
16374 File names starting with either a forward slash, or a drive letter
16375 followed by a colon (e.g., @samp{c:}), are considered absolute, and
16376 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
16377 considered directory separators.
16380 Instruct @value{GDBN} to use the file system kind associated with the
16381 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
16382 This is the default.
16386 @cindex file name canonicalization
16387 @cindex base name differences
16388 When processing file names provided by the user, @value{GDBN}
16389 frequently needs to compare them to the file names recorded in the
16390 program's debug info. Normally, @value{GDBN} compares just the
16391 @dfn{base names} of the files as strings, which is reasonably fast
16392 even for very large programs. (The base name of a file is the last
16393 portion of its name, after stripping all the leading directories.)
16394 This shortcut in comparison is based upon the assumption that files
16395 cannot have more than one base name. This is usually true, but
16396 references to files that use symlinks or similar filesystem
16397 facilities violate that assumption. If your program records files
16398 using such facilities, or if you provide file names to @value{GDBN}
16399 using symlinks etc., you can set @code{basenames-may-differ} to
16400 @code{true} to instruct @value{GDBN} to completely canonicalize each
16401 pair of file names it needs to compare. This will make file-name
16402 comparisons accurate, but at a price of a significant slowdown.
16405 @item set basenames-may-differ
16406 @kindex set basenames-may-differ
16407 Set whether a source file may have multiple base names.
16409 @item show basenames-may-differ
16410 @kindex show basenames-may-differ
16411 Show whether a source file may have multiple base names.
16414 @node Separate Debug Files
16415 @section Debugging Information in Separate Files
16416 @cindex separate debugging information files
16417 @cindex debugging information in separate files
16418 @cindex @file{.debug} subdirectories
16419 @cindex debugging information directory, global
16420 @cindex global debugging information directories
16421 @cindex build ID, and separate debugging files
16422 @cindex @file{.build-id} directory
16424 @value{GDBN} allows you to put a program's debugging information in a
16425 file separate from the executable itself, in a way that allows
16426 @value{GDBN} to find and load the debugging information automatically.
16427 Since debugging information can be very large---sometimes larger
16428 than the executable code itself---some systems distribute debugging
16429 information for their executables in separate files, which users can
16430 install only when they need to debug a problem.
16432 @value{GDBN} supports two ways of specifying the separate debug info
16437 The executable contains a @dfn{debug link} that specifies the name of
16438 the separate debug info file. The separate debug file's name is
16439 usually @file{@var{executable}.debug}, where @var{executable} is the
16440 name of the corresponding executable file without leading directories
16441 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
16442 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
16443 checksum for the debug file, which @value{GDBN} uses to validate that
16444 the executable and the debug file came from the same build.
16447 The executable contains a @dfn{build ID}, a unique bit string that is
16448 also present in the corresponding debug info file. (This is supported
16449 only on some operating systems, notably those which use the ELF format
16450 for binary files and the @sc{gnu} Binutils.) For more details about
16451 this feature, see the description of the @option{--build-id}
16452 command-line option in @ref{Options, , Command Line Options, ld.info,
16453 The GNU Linker}. The debug info file's name is not specified
16454 explicitly by the build ID, but can be computed from the build ID, see
16458 Depending on the way the debug info file is specified, @value{GDBN}
16459 uses two different methods of looking for the debug file:
16463 For the ``debug link'' method, @value{GDBN} looks up the named file in
16464 the directory of the executable file, then in a subdirectory of that
16465 directory named @file{.debug}, and finally under each one of the global debug
16466 directories, in a subdirectory whose name is identical to the leading
16467 directories of the executable's absolute file name.
16470 For the ``build ID'' method, @value{GDBN} looks in the
16471 @file{.build-id} subdirectory of each one of the global debug directories for
16472 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
16473 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
16474 are the rest of the bit string. (Real build ID strings are 32 or more
16475 hex characters, not 10.)
16478 So, for example, suppose you ask @value{GDBN} to debug
16479 @file{/usr/bin/ls}, which has a debug link that specifies the
16480 file @file{ls.debug}, and a build ID whose value in hex is
16481 @code{abcdef1234}. If the list of the global debug directories includes
16482 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
16483 debug information files, in the indicated order:
16487 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
16489 @file{/usr/bin/ls.debug}
16491 @file{/usr/bin/.debug/ls.debug}
16493 @file{/usr/lib/debug/usr/bin/ls.debug}.
16496 @anchor{debug-file-directory}
16497 Global debugging info directories default to what is set by @value{GDBN}
16498 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
16499 you can also set the global debugging info directories, and view the list
16500 @value{GDBN} is currently using.
16504 @kindex set debug-file-directory
16505 @item set debug-file-directory @var{directories}
16506 Set the directories which @value{GDBN} searches for separate debugging
16507 information files to @var{directory}. Multiple path components can be set
16508 concatenating them by a path separator.
16510 @kindex show debug-file-directory
16511 @item show debug-file-directory
16512 Show the directories @value{GDBN} searches for separate debugging
16517 @cindex @code{.gnu_debuglink} sections
16518 @cindex debug link sections
16519 A debug link is a special section of the executable file named
16520 @code{.gnu_debuglink}. The section must contain:
16524 A filename, with any leading directory components removed, followed by
16527 zero to three bytes of padding, as needed to reach the next four-byte
16528 boundary within the section, and
16530 a four-byte CRC checksum, stored in the same endianness used for the
16531 executable file itself. The checksum is computed on the debugging
16532 information file's full contents by the function given below, passing
16533 zero as the @var{crc} argument.
16536 Any executable file format can carry a debug link, as long as it can
16537 contain a section named @code{.gnu_debuglink} with the contents
16540 @cindex @code{.note.gnu.build-id} sections
16541 @cindex build ID sections
16542 The build ID is a special section in the executable file (and in other
16543 ELF binary files that @value{GDBN} may consider). This section is
16544 often named @code{.note.gnu.build-id}, but that name is not mandatory.
16545 It contains unique identification for the built files---the ID remains
16546 the same across multiple builds of the same build tree. The default
16547 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
16548 content for the build ID string. The same section with an identical
16549 value is present in the original built binary with symbols, in its
16550 stripped variant, and in the separate debugging information file.
16552 The debugging information file itself should be an ordinary
16553 executable, containing a full set of linker symbols, sections, and
16554 debugging information. The sections of the debugging information file
16555 should have the same names, addresses, and sizes as the original file,
16556 but they need not contain any data---much like a @code{.bss} section
16557 in an ordinary executable.
16559 The @sc{gnu} binary utilities (Binutils) package includes the
16560 @samp{objcopy} utility that can produce
16561 the separated executable / debugging information file pairs using the
16562 following commands:
16565 @kbd{objcopy --only-keep-debug foo foo.debug}
16570 These commands remove the debugging
16571 information from the executable file @file{foo} and place it in the file
16572 @file{foo.debug}. You can use the first, second or both methods to link the
16577 The debug link method needs the following additional command to also leave
16578 behind a debug link in @file{foo}:
16581 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
16584 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
16585 a version of the @code{strip} command such that the command @kbd{strip foo -f
16586 foo.debug} has the same functionality as the two @code{objcopy} commands and
16587 the @code{ln -s} command above, together.
16590 Build ID gets embedded into the main executable using @code{ld --build-id} or
16591 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
16592 compatibility fixes for debug files separation are present in @sc{gnu} binary
16593 utilities (Binutils) package since version 2.18.
16598 @cindex CRC algorithm definition
16599 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
16600 IEEE 802.3 using the polynomial:
16602 @c TexInfo requires naked braces for multi-digit exponents for Tex
16603 @c output, but this causes HTML output to barf. HTML has to be set using
16604 @c raw commands. So we end up having to specify this equation in 2
16609 <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>
16610 + <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
16616 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
16617 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
16621 The function is computed byte at a time, taking the least
16622 significant bit of each byte first. The initial pattern
16623 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
16624 the final result is inverted to ensure trailing zeros also affect the
16627 @emph{Note:} This is the same CRC polynomial as used in handling the
16628 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{Remote Protocol,
16629 , @value{GDBN} Remote Serial Protocol}). However in the
16630 case of the Remote Serial Protocol, the CRC is computed @emph{most}
16631 significant bit first, and the result is not inverted, so trailing
16632 zeros have no effect on the CRC value.
16634 To complete the description, we show below the code of the function
16635 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
16636 initially supplied @code{crc} argument means that an initial call to
16637 this function passing in zero will start computing the CRC using
16640 @kindex gnu_debuglink_crc32
16643 gnu_debuglink_crc32 (unsigned long crc,
16644 unsigned char *buf, size_t len)
16646 static const unsigned long crc32_table[256] =
16648 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
16649 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
16650 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
16651 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
16652 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
16653 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
16654 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
16655 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
16656 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
16657 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
16658 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
16659 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
16660 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
16661 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
16662 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
16663 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
16664 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
16665 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
16666 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
16667 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
16668 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
16669 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
16670 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
16671 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
16672 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
16673 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
16674 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
16675 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
16676 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
16677 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
16678 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
16679 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
16680 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
16681 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
16682 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
16683 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
16684 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
16685 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
16686 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
16687 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
16688 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
16689 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
16690 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
16691 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
16692 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
16693 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
16694 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
16695 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
16696 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
16697 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
16698 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
16701 unsigned char *end;
16703 crc = ~crc & 0xffffffff;
16704 for (end = buf + len; buf < end; ++buf)
16705 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
16706 return ~crc & 0xffffffff;
16711 This computation does not apply to the ``build ID'' method.
16715 @section Index Files Speed Up @value{GDBN}
16716 @cindex index files
16717 @cindex @samp{.gdb_index} section
16719 When @value{GDBN} finds a symbol file, it scans the symbols in the
16720 file in order to construct an internal symbol table. This lets most
16721 @value{GDBN} operations work quickly---at the cost of a delay early
16722 on. For large programs, this delay can be quite lengthy, so
16723 @value{GDBN} provides a way to build an index, which speeds up
16726 The index is stored as a section in the symbol file. @value{GDBN} can
16727 write the index to a file, then you can put it into the symbol file
16728 using @command{objcopy}.
16730 To create an index file, use the @code{save gdb-index} command:
16733 @item save gdb-index @var{directory}
16734 @kindex save gdb-index
16735 Create an index file for each symbol file currently known by
16736 @value{GDBN}. Each file is named after its corresponding symbol file,
16737 with @samp{.gdb-index} appended, and is written into the given
16741 Once you have created an index file you can merge it into your symbol
16742 file, here named @file{symfile}, using @command{objcopy}:
16745 $ objcopy --add-section .gdb_index=symfile.gdb-index \
16746 --set-section-flags .gdb_index=readonly symfile symfile
16749 There are currently some limitation on indices. They only work when
16750 for DWARF debugging information, not stabs. And, they do not
16751 currently work for programs using Ada.
16753 @node Symbol Errors
16754 @section Errors Reading Symbol Files
16756 While reading a symbol file, @value{GDBN} occasionally encounters problems,
16757 such as symbol types it does not recognize, or known bugs in compiler
16758 output. By default, @value{GDBN} does not notify you of such problems, since
16759 they are relatively common and primarily of interest to people
16760 debugging compilers. If you are interested in seeing information
16761 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
16762 only one message about each such type of problem, no matter how many
16763 times the problem occurs; or you can ask @value{GDBN} to print more messages,
16764 to see how many times the problems occur, with the @code{set
16765 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
16768 The messages currently printed, and their meanings, include:
16771 @item inner block not inside outer block in @var{symbol}
16773 The symbol information shows where symbol scopes begin and end
16774 (such as at the start of a function or a block of statements). This
16775 error indicates that an inner scope block is not fully contained
16776 in its outer scope blocks.
16778 @value{GDBN} circumvents the problem by treating the inner block as if it had
16779 the same scope as the outer block. In the error message, @var{symbol}
16780 may be shown as ``@code{(don't know)}'' if the outer block is not a
16783 @item block at @var{address} out of order
16785 The symbol information for symbol scope blocks should occur in
16786 order of increasing addresses. This error indicates that it does not
16789 @value{GDBN} does not circumvent this problem, and has trouble
16790 locating symbols in the source file whose symbols it is reading. (You
16791 can often determine what source file is affected by specifying
16792 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
16795 @item bad block start address patched
16797 The symbol information for a symbol scope block has a start address
16798 smaller than the address of the preceding source line. This is known
16799 to occur in the SunOS 4.1.1 (and earlier) C compiler.
16801 @value{GDBN} circumvents the problem by treating the symbol scope block as
16802 starting on the previous source line.
16804 @item bad string table offset in symbol @var{n}
16807 Symbol number @var{n} contains a pointer into the string table which is
16808 larger than the size of the string table.
16810 @value{GDBN} circumvents the problem by considering the symbol to have the
16811 name @code{foo}, which may cause other problems if many symbols end up
16814 @item unknown symbol type @code{0x@var{nn}}
16816 The symbol information contains new data types that @value{GDBN} does
16817 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
16818 uncomprehended information, in hexadecimal.
16820 @value{GDBN} circumvents the error by ignoring this symbol information.
16821 This usually allows you to debug your program, though certain symbols
16822 are not accessible. If you encounter such a problem and feel like
16823 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
16824 on @code{complain}, then go up to the function @code{read_dbx_symtab}
16825 and examine @code{*bufp} to see the symbol.
16827 @item stub type has NULL name
16829 @value{GDBN} could not find the full definition for a struct or class.
16831 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
16832 The symbol information for a C@t{++} member function is missing some
16833 information that recent versions of the compiler should have output for
16836 @item info mismatch between compiler and debugger
16838 @value{GDBN} could not parse a type specification output by the compiler.
16843 @section GDB Data Files
16845 @cindex prefix for data files
16846 @value{GDBN} will sometimes read an auxiliary data file. These files
16847 are kept in a directory known as the @dfn{data directory}.
16849 You can set the data directory's name, and view the name @value{GDBN}
16850 is currently using.
16853 @kindex set data-directory
16854 @item set data-directory @var{directory}
16855 Set the directory which @value{GDBN} searches for auxiliary data files
16856 to @var{directory}.
16858 @kindex show data-directory
16859 @item show data-directory
16860 Show the directory @value{GDBN} searches for auxiliary data files.
16863 @cindex default data directory
16864 @cindex @samp{--with-gdb-datadir}
16865 You can set the default data directory by using the configure-time
16866 @samp{--with-gdb-datadir} option. If the data directory is inside
16867 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
16868 @samp{--exec-prefix}), then the default data directory will be updated
16869 automatically if the installed @value{GDBN} is moved to a new
16872 The data directory may also be specified with the
16873 @code{--data-directory} command line option.
16874 @xref{Mode Options}.
16877 @chapter Specifying a Debugging Target
16879 @cindex debugging target
16880 A @dfn{target} is the execution environment occupied by your program.
16882 Often, @value{GDBN} runs in the same host environment as your program;
16883 in that case, the debugging target is specified as a side effect when
16884 you use the @code{file} or @code{core} commands. When you need more
16885 flexibility---for example, running @value{GDBN} on a physically separate
16886 host, or controlling a standalone system over a serial port or a
16887 realtime system over a TCP/IP connection---you can use the @code{target}
16888 command to specify one of the target types configured for @value{GDBN}
16889 (@pxref{Target Commands, ,Commands for Managing Targets}).
16891 @cindex target architecture
16892 It is possible to build @value{GDBN} for several different @dfn{target
16893 architectures}. When @value{GDBN} is built like that, you can choose
16894 one of the available architectures with the @kbd{set architecture}
16898 @kindex set architecture
16899 @kindex show architecture
16900 @item set architecture @var{arch}
16901 This command sets the current target architecture to @var{arch}. The
16902 value of @var{arch} can be @code{"auto"}, in addition to one of the
16903 supported architectures.
16905 @item show architecture
16906 Show the current target architecture.
16908 @item set processor
16910 @kindex set processor
16911 @kindex show processor
16912 These are alias commands for, respectively, @code{set architecture}
16913 and @code{show architecture}.
16917 * Active Targets:: Active targets
16918 * Target Commands:: Commands for managing targets
16919 * Byte Order:: Choosing target byte order
16922 @node Active Targets
16923 @section Active Targets
16925 @cindex stacking targets
16926 @cindex active targets
16927 @cindex multiple targets
16929 There are multiple classes of targets such as: processes, executable files or
16930 recording sessions. Core files belong to the process class, making core file
16931 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
16932 on multiple active targets, one in each class. This allows you to (for
16933 example) start a process and inspect its activity, while still having access to
16934 the executable file after the process finishes. Or if you start process
16935 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
16936 presented a virtual layer of the recording target, while the process target
16937 remains stopped at the chronologically last point of the process execution.
16939 Use the @code{core-file} and @code{exec-file} commands to select a new core
16940 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
16941 specify as a target a process that is already running, use the @code{attach}
16942 command (@pxref{Attach, ,Debugging an Already-running Process}).
16944 @node Target Commands
16945 @section Commands for Managing Targets
16948 @item target @var{type} @var{parameters}
16949 Connects the @value{GDBN} host environment to a target machine or
16950 process. A target is typically a protocol for talking to debugging
16951 facilities. You use the argument @var{type} to specify the type or
16952 protocol of the target machine.
16954 Further @var{parameters} are interpreted by the target protocol, but
16955 typically include things like device names or host names to connect
16956 with, process numbers, and baud rates.
16958 The @code{target} command does not repeat if you press @key{RET} again
16959 after executing the command.
16961 @kindex help target
16963 Displays the names of all targets available. To display targets
16964 currently selected, use either @code{info target} or @code{info files}
16965 (@pxref{Files, ,Commands to Specify Files}).
16967 @item help target @var{name}
16968 Describe a particular target, including any parameters necessary to
16971 @kindex set gnutarget
16972 @item set gnutarget @var{args}
16973 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
16974 knows whether it is reading an @dfn{executable},
16975 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
16976 with the @code{set gnutarget} command. Unlike most @code{target} commands,
16977 with @code{gnutarget} the @code{target} refers to a program, not a machine.
16980 @emph{Warning:} To specify a file format with @code{set gnutarget},
16981 you must know the actual BFD name.
16985 @xref{Files, , Commands to Specify Files}.
16987 @kindex show gnutarget
16988 @item show gnutarget
16989 Use the @code{show gnutarget} command to display what file format
16990 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
16991 @value{GDBN} will determine the file format for each file automatically,
16992 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
16995 @cindex common targets
16996 Here are some common targets (available, or not, depending on the GDB
17001 @item target exec @var{program}
17002 @cindex executable file target
17003 An executable file. @samp{target exec @var{program}} is the same as
17004 @samp{exec-file @var{program}}.
17006 @item target core @var{filename}
17007 @cindex core dump file target
17008 A core dump file. @samp{target core @var{filename}} is the same as
17009 @samp{core-file @var{filename}}.
17011 @item target remote @var{medium}
17012 @cindex remote target
17013 A remote system connected to @value{GDBN} via a serial line or network
17014 connection. This command tells @value{GDBN} to use its own remote
17015 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
17017 For example, if you have a board connected to @file{/dev/ttya} on the
17018 machine running @value{GDBN}, you could say:
17021 target remote /dev/ttya
17024 @code{target remote} supports the @code{load} command. This is only
17025 useful if you have some other way of getting the stub to the target
17026 system, and you can put it somewhere in memory where it won't get
17027 clobbered by the download.
17029 @item target sim @r{[}@var{simargs}@r{]} @dots{}
17030 @cindex built-in simulator target
17031 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
17039 works; however, you cannot assume that a specific memory map, device
17040 drivers, or even basic I/O is available, although some simulators do
17041 provide these. For info about any processor-specific simulator details,
17042 see the appropriate section in @ref{Embedded Processors, ,Embedded
17047 Some configurations may include these targets as well:
17051 @item target nrom @var{dev}
17052 @cindex NetROM ROM emulator target
17053 NetROM ROM emulator. This target only supports downloading.
17057 Different targets are available on different configurations of @value{GDBN};
17058 your configuration may have more or fewer targets.
17060 Many remote targets require you to download the executable's code once
17061 you've successfully established a connection. You may wish to control
17062 various aspects of this process.
17067 @kindex set hash@r{, for remote monitors}
17068 @cindex hash mark while downloading
17069 This command controls whether a hash mark @samp{#} is displayed while
17070 downloading a file to the remote monitor. If on, a hash mark is
17071 displayed after each S-record is successfully downloaded to the
17075 @kindex show hash@r{, for remote monitors}
17076 Show the current status of displaying the hash mark.
17078 @item set debug monitor
17079 @kindex set debug monitor
17080 @cindex display remote monitor communications
17081 Enable or disable display of communications messages between
17082 @value{GDBN} and the remote monitor.
17084 @item show debug monitor
17085 @kindex show debug monitor
17086 Show the current status of displaying communications between
17087 @value{GDBN} and the remote monitor.
17092 @kindex load @var{filename}
17093 @item load @var{filename}
17095 Depending on what remote debugging facilities are configured into
17096 @value{GDBN}, the @code{load} command may be available. Where it exists, it
17097 is meant to make @var{filename} (an executable) available for debugging
17098 on the remote system---by downloading, or dynamic linking, for example.
17099 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
17100 the @code{add-symbol-file} command.
17102 If your @value{GDBN} does not have a @code{load} command, attempting to
17103 execute it gets the error message ``@code{You can't do that when your
17104 target is @dots{}}''
17106 The file is loaded at whatever address is specified in the executable.
17107 For some object file formats, you can specify the load address when you
17108 link the program; for other formats, like a.out, the object file format
17109 specifies a fixed address.
17110 @c FIXME! This would be a good place for an xref to the GNU linker doc.
17112 Depending on the remote side capabilities, @value{GDBN} may be able to
17113 load programs into flash memory.
17115 @code{load} does not repeat if you press @key{RET} again after using it.
17119 @section Choosing Target Byte Order
17121 @cindex choosing target byte order
17122 @cindex target byte order
17124 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
17125 offer the ability to run either big-endian or little-endian byte
17126 orders. Usually the executable or symbol will include a bit to
17127 designate the endian-ness, and you will not need to worry about
17128 which to use. However, you may still find it useful to adjust
17129 @value{GDBN}'s idea of processor endian-ness manually.
17133 @item set endian big
17134 Instruct @value{GDBN} to assume the target is big-endian.
17136 @item set endian little
17137 Instruct @value{GDBN} to assume the target is little-endian.
17139 @item set endian auto
17140 Instruct @value{GDBN} to use the byte order associated with the
17144 Display @value{GDBN}'s current idea of the target byte order.
17148 Note that these commands merely adjust interpretation of symbolic
17149 data on the host, and that they have absolutely no effect on the
17153 @node Remote Debugging
17154 @chapter Debugging Remote Programs
17155 @cindex remote debugging
17157 If you are trying to debug a program running on a machine that cannot run
17158 @value{GDBN} in the usual way, it is often useful to use remote debugging.
17159 For example, you might use remote debugging on an operating system kernel,
17160 or on a small system which does not have a general purpose operating system
17161 powerful enough to run a full-featured debugger.
17163 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
17164 to make this work with particular debugging targets. In addition,
17165 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
17166 but not specific to any particular target system) which you can use if you
17167 write the remote stubs---the code that runs on the remote system to
17168 communicate with @value{GDBN}.
17170 Other remote targets may be available in your
17171 configuration of @value{GDBN}; use @code{help target} to list them.
17174 * Connecting:: Connecting to a remote target
17175 * File Transfer:: Sending files to a remote system
17176 * Server:: Using the gdbserver program
17177 * Remote Configuration:: Remote configuration
17178 * Remote Stub:: Implementing a remote stub
17182 @section Connecting to a Remote Target
17184 On the @value{GDBN} host machine, you will need an unstripped copy of
17185 your program, since @value{GDBN} needs symbol and debugging information.
17186 Start up @value{GDBN} as usual, using the name of the local copy of your
17187 program as the first argument.
17189 @cindex @code{target remote}
17190 @value{GDBN} can communicate with the target over a serial line, or
17191 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
17192 each case, @value{GDBN} uses the same protocol for debugging your
17193 program; only the medium carrying the debugging packets varies. The
17194 @code{target remote} command establishes a connection to the target.
17195 Its arguments indicate which medium to use:
17199 @item target remote @var{serial-device}
17200 @cindex serial line, @code{target remote}
17201 Use @var{serial-device} to communicate with the target. For example,
17202 to use a serial line connected to the device named @file{/dev/ttyb}:
17205 target remote /dev/ttyb
17208 If you're using a serial line, you may want to give @value{GDBN} the
17209 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
17210 (@pxref{Remote Configuration, set remotebaud}) before the
17211 @code{target} command.
17213 @item target remote @code{@var{host}:@var{port}}
17214 @itemx target remote @code{tcp:@var{host}:@var{port}}
17215 @cindex @acronym{TCP} port, @code{target remote}
17216 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
17217 The @var{host} may be either a host name or a numeric @acronym{IP}
17218 address; @var{port} must be a decimal number. The @var{host} could be
17219 the target machine itself, if it is directly connected to the net, or
17220 it might be a terminal server which in turn has a serial line to the
17223 For example, to connect to port 2828 on a terminal server named
17227 target remote manyfarms:2828
17230 If your remote target is actually running on the same machine as your
17231 debugger session (e.g.@: a simulator for your target running on the
17232 same host), you can omit the hostname. For example, to connect to
17233 port 1234 on your local machine:
17236 target remote :1234
17240 Note that the colon is still required here.
17242 @item target remote @code{udp:@var{host}:@var{port}}
17243 @cindex @acronym{UDP} port, @code{target remote}
17244 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
17245 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
17248 target remote udp:manyfarms:2828
17251 When using a @acronym{UDP} connection for remote debugging, you should
17252 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
17253 can silently drop packets on busy or unreliable networks, which will
17254 cause havoc with your debugging session.
17256 @item target remote | @var{command}
17257 @cindex pipe, @code{target remote} to
17258 Run @var{command} in the background and communicate with it using a
17259 pipe. The @var{command} is a shell command, to be parsed and expanded
17260 by the system's command shell, @code{/bin/sh}; it should expect remote
17261 protocol packets on its standard input, and send replies on its
17262 standard output. You could use this to run a stand-alone simulator
17263 that speaks the remote debugging protocol, to make net connections
17264 using programs like @code{ssh}, or for other similar tricks.
17266 If @var{command} closes its standard output (perhaps by exiting),
17267 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
17268 program has already exited, this will have no effect.)
17272 Once the connection has been established, you can use all the usual
17273 commands to examine and change data. The remote program is already
17274 running; you can use @kbd{step} and @kbd{continue}, and you do not
17275 need to use @kbd{run}.
17277 @cindex interrupting remote programs
17278 @cindex remote programs, interrupting
17279 Whenever @value{GDBN} is waiting for the remote program, if you type the
17280 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
17281 program. This may or may not succeed, depending in part on the hardware
17282 and the serial drivers the remote system uses. If you type the
17283 interrupt character once again, @value{GDBN} displays this prompt:
17286 Interrupted while waiting for the program.
17287 Give up (and stop debugging it)? (y or n)
17290 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
17291 (If you decide you want to try again later, you can use @samp{target
17292 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
17293 goes back to waiting.
17296 @kindex detach (remote)
17298 When you have finished debugging the remote program, you can use the
17299 @code{detach} command to release it from @value{GDBN} control.
17300 Detaching from the target normally resumes its execution, but the results
17301 will depend on your particular remote stub. After the @code{detach}
17302 command, @value{GDBN} is free to connect to another target.
17306 The @code{disconnect} command behaves like @code{detach}, except that
17307 the target is generally not resumed. It will wait for @value{GDBN}
17308 (this instance or another one) to connect and continue debugging. After
17309 the @code{disconnect} command, @value{GDBN} is again free to connect to
17312 @cindex send command to remote monitor
17313 @cindex extend @value{GDBN} for remote targets
17314 @cindex add new commands for external monitor
17316 @item monitor @var{cmd}
17317 This command allows you to send arbitrary commands directly to the
17318 remote monitor. Since @value{GDBN} doesn't care about the commands it
17319 sends like this, this command is the way to extend @value{GDBN}---you
17320 can add new commands that only the external monitor will understand
17324 @node File Transfer
17325 @section Sending files to a remote system
17326 @cindex remote target, file transfer
17327 @cindex file transfer
17328 @cindex sending files to remote systems
17330 Some remote targets offer the ability to transfer files over the same
17331 connection used to communicate with @value{GDBN}. This is convenient
17332 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
17333 running @code{gdbserver} over a network interface. For other targets,
17334 e.g.@: embedded devices with only a single serial port, this may be
17335 the only way to upload or download files.
17337 Not all remote targets support these commands.
17341 @item remote put @var{hostfile} @var{targetfile}
17342 Copy file @var{hostfile} from the host system (the machine running
17343 @value{GDBN}) to @var{targetfile} on the target system.
17346 @item remote get @var{targetfile} @var{hostfile}
17347 Copy file @var{targetfile} from the target system to @var{hostfile}
17348 on the host system.
17350 @kindex remote delete
17351 @item remote delete @var{targetfile}
17352 Delete @var{targetfile} from the target system.
17357 @section Using the @code{gdbserver} Program
17360 @cindex remote connection without stubs
17361 @code{gdbserver} is a control program for Unix-like systems, which
17362 allows you to connect your program with a remote @value{GDBN} via
17363 @code{target remote}---but without linking in the usual debugging stub.
17365 @code{gdbserver} is not a complete replacement for the debugging stubs,
17366 because it requires essentially the same operating-system facilities
17367 that @value{GDBN} itself does. In fact, a system that can run
17368 @code{gdbserver} to connect to a remote @value{GDBN} could also run
17369 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
17370 because it is a much smaller program than @value{GDBN} itself. It is
17371 also easier to port than all of @value{GDBN}, so you may be able to get
17372 started more quickly on a new system by using @code{gdbserver}.
17373 Finally, if you develop code for real-time systems, you may find that
17374 the tradeoffs involved in real-time operation make it more convenient to
17375 do as much development work as possible on another system, for example
17376 by cross-compiling. You can use @code{gdbserver} to make a similar
17377 choice for debugging.
17379 @value{GDBN} and @code{gdbserver} communicate via either a serial line
17380 or a TCP connection, using the standard @value{GDBN} remote serial
17384 @emph{Warning:} @code{gdbserver} does not have any built-in security.
17385 Do not run @code{gdbserver} connected to any public network; a
17386 @value{GDBN} connection to @code{gdbserver} provides access to the
17387 target system with the same privileges as the user running
17391 @subsection Running @code{gdbserver}
17392 @cindex arguments, to @code{gdbserver}
17393 @cindex @code{gdbserver}, command-line arguments
17395 Run @code{gdbserver} on the target system. You need a copy of the
17396 program you want to debug, including any libraries it requires.
17397 @code{gdbserver} does not need your program's symbol table, so you can
17398 strip the program if necessary to save space. @value{GDBN} on the host
17399 system does all the symbol handling.
17401 To use the server, you must tell it how to communicate with @value{GDBN};
17402 the name of your program; and the arguments for your program. The usual
17406 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
17409 @var{comm} is either a device name (to use a serial line), or a TCP
17410 hostname and portnumber, or @code{-} or @code{stdio} to use
17411 stdin/stdout of @code{gdbserver}.
17412 For example, to debug Emacs with the argument
17413 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
17417 target> gdbserver /dev/com1 emacs foo.txt
17420 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
17423 To use a TCP connection instead of a serial line:
17426 target> gdbserver host:2345 emacs foo.txt
17429 The only difference from the previous example is the first argument,
17430 specifying that you are communicating with the host @value{GDBN} via
17431 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
17432 expect a TCP connection from machine @samp{host} to local TCP port 2345.
17433 (Currently, the @samp{host} part is ignored.) You can choose any number
17434 you want for the port number as long as it does not conflict with any
17435 TCP ports already in use on the target system (for example, @code{23} is
17436 reserved for @code{telnet}).@footnote{If you choose a port number that
17437 conflicts with another service, @code{gdbserver} prints an error message
17438 and exits.} You must use the same port number with the host @value{GDBN}
17439 @code{target remote} command.
17441 The @code{stdio} connection is useful when starting @code{gdbserver}
17445 (gdb) target remote | ssh -T hostname gdbserver - hello
17448 The @samp{-T} option to ssh is provided because we don't need a remote pty,
17449 and we don't want escape-character handling. Ssh does this by default when
17450 a command is provided, the flag is provided to make it explicit.
17451 You could elide it if you want to.
17453 Programs started with stdio-connected gdbserver have @file{/dev/null} for
17454 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
17455 display through a pipe connected to gdbserver.
17456 Both @code{stdout} and @code{stderr} use the same pipe.
17458 @subsubsection Attaching to a Running Program
17459 @cindex attach to a program, @code{gdbserver}
17460 @cindex @option{--attach}, @code{gdbserver} option
17462 On some targets, @code{gdbserver} can also attach to running programs.
17463 This is accomplished via the @code{--attach} argument. The syntax is:
17466 target> gdbserver --attach @var{comm} @var{pid}
17469 @var{pid} is the process ID of a currently running process. It isn't necessary
17470 to point @code{gdbserver} at a binary for the running process.
17473 You can debug processes by name instead of process ID if your target has the
17474 @code{pidof} utility:
17477 target> gdbserver --attach @var{comm} `pidof @var{program}`
17480 In case more than one copy of @var{program} is running, or @var{program}
17481 has multiple threads, most versions of @code{pidof} support the
17482 @code{-s} option to only return the first process ID.
17484 @subsubsection Multi-Process Mode for @code{gdbserver}
17485 @cindex @code{gdbserver}, multiple processes
17486 @cindex multiple processes with @code{gdbserver}
17488 When you connect to @code{gdbserver} using @code{target remote},
17489 @code{gdbserver} debugs the specified program only once. When the
17490 program exits, or you detach from it, @value{GDBN} closes the connection
17491 and @code{gdbserver} exits.
17493 If you connect using @kbd{target extended-remote}, @code{gdbserver}
17494 enters multi-process mode. When the debugged program exits, or you
17495 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
17496 though no program is running. The @code{run} and @code{attach}
17497 commands instruct @code{gdbserver} to run or attach to a new program.
17498 The @code{run} command uses @code{set remote exec-file} (@pxref{set
17499 remote exec-file}) to select the program to run. Command line
17500 arguments are supported, except for wildcard expansion and I/O
17501 redirection (@pxref{Arguments}).
17503 @cindex @option{--multi}, @code{gdbserver} option
17504 To start @code{gdbserver} without supplying an initial command to run
17505 or process ID to attach, use the @option{--multi} command line option.
17506 Then you can connect using @kbd{target extended-remote} and start
17507 the program you want to debug.
17509 In multi-process mode @code{gdbserver} does not automatically exit unless you
17510 use the option @option{--once}. You can terminate it by using
17511 @code{monitor exit} (@pxref{Monitor Commands for gdbserver}). Note that the
17512 conditions under which @code{gdbserver} terminates depend on how @value{GDBN}
17513 connects to it (@kbd{target remote} or @kbd{target extended-remote}). The
17514 @option{--multi} option to @code{gdbserver} has no influence on that.
17516 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
17518 This section applies only when @code{gdbserver} is run to listen on a TCP port.
17520 @code{gdbserver} normally terminates after all of its debugged processes have
17521 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
17522 extended-remote}, @code{gdbserver} stays running even with no processes left.
17523 @value{GDBN} normally terminates the spawned debugged process on its exit,
17524 which normally also terminates @code{gdbserver} in the @kbd{target remote}
17525 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
17526 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
17527 stays running even in the @kbd{target remote} mode.
17529 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
17530 Such reconnecting is useful for features like @ref{disconnected tracing}. For
17531 completeness, at most one @value{GDBN} can be connected at a time.
17533 @cindex @option{--once}, @code{gdbserver} option
17534 By default, @code{gdbserver} keeps the listening TCP port open, so that
17535 additional connections are possible. However, if you start @code{gdbserver}
17536 with the @option{--once} option, it will stop listening for any further
17537 connection attempts after connecting to the first @value{GDBN} session. This
17538 means no further connections to @code{gdbserver} will be possible after the
17539 first one. It also means @code{gdbserver} will terminate after the first
17540 connection with remote @value{GDBN} has closed, even for unexpectedly closed
17541 connections and even in the @kbd{target extended-remote} mode. The
17542 @option{--once} option allows reusing the same port number for connecting to
17543 multiple instances of @code{gdbserver} running on the same host, since each
17544 instance closes its port after the first connection.
17546 @subsubsection Other Command-Line Arguments for @code{gdbserver}
17548 @cindex @option{--debug}, @code{gdbserver} option
17549 The @option{--debug} option tells @code{gdbserver} to display extra
17550 status information about the debugging process.
17551 @cindex @option{--remote-debug}, @code{gdbserver} option
17552 The @option{--remote-debug} option tells @code{gdbserver} to display
17553 remote protocol debug output. These options are intended for
17554 @code{gdbserver} development and for bug reports to the developers.
17556 @cindex @option{--wrapper}, @code{gdbserver} option
17557 The @option{--wrapper} option specifies a wrapper to launch programs
17558 for debugging. The option should be followed by the name of the
17559 wrapper, then any command-line arguments to pass to the wrapper, then
17560 @kbd{--} indicating the end of the wrapper arguments.
17562 @code{gdbserver} runs the specified wrapper program with a combined
17563 command line including the wrapper arguments, then the name of the
17564 program to debug, then any arguments to the program. The wrapper
17565 runs until it executes your program, and then @value{GDBN} gains control.
17567 You can use any program that eventually calls @code{execve} with
17568 its arguments as a wrapper. Several standard Unix utilities do
17569 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
17570 with @code{exec "$@@"} will also work.
17572 For example, you can use @code{env} to pass an environment variable to
17573 the debugged program, without setting the variable in @code{gdbserver}'s
17577 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
17580 @subsection Connecting to @code{gdbserver}
17582 Run @value{GDBN} on the host system.
17584 First make sure you have the necessary symbol files. Load symbols for
17585 your application using the @code{file} command before you connect. Use
17586 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
17587 was compiled with the correct sysroot using @code{--with-sysroot}).
17589 The symbol file and target libraries must exactly match the executable
17590 and libraries on the target, with one exception: the files on the host
17591 system should not be stripped, even if the files on the target system
17592 are. Mismatched or missing files will lead to confusing results
17593 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
17594 files may also prevent @code{gdbserver} from debugging multi-threaded
17597 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
17598 For TCP connections, you must start up @code{gdbserver} prior to using
17599 the @code{target remote} command. Otherwise you may get an error whose
17600 text depends on the host system, but which usually looks something like
17601 @samp{Connection refused}. Don't use the @code{load}
17602 command in @value{GDBN} when using @code{gdbserver}, since the program is
17603 already on the target.
17605 @subsection Monitor Commands for @code{gdbserver}
17606 @cindex monitor commands, for @code{gdbserver}
17607 @anchor{Monitor Commands for gdbserver}
17609 During a @value{GDBN} session using @code{gdbserver}, you can use the
17610 @code{monitor} command to send special requests to @code{gdbserver}.
17611 Here are the available commands.
17615 List the available monitor commands.
17617 @item monitor set debug 0
17618 @itemx monitor set debug 1
17619 Disable or enable general debugging messages.
17621 @item monitor set remote-debug 0
17622 @itemx monitor set remote-debug 1
17623 Disable or enable specific debugging messages associated with the remote
17624 protocol (@pxref{Remote Protocol}).
17626 @item monitor set libthread-db-search-path [PATH]
17627 @cindex gdbserver, search path for @code{libthread_db}
17628 When this command is issued, @var{path} is a colon-separated list of
17629 directories to search for @code{libthread_db} (@pxref{Threads,,set
17630 libthread-db-search-path}). If you omit @var{path},
17631 @samp{libthread-db-search-path} will be reset to its default value.
17633 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
17634 not supported in @code{gdbserver}.
17637 Tell gdbserver to exit immediately. This command should be followed by
17638 @code{disconnect} to close the debugging session. @code{gdbserver} will
17639 detach from any attached processes and kill any processes it created.
17640 Use @code{monitor exit} to terminate @code{gdbserver} at the end
17641 of a multi-process mode debug session.
17645 @subsection Tracepoints support in @code{gdbserver}
17646 @cindex tracepoints support in @code{gdbserver}
17648 On some targets, @code{gdbserver} supports tracepoints, fast
17649 tracepoints and static tracepoints.
17651 For fast or static tracepoints to work, a special library called the
17652 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
17653 This library is built and distributed as an integral part of
17654 @code{gdbserver}. In addition, support for static tracepoints
17655 requires building the in-process agent library with static tracepoints
17656 support. At present, the UST (LTTng Userspace Tracer,
17657 @url{http://lttng.org/ust}) tracing engine is supported. This support
17658 is automatically available if UST development headers are found in the
17659 standard include path when @code{gdbserver} is built, or if
17660 @code{gdbserver} was explicitly configured using @option{--with-ust}
17661 to point at such headers. You can explicitly disable the support
17662 using @option{--with-ust=no}.
17664 There are several ways to load the in-process agent in your program:
17667 @item Specifying it as dependency at link time
17669 You can link your program dynamically with the in-process agent
17670 library. On most systems, this is accomplished by adding
17671 @code{-linproctrace} to the link command.
17673 @item Using the system's preloading mechanisms
17675 You can force loading the in-process agent at startup time by using
17676 your system's support for preloading shared libraries. Many Unixes
17677 support the concept of preloading user defined libraries. In most
17678 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
17679 in the environment. See also the description of @code{gdbserver}'s
17680 @option{--wrapper} command line option.
17682 @item Using @value{GDBN} to force loading the agent at run time
17684 On some systems, you can force the inferior to load a shared library,
17685 by calling a dynamic loader function in the inferior that takes care
17686 of dynamically looking up and loading a shared library. On most Unix
17687 systems, the function is @code{dlopen}. You'll use the @code{call}
17688 command for that. For example:
17691 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
17694 Note that on most Unix systems, for the @code{dlopen} function to be
17695 available, the program needs to be linked with @code{-ldl}.
17698 On systems that have a userspace dynamic loader, like most Unix
17699 systems, when you connect to @code{gdbserver} using @code{target
17700 remote}, you'll find that the program is stopped at the dynamic
17701 loader's entry point, and no shared library has been loaded in the
17702 program's address space yet, including the in-process agent. In that
17703 case, before being able to use any of the fast or static tracepoints
17704 features, you need to let the loader run and load the shared
17705 libraries. The simplest way to do that is to run the program to the
17706 main procedure. E.g., if debugging a C or C@t{++} program, start
17707 @code{gdbserver} like so:
17710 $ gdbserver :9999 myprogram
17713 Start GDB and connect to @code{gdbserver} like so, and run to main:
17717 (@value{GDBP}) target remote myhost:9999
17718 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
17719 (@value{GDBP}) b main
17720 (@value{GDBP}) continue
17723 The in-process tracing agent library should now be loaded into the
17724 process; you can confirm it with the @code{info sharedlibrary}
17725 command, which will list @file{libinproctrace.so} as loaded in the
17726 process. You are now ready to install fast tracepoints, list static
17727 tracepoint markers, probe static tracepoints markers, and start
17730 @node Remote Configuration
17731 @section Remote Configuration
17734 @kindex show remote
17735 This section documents the configuration options available when
17736 debugging remote programs. For the options related to the File I/O
17737 extensions of the remote protocol, see @ref{system,
17738 system-call-allowed}.
17741 @item set remoteaddresssize @var{bits}
17742 @cindex address size for remote targets
17743 @cindex bits in remote address
17744 Set the maximum size of address in a memory packet to the specified
17745 number of bits. @value{GDBN} will mask off the address bits above
17746 that number, when it passes addresses to the remote target. The
17747 default value is the number of bits in the target's address.
17749 @item show remoteaddresssize
17750 Show the current value of remote address size in bits.
17752 @item set remotebaud @var{n}
17753 @cindex baud rate for remote targets
17754 Set the baud rate for the remote serial I/O to @var{n} baud. The
17755 value is used to set the speed of the serial port used for debugging
17758 @item show remotebaud
17759 Show the current speed of the remote connection.
17761 @item set remotebreak
17762 @cindex interrupt remote programs
17763 @cindex BREAK signal instead of Ctrl-C
17764 @anchor{set remotebreak}
17765 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
17766 when you type @kbd{Ctrl-c} to interrupt the program running
17767 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
17768 character instead. The default is off, since most remote systems
17769 expect to see @samp{Ctrl-C} as the interrupt signal.
17771 @item show remotebreak
17772 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
17773 interrupt the remote program.
17775 @item set remoteflow on
17776 @itemx set remoteflow off
17777 @kindex set remoteflow
17778 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
17779 on the serial port used to communicate to the remote target.
17781 @item show remoteflow
17782 @kindex show remoteflow
17783 Show the current setting of hardware flow control.
17785 @item set remotelogbase @var{base}
17786 Set the base (a.k.a.@: radix) of logging serial protocol
17787 communications to @var{base}. Supported values of @var{base} are:
17788 @code{ascii}, @code{octal}, and @code{hex}. The default is
17791 @item show remotelogbase
17792 Show the current setting of the radix for logging remote serial
17795 @item set remotelogfile @var{file}
17796 @cindex record serial communications on file
17797 Record remote serial communications on the named @var{file}. The
17798 default is not to record at all.
17800 @item show remotelogfile.
17801 Show the current setting of the file name on which to record the
17802 serial communications.
17804 @item set remotetimeout @var{num}
17805 @cindex timeout for serial communications
17806 @cindex remote timeout
17807 Set the timeout limit to wait for the remote target to respond to
17808 @var{num} seconds. The default is 2 seconds.
17810 @item show remotetimeout
17811 Show the current number of seconds to wait for the remote target
17814 @cindex limit hardware breakpoints and watchpoints
17815 @cindex remote target, limit break- and watchpoints
17816 @anchor{set remote hardware-watchpoint-limit}
17817 @anchor{set remote hardware-breakpoint-limit}
17818 @item set remote hardware-watchpoint-limit @var{limit}
17819 @itemx set remote hardware-breakpoint-limit @var{limit}
17820 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
17821 watchpoints. A limit of -1, the default, is treated as unlimited.
17823 @cindex limit hardware watchpoints length
17824 @cindex remote target, limit watchpoints length
17825 @anchor{set remote hardware-watchpoint-length-limit}
17826 @item set remote hardware-watchpoint-length-limit @var{limit}
17827 Restrict @value{GDBN} to using @var{limit} bytes for the maximum length of
17828 a remote hardware watchpoint. A limit of -1, the default, is treated
17831 @item show remote hardware-watchpoint-length-limit
17832 Show the current limit (in bytes) of the maximum length of
17833 a remote hardware watchpoint.
17835 @item set remote exec-file @var{filename}
17836 @itemx show remote exec-file
17837 @anchor{set remote exec-file}
17838 @cindex executable file, for remote target
17839 Select the file used for @code{run} with @code{target
17840 extended-remote}. This should be set to a filename valid on the
17841 target system. If it is not set, the target will use a default
17842 filename (e.g.@: the last program run).
17844 @item set remote interrupt-sequence
17845 @cindex interrupt remote programs
17846 @cindex select Ctrl-C, BREAK or BREAK-g
17847 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
17848 @samp{BREAK-g} as the
17849 sequence to the remote target in order to interrupt the execution.
17850 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
17851 is high level of serial line for some certain time.
17852 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
17853 It is @code{BREAK} signal followed by character @code{g}.
17855 @item show interrupt-sequence
17856 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
17857 is sent by @value{GDBN} to interrupt the remote program.
17858 @code{BREAK-g} is BREAK signal followed by @code{g} and
17859 also known as Magic SysRq g.
17861 @item set remote interrupt-on-connect
17862 @cindex send interrupt-sequence on start
17863 Specify whether interrupt-sequence is sent to remote target when
17864 @value{GDBN} connects to it. This is mostly needed when you debug
17865 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
17866 which is known as Magic SysRq g in order to connect @value{GDBN}.
17868 @item show interrupt-on-connect
17869 Show whether interrupt-sequence is sent
17870 to remote target when @value{GDBN} connects to it.
17874 @item set tcp auto-retry on
17875 @cindex auto-retry, for remote TCP target
17876 Enable auto-retry for remote TCP connections. This is useful if the remote
17877 debugging agent is launched in parallel with @value{GDBN}; there is a race
17878 condition because the agent may not become ready to accept the connection
17879 before @value{GDBN} attempts to connect. When auto-retry is
17880 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
17881 to establish the connection using the timeout specified by
17882 @code{set tcp connect-timeout}.
17884 @item set tcp auto-retry off
17885 Do not auto-retry failed TCP connections.
17887 @item show tcp auto-retry
17888 Show the current auto-retry setting.
17890 @item set tcp connect-timeout @var{seconds}
17891 @cindex connection timeout, for remote TCP target
17892 @cindex timeout, for remote target connection
17893 Set the timeout for establishing a TCP connection to the remote target to
17894 @var{seconds}. The timeout affects both polling to retry failed connections
17895 (enabled by @code{set tcp auto-retry on}) and waiting for connections
17896 that are merely slow to complete, and represents an approximate cumulative
17899 @item show tcp connect-timeout
17900 Show the current connection timeout setting.
17903 @cindex remote packets, enabling and disabling
17904 The @value{GDBN} remote protocol autodetects the packets supported by
17905 your debugging stub. If you need to override the autodetection, you
17906 can use these commands to enable or disable individual packets. Each
17907 packet can be set to @samp{on} (the remote target supports this
17908 packet), @samp{off} (the remote target does not support this packet),
17909 or @samp{auto} (detect remote target support for this packet). They
17910 all default to @samp{auto}. For more information about each packet,
17911 see @ref{Remote Protocol}.
17913 During normal use, you should not have to use any of these commands.
17914 If you do, that may be a bug in your remote debugging stub, or a bug
17915 in @value{GDBN}. You may want to report the problem to the
17916 @value{GDBN} developers.
17918 For each packet @var{name}, the command to enable or disable the
17919 packet is @code{set remote @var{name}-packet}. The available settings
17922 @multitable @columnfractions 0.28 0.32 0.25
17925 @tab Related Features
17927 @item @code{fetch-register}
17929 @tab @code{info registers}
17931 @item @code{set-register}
17935 @item @code{binary-download}
17937 @tab @code{load}, @code{set}
17939 @item @code{read-aux-vector}
17940 @tab @code{qXfer:auxv:read}
17941 @tab @code{info auxv}
17943 @item @code{symbol-lookup}
17944 @tab @code{qSymbol}
17945 @tab Detecting multiple threads
17947 @item @code{attach}
17948 @tab @code{vAttach}
17951 @item @code{verbose-resume}
17953 @tab Stepping or resuming multiple threads
17959 @item @code{software-breakpoint}
17963 @item @code{hardware-breakpoint}
17967 @item @code{write-watchpoint}
17971 @item @code{read-watchpoint}
17975 @item @code{access-watchpoint}
17979 @item @code{target-features}
17980 @tab @code{qXfer:features:read}
17981 @tab @code{set architecture}
17983 @item @code{library-info}
17984 @tab @code{qXfer:libraries:read}
17985 @tab @code{info sharedlibrary}
17987 @item @code{memory-map}
17988 @tab @code{qXfer:memory-map:read}
17989 @tab @code{info mem}
17991 @item @code{read-sdata-object}
17992 @tab @code{qXfer:sdata:read}
17993 @tab @code{print $_sdata}
17995 @item @code{read-spu-object}
17996 @tab @code{qXfer:spu:read}
17997 @tab @code{info spu}
17999 @item @code{write-spu-object}
18000 @tab @code{qXfer:spu:write}
18001 @tab @code{info spu}
18003 @item @code{read-siginfo-object}
18004 @tab @code{qXfer:siginfo:read}
18005 @tab @code{print $_siginfo}
18007 @item @code{write-siginfo-object}
18008 @tab @code{qXfer:siginfo:write}
18009 @tab @code{set $_siginfo}
18011 @item @code{threads}
18012 @tab @code{qXfer:threads:read}
18013 @tab @code{info threads}
18015 @item @code{get-thread-local-@*storage-address}
18016 @tab @code{qGetTLSAddr}
18017 @tab Displaying @code{__thread} variables
18019 @item @code{get-thread-information-block-address}
18020 @tab @code{qGetTIBAddr}
18021 @tab Display MS-Windows Thread Information Block.
18023 @item @code{search-memory}
18024 @tab @code{qSearch:memory}
18027 @item @code{supported-packets}
18028 @tab @code{qSupported}
18029 @tab Remote communications parameters
18031 @item @code{pass-signals}
18032 @tab @code{QPassSignals}
18033 @tab @code{handle @var{signal}}
18035 @item @code{program-signals}
18036 @tab @code{QProgramSignals}
18037 @tab @code{handle @var{signal}}
18039 @item @code{hostio-close-packet}
18040 @tab @code{vFile:close}
18041 @tab @code{remote get}, @code{remote put}
18043 @item @code{hostio-open-packet}
18044 @tab @code{vFile:open}
18045 @tab @code{remote get}, @code{remote put}
18047 @item @code{hostio-pread-packet}
18048 @tab @code{vFile:pread}
18049 @tab @code{remote get}, @code{remote put}
18051 @item @code{hostio-pwrite-packet}
18052 @tab @code{vFile:pwrite}
18053 @tab @code{remote get}, @code{remote put}
18055 @item @code{hostio-unlink-packet}
18056 @tab @code{vFile:unlink}
18057 @tab @code{remote delete}
18059 @item @code{hostio-readlink-packet}
18060 @tab @code{vFile:readlink}
18063 @item @code{noack-packet}
18064 @tab @code{QStartNoAckMode}
18065 @tab Packet acknowledgment
18067 @item @code{osdata}
18068 @tab @code{qXfer:osdata:read}
18069 @tab @code{info os}
18071 @item @code{query-attached}
18072 @tab @code{qAttached}
18073 @tab Querying remote process attach state.
18075 @item @code{traceframe-info}
18076 @tab @code{qXfer:traceframe-info:read}
18077 @tab Traceframe info
18079 @item @code{install-in-trace}
18080 @tab @code{InstallInTrace}
18081 @tab Install tracepoint in tracing
18083 @item @code{disable-randomization}
18084 @tab @code{QDisableRandomization}
18085 @tab @code{set disable-randomization}
18087 @item @code{conditional-breakpoints-packet}
18088 @tab @code{Z0 and Z1}
18089 @tab @code{Support for target-side breakpoint condition evaluation}
18093 @section Implementing a Remote Stub
18095 @cindex debugging stub, example
18096 @cindex remote stub, example
18097 @cindex stub example, remote debugging
18098 The stub files provided with @value{GDBN} implement the target side of the
18099 communication protocol, and the @value{GDBN} side is implemented in the
18100 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
18101 these subroutines to communicate, and ignore the details. (If you're
18102 implementing your own stub file, you can still ignore the details: start
18103 with one of the existing stub files. @file{sparc-stub.c} is the best
18104 organized, and therefore the easiest to read.)
18106 @cindex remote serial debugging, overview
18107 To debug a program running on another machine (the debugging
18108 @dfn{target} machine), you must first arrange for all the usual
18109 prerequisites for the program to run by itself. For example, for a C
18114 A startup routine to set up the C runtime environment; these usually
18115 have a name like @file{crt0}. The startup routine may be supplied by
18116 your hardware supplier, or you may have to write your own.
18119 A C subroutine library to support your program's
18120 subroutine calls, notably managing input and output.
18123 A way of getting your program to the other machine---for example, a
18124 download program. These are often supplied by the hardware
18125 manufacturer, but you may have to write your own from hardware
18129 The next step is to arrange for your program to use a serial port to
18130 communicate with the machine where @value{GDBN} is running (the @dfn{host}
18131 machine). In general terms, the scheme looks like this:
18135 @value{GDBN} already understands how to use this protocol; when everything
18136 else is set up, you can simply use the @samp{target remote} command
18137 (@pxref{Targets,,Specifying a Debugging Target}).
18139 @item On the target,
18140 you must link with your program a few special-purpose subroutines that
18141 implement the @value{GDBN} remote serial protocol. The file containing these
18142 subroutines is called a @dfn{debugging stub}.
18144 On certain remote targets, you can use an auxiliary program
18145 @code{gdbserver} instead of linking a stub into your program.
18146 @xref{Server,,Using the @code{gdbserver} Program}, for details.
18149 The debugging stub is specific to the architecture of the remote
18150 machine; for example, use @file{sparc-stub.c} to debug programs on
18153 @cindex remote serial stub list
18154 These working remote stubs are distributed with @value{GDBN}:
18159 @cindex @file{i386-stub.c}
18162 For Intel 386 and compatible architectures.
18165 @cindex @file{m68k-stub.c}
18166 @cindex Motorola 680x0
18168 For Motorola 680x0 architectures.
18171 @cindex @file{sh-stub.c}
18174 For Renesas SH architectures.
18177 @cindex @file{sparc-stub.c}
18179 For @sc{sparc} architectures.
18181 @item sparcl-stub.c
18182 @cindex @file{sparcl-stub.c}
18185 For Fujitsu @sc{sparclite} architectures.
18189 The @file{README} file in the @value{GDBN} distribution may list other
18190 recently added stubs.
18193 * Stub Contents:: What the stub can do for you
18194 * Bootstrapping:: What you must do for the stub
18195 * Debug Session:: Putting it all together
18198 @node Stub Contents
18199 @subsection What the Stub Can Do for You
18201 @cindex remote serial stub
18202 The debugging stub for your architecture supplies these three
18206 @item set_debug_traps
18207 @findex set_debug_traps
18208 @cindex remote serial stub, initialization
18209 This routine arranges for @code{handle_exception} to run when your
18210 program stops. You must call this subroutine explicitly in your
18211 program's startup code.
18213 @item handle_exception
18214 @findex handle_exception
18215 @cindex remote serial stub, main routine
18216 This is the central workhorse, but your program never calls it
18217 explicitly---the setup code arranges for @code{handle_exception} to
18218 run when a trap is triggered.
18220 @code{handle_exception} takes control when your program stops during
18221 execution (for example, on a breakpoint), and mediates communications
18222 with @value{GDBN} on the host machine. This is where the communications
18223 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
18224 representative on the target machine. It begins by sending summary
18225 information on the state of your program, then continues to execute,
18226 retrieving and transmitting any information @value{GDBN} needs, until you
18227 execute a @value{GDBN} command that makes your program resume; at that point,
18228 @code{handle_exception} returns control to your own code on the target
18232 @cindex @code{breakpoint} subroutine, remote
18233 Use this auxiliary subroutine to make your program contain a
18234 breakpoint. Depending on the particular situation, this may be the only
18235 way for @value{GDBN} to get control. For instance, if your target
18236 machine has some sort of interrupt button, you won't need to call this;
18237 pressing the interrupt button transfers control to
18238 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
18239 simply receiving characters on the serial port may also trigger a trap;
18240 again, in that situation, you don't need to call @code{breakpoint} from
18241 your own program---simply running @samp{target remote} from the host
18242 @value{GDBN} session gets control.
18244 Call @code{breakpoint} if none of these is true, or if you simply want
18245 to make certain your program stops at a predetermined point for the
18246 start of your debugging session.
18249 @node Bootstrapping
18250 @subsection What You Must Do for the Stub
18252 @cindex remote stub, support routines
18253 The debugging stubs that come with @value{GDBN} are set up for a particular
18254 chip architecture, but they have no information about the rest of your
18255 debugging target machine.
18257 First of all you need to tell the stub how to communicate with the
18261 @item int getDebugChar()
18262 @findex getDebugChar
18263 Write this subroutine to read a single character from the serial port.
18264 It may be identical to @code{getchar} for your target system; a
18265 different name is used to allow you to distinguish the two if you wish.
18267 @item void putDebugChar(int)
18268 @findex putDebugChar
18269 Write this subroutine to write a single character to the serial port.
18270 It may be identical to @code{putchar} for your target system; a
18271 different name is used to allow you to distinguish the two if you wish.
18274 @cindex control C, and remote debugging
18275 @cindex interrupting remote targets
18276 If you want @value{GDBN} to be able to stop your program while it is
18277 running, you need to use an interrupt-driven serial driver, and arrange
18278 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
18279 character). That is the character which @value{GDBN} uses to tell the
18280 remote system to stop.
18282 Getting the debugging target to return the proper status to @value{GDBN}
18283 probably requires changes to the standard stub; one quick and dirty way
18284 is to just execute a breakpoint instruction (the ``dirty'' part is that
18285 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
18287 Other routines you need to supply are:
18290 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
18291 @findex exceptionHandler
18292 Write this function to install @var{exception_address} in the exception
18293 handling tables. You need to do this because the stub does not have any
18294 way of knowing what the exception handling tables on your target system
18295 are like (for example, the processor's table might be in @sc{rom},
18296 containing entries which point to a table in @sc{ram}).
18297 @var{exception_number} is the exception number which should be changed;
18298 its meaning is architecture-dependent (for example, different numbers
18299 might represent divide by zero, misaligned access, etc). When this
18300 exception occurs, control should be transferred directly to
18301 @var{exception_address}, and the processor state (stack, registers,
18302 and so on) should be just as it is when a processor exception occurs. So if
18303 you want to use a jump instruction to reach @var{exception_address}, it
18304 should be a simple jump, not a jump to subroutine.
18306 For the 386, @var{exception_address} should be installed as an interrupt
18307 gate so that interrupts are masked while the handler runs. The gate
18308 should be at privilege level 0 (the most privileged level). The
18309 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
18310 help from @code{exceptionHandler}.
18312 @item void flush_i_cache()
18313 @findex flush_i_cache
18314 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
18315 instruction cache, if any, on your target machine. If there is no
18316 instruction cache, this subroutine may be a no-op.
18318 On target machines that have instruction caches, @value{GDBN} requires this
18319 function to make certain that the state of your program is stable.
18323 You must also make sure this library routine is available:
18326 @item void *memset(void *, int, int)
18328 This is the standard library function @code{memset} that sets an area of
18329 memory to a known value. If you have one of the free versions of
18330 @code{libc.a}, @code{memset} can be found there; otherwise, you must
18331 either obtain it from your hardware manufacturer, or write your own.
18334 If you do not use the GNU C compiler, you may need other standard
18335 library subroutines as well; this varies from one stub to another,
18336 but in general the stubs are likely to use any of the common library
18337 subroutines which @code{@value{NGCC}} generates as inline code.
18340 @node Debug Session
18341 @subsection Putting it All Together
18343 @cindex remote serial debugging summary
18344 In summary, when your program is ready to debug, you must follow these
18349 Make sure you have defined the supporting low-level routines
18350 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
18352 @code{getDebugChar}, @code{putDebugChar},
18353 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
18357 Insert these lines in your program's startup code, before the main
18358 procedure is called:
18365 On some machines, when a breakpoint trap is raised, the hardware
18366 automatically makes the PC point to the instruction after the
18367 breakpoint. If your machine doesn't do that, you may need to adjust
18368 @code{handle_exception} to arrange for it to return to the instruction
18369 after the breakpoint on this first invocation, so that your program
18370 doesn't keep hitting the initial breakpoint instead of making
18374 For the 680x0 stub only, you need to provide a variable called
18375 @code{exceptionHook}. Normally you just use:
18378 void (*exceptionHook)() = 0;
18382 but if before calling @code{set_debug_traps}, you set it to point to a
18383 function in your program, that function is called when
18384 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
18385 error). The function indicated by @code{exceptionHook} is called with
18386 one parameter: an @code{int} which is the exception number.
18389 Compile and link together: your program, the @value{GDBN} debugging stub for
18390 your target architecture, and the supporting subroutines.
18393 Make sure you have a serial connection between your target machine and
18394 the @value{GDBN} host, and identify the serial port on the host.
18397 @c The "remote" target now provides a `load' command, so we should
18398 @c document that. FIXME.
18399 Download your program to your target machine (or get it there by
18400 whatever means the manufacturer provides), and start it.
18403 Start @value{GDBN} on the host, and connect to the target
18404 (@pxref{Connecting,,Connecting to a Remote Target}).
18408 @node Configurations
18409 @chapter Configuration-Specific Information
18411 While nearly all @value{GDBN} commands are available for all native and
18412 cross versions of the debugger, there are some exceptions. This chapter
18413 describes things that are only available in certain configurations.
18415 There are three major categories of configurations: native
18416 configurations, where the host and target are the same, embedded
18417 operating system configurations, which are usually the same for several
18418 different processor architectures, and bare embedded processors, which
18419 are quite different from each other.
18424 * Embedded Processors::
18431 This section describes details specific to particular native
18436 * BSD libkvm Interface:: Debugging BSD kernel memory images
18437 * SVR4 Process Information:: SVR4 process information
18438 * DJGPP Native:: Features specific to the DJGPP port
18439 * Cygwin Native:: Features specific to the Cygwin port
18440 * Hurd Native:: Features specific to @sc{gnu} Hurd
18441 * Neutrino:: Features specific to QNX Neutrino
18442 * Darwin:: Features specific to Darwin
18448 On HP-UX systems, if you refer to a function or variable name that
18449 begins with a dollar sign, @value{GDBN} searches for a user or system
18450 name first, before it searches for a convenience variable.
18453 @node BSD libkvm Interface
18454 @subsection BSD libkvm Interface
18457 @cindex kernel memory image
18458 @cindex kernel crash dump
18460 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
18461 interface that provides a uniform interface for accessing kernel virtual
18462 memory images, including live systems and crash dumps. @value{GDBN}
18463 uses this interface to allow you to debug live kernels and kernel crash
18464 dumps on many native BSD configurations. This is implemented as a
18465 special @code{kvm} debugging target. For debugging a live system, load
18466 the currently running kernel into @value{GDBN} and connect to the
18470 (@value{GDBP}) @b{target kvm}
18473 For debugging crash dumps, provide the file name of the crash dump as an
18477 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
18480 Once connected to the @code{kvm} target, the following commands are
18486 Set current context from the @dfn{Process Control Block} (PCB) address.
18489 Set current context from proc address. This command isn't available on
18490 modern FreeBSD systems.
18493 @node SVR4 Process Information
18494 @subsection SVR4 Process Information
18496 @cindex examine process image
18497 @cindex process info via @file{/proc}
18499 Many versions of SVR4 and compatible systems provide a facility called
18500 @samp{/proc} that can be used to examine the image of a running
18501 process using file-system subroutines. If @value{GDBN} is configured
18502 for an operating system with this facility, the command @code{info
18503 proc} is available to report information about the process running
18504 your program, or about any process running on your system. @code{info
18505 proc} works only on SVR4 systems that include the @code{procfs} code.
18506 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
18507 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
18513 @itemx info proc @var{process-id}
18514 Summarize available information about any running process. If a
18515 process ID is specified by @var{process-id}, display information about
18516 that process; otherwise display information about the program being
18517 debugged. The summary includes the debugged process ID, the command
18518 line used to invoke it, its current working directory, and its
18519 executable file's absolute file name.
18521 On some systems, @var{process-id} can be of the form
18522 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
18523 within a process. If the optional @var{pid} part is missing, it means
18524 a thread from the process being debugged (the leading @samp{/} still
18525 needs to be present, or else @value{GDBN} will interpret the number as
18526 a process ID rather than a thread ID).
18528 @item info proc mappings
18529 @cindex memory address space mappings
18530 Report the memory address space ranges accessible in the program, with
18531 information on whether the process has read, write, or execute access
18532 rights to each range. On @sc{gnu}/Linux systems, each memory range
18533 includes the object file which is mapped to that range, instead of the
18534 memory access rights to that range.
18536 @item info proc stat
18537 @itemx info proc status
18538 @cindex process detailed status information
18539 These subcommands are specific to @sc{gnu}/Linux systems. They show
18540 the process-related information, including the user ID and group ID;
18541 how many threads are there in the process; its virtual memory usage;
18542 the signals that are pending, blocked, and ignored; its TTY; its
18543 consumption of system and user time; its stack size; its @samp{nice}
18544 value; etc. For more information, see the @samp{proc} man page
18545 (type @kbd{man 5 proc} from your shell prompt).
18547 @item info proc all
18548 Show all the information about the process described under all of the
18549 above @code{info proc} subcommands.
18552 @comment These sub-options of 'info proc' were not included when
18553 @comment procfs.c was re-written. Keep their descriptions around
18554 @comment against the day when someone finds the time to put them back in.
18555 @kindex info proc times
18556 @item info proc times
18557 Starting time, user CPU time, and system CPU time for your program and
18560 @kindex info proc id
18562 Report on the process IDs related to your program: its own process ID,
18563 the ID of its parent, the process group ID, and the session ID.
18566 @item set procfs-trace
18567 @kindex set procfs-trace
18568 @cindex @code{procfs} API calls
18569 This command enables and disables tracing of @code{procfs} API calls.
18571 @item show procfs-trace
18572 @kindex show procfs-trace
18573 Show the current state of @code{procfs} API call tracing.
18575 @item set procfs-file @var{file}
18576 @kindex set procfs-file
18577 Tell @value{GDBN} to write @code{procfs} API trace to the named
18578 @var{file}. @value{GDBN} appends the trace info to the previous
18579 contents of the file. The default is to display the trace on the
18582 @item show procfs-file
18583 @kindex show procfs-file
18584 Show the file to which @code{procfs} API trace is written.
18586 @item proc-trace-entry
18587 @itemx proc-trace-exit
18588 @itemx proc-untrace-entry
18589 @itemx proc-untrace-exit
18590 @kindex proc-trace-entry
18591 @kindex proc-trace-exit
18592 @kindex proc-untrace-entry
18593 @kindex proc-untrace-exit
18594 These commands enable and disable tracing of entries into and exits
18595 from the @code{syscall} interface.
18598 @kindex info pidlist
18599 @cindex process list, QNX Neutrino
18600 For QNX Neutrino only, this command displays the list of all the
18601 processes and all the threads within each process.
18604 @kindex info meminfo
18605 @cindex mapinfo list, QNX Neutrino
18606 For QNX Neutrino only, this command displays the list of all mapinfos.
18610 @subsection Features for Debugging @sc{djgpp} Programs
18611 @cindex @sc{djgpp} debugging
18612 @cindex native @sc{djgpp} debugging
18613 @cindex MS-DOS-specific commands
18616 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
18617 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
18618 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
18619 top of real-mode DOS systems and their emulations.
18621 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
18622 defines a few commands specific to the @sc{djgpp} port. This
18623 subsection describes those commands.
18628 This is a prefix of @sc{djgpp}-specific commands which print
18629 information about the target system and important OS structures.
18632 @cindex MS-DOS system info
18633 @cindex free memory information (MS-DOS)
18634 @item info dos sysinfo
18635 This command displays assorted information about the underlying
18636 platform: the CPU type and features, the OS version and flavor, the
18637 DPMI version, and the available conventional and DPMI memory.
18642 @cindex segment descriptor tables
18643 @cindex descriptor tables display
18645 @itemx info dos ldt
18646 @itemx info dos idt
18647 These 3 commands display entries from, respectively, Global, Local,
18648 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
18649 tables are data structures which store a descriptor for each segment
18650 that is currently in use. The segment's selector is an index into a
18651 descriptor table; the table entry for that index holds the
18652 descriptor's base address and limit, and its attributes and access
18655 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
18656 segment (used for both data and the stack), and a DOS segment (which
18657 allows access to DOS/BIOS data structures and absolute addresses in
18658 conventional memory). However, the DPMI host will usually define
18659 additional segments in order to support the DPMI environment.
18661 @cindex garbled pointers
18662 These commands allow to display entries from the descriptor tables.
18663 Without an argument, all entries from the specified table are
18664 displayed. An argument, which should be an integer expression, means
18665 display a single entry whose index is given by the argument. For
18666 example, here's a convenient way to display information about the
18667 debugged program's data segment:
18670 @exdent @code{(@value{GDBP}) info dos ldt $ds}
18671 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
18675 This comes in handy when you want to see whether a pointer is outside
18676 the data segment's limit (i.e.@: @dfn{garbled}).
18678 @cindex page tables display (MS-DOS)
18680 @itemx info dos pte
18681 These two commands display entries from, respectively, the Page
18682 Directory and the Page Tables. Page Directories and Page Tables are
18683 data structures which control how virtual memory addresses are mapped
18684 into physical addresses. A Page Table includes an entry for every
18685 page of memory that is mapped into the program's address space; there
18686 may be several Page Tables, each one holding up to 4096 entries. A
18687 Page Directory has up to 4096 entries, one each for every Page Table
18688 that is currently in use.
18690 Without an argument, @kbd{info dos pde} displays the entire Page
18691 Directory, and @kbd{info dos pte} displays all the entries in all of
18692 the Page Tables. An argument, an integer expression, given to the
18693 @kbd{info dos pde} command means display only that entry from the Page
18694 Directory table. An argument given to the @kbd{info dos pte} command
18695 means display entries from a single Page Table, the one pointed to by
18696 the specified entry in the Page Directory.
18698 @cindex direct memory access (DMA) on MS-DOS
18699 These commands are useful when your program uses @dfn{DMA} (Direct
18700 Memory Access), which needs physical addresses to program the DMA
18703 These commands are supported only with some DPMI servers.
18705 @cindex physical address from linear address
18706 @item info dos address-pte @var{addr}
18707 This command displays the Page Table entry for a specified linear
18708 address. The argument @var{addr} is a linear address which should
18709 already have the appropriate segment's base address added to it,
18710 because this command accepts addresses which may belong to @emph{any}
18711 segment. For example, here's how to display the Page Table entry for
18712 the page where a variable @code{i} is stored:
18715 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
18716 @exdent @code{Page Table entry for address 0x11a00d30:}
18717 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
18721 This says that @code{i} is stored at offset @code{0xd30} from the page
18722 whose physical base address is @code{0x02698000}, and shows all the
18723 attributes of that page.
18725 Note that you must cast the addresses of variables to a @code{char *},
18726 since otherwise the value of @code{__djgpp_base_address}, the base
18727 address of all variables and functions in a @sc{djgpp} program, will
18728 be added using the rules of C pointer arithmetics: if @code{i} is
18729 declared an @code{int}, @value{GDBN} will add 4 times the value of
18730 @code{__djgpp_base_address} to the address of @code{i}.
18732 Here's another example, it displays the Page Table entry for the
18736 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
18737 @exdent @code{Page Table entry for address 0x29110:}
18738 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
18742 (The @code{+ 3} offset is because the transfer buffer's address is the
18743 3rd member of the @code{_go32_info_block} structure.) The output
18744 clearly shows that this DPMI server maps the addresses in conventional
18745 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
18746 linear (@code{0x29110}) addresses are identical.
18748 This command is supported only with some DPMI servers.
18751 @cindex DOS serial data link, remote debugging
18752 In addition to native debugging, the DJGPP port supports remote
18753 debugging via a serial data link. The following commands are specific
18754 to remote serial debugging in the DJGPP port of @value{GDBN}.
18757 @kindex set com1base
18758 @kindex set com1irq
18759 @kindex set com2base
18760 @kindex set com2irq
18761 @kindex set com3base
18762 @kindex set com3irq
18763 @kindex set com4base
18764 @kindex set com4irq
18765 @item set com1base @var{addr}
18766 This command sets the base I/O port address of the @file{COM1} serial
18769 @item set com1irq @var{irq}
18770 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
18771 for the @file{COM1} serial port.
18773 There are similar commands @samp{set com2base}, @samp{set com3irq},
18774 etc.@: for setting the port address and the @code{IRQ} lines for the
18777 @kindex show com1base
18778 @kindex show com1irq
18779 @kindex show com2base
18780 @kindex show com2irq
18781 @kindex show com3base
18782 @kindex show com3irq
18783 @kindex show com4base
18784 @kindex show com4irq
18785 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
18786 display the current settings of the base address and the @code{IRQ}
18787 lines used by the COM ports.
18790 @kindex info serial
18791 @cindex DOS serial port status
18792 This command prints the status of the 4 DOS serial ports. For each
18793 port, it prints whether it's active or not, its I/O base address and
18794 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
18795 counts of various errors encountered so far.
18799 @node Cygwin Native
18800 @subsection Features for Debugging MS Windows PE Executables
18801 @cindex MS Windows debugging
18802 @cindex native Cygwin debugging
18803 @cindex Cygwin-specific commands
18805 @value{GDBN} supports native debugging of MS Windows programs, including
18806 DLLs with and without symbolic debugging information.
18808 @cindex Ctrl-BREAK, MS-Windows
18809 @cindex interrupt debuggee on MS-Windows
18810 MS-Windows programs that call @code{SetConsoleMode} to switch off the
18811 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
18812 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
18813 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
18814 sequence, which can be used to interrupt the debuggee even if it
18817 There are various additional Cygwin-specific commands, described in
18818 this section. Working with DLLs that have no debugging symbols is
18819 described in @ref{Non-debug DLL Symbols}.
18824 This is a prefix of MS Windows-specific commands which print
18825 information about the target system and important OS structures.
18827 @item info w32 selector
18828 This command displays information returned by
18829 the Win32 API @code{GetThreadSelectorEntry} function.
18830 It takes an optional argument that is evaluated to
18831 a long value to give the information about this given selector.
18832 Without argument, this command displays information
18833 about the six segment registers.
18835 @item info w32 thread-information-block
18836 This command displays thread specific information stored in the
18837 Thread Information Block (readable on the X86 CPU family using @code{$fs}
18838 selector for 32-bit programs and @code{$gs} for 64-bit programs).
18842 This is a Cygwin-specific alias of @code{info shared}.
18844 @kindex dll-symbols
18846 This command loads symbols from a dll similarly to
18847 add-sym command but without the need to specify a base address.
18849 @kindex set cygwin-exceptions
18850 @cindex debugging the Cygwin DLL
18851 @cindex Cygwin DLL, debugging
18852 @item set cygwin-exceptions @var{mode}
18853 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
18854 happen inside the Cygwin DLL. If @var{mode} is @code{off},
18855 @value{GDBN} will delay recognition of exceptions, and may ignore some
18856 exceptions which seem to be caused by internal Cygwin DLL
18857 ``bookkeeping''. This option is meant primarily for debugging the
18858 Cygwin DLL itself; the default value is @code{off} to avoid annoying
18859 @value{GDBN} users with false @code{SIGSEGV} signals.
18861 @kindex show cygwin-exceptions
18862 @item show cygwin-exceptions
18863 Displays whether @value{GDBN} will break on exceptions that happen
18864 inside the Cygwin DLL itself.
18866 @kindex set new-console
18867 @item set new-console @var{mode}
18868 If @var{mode} is @code{on} the debuggee will
18869 be started in a new console on next start.
18870 If @var{mode} is @code{off}, the debuggee will
18871 be started in the same console as the debugger.
18873 @kindex show new-console
18874 @item show new-console
18875 Displays whether a new console is used
18876 when the debuggee is started.
18878 @kindex set new-group
18879 @item set new-group @var{mode}
18880 This boolean value controls whether the debuggee should
18881 start a new group or stay in the same group as the debugger.
18882 This affects the way the Windows OS handles
18885 @kindex show new-group
18886 @item show new-group
18887 Displays current value of new-group boolean.
18889 @kindex set debugevents
18890 @item set debugevents
18891 This boolean value adds debug output concerning kernel events related
18892 to the debuggee seen by the debugger. This includes events that
18893 signal thread and process creation and exit, DLL loading and
18894 unloading, console interrupts, and debugging messages produced by the
18895 Windows @code{OutputDebugString} API call.
18897 @kindex set debugexec
18898 @item set debugexec
18899 This boolean value adds debug output concerning execute events
18900 (such as resume thread) seen by the debugger.
18902 @kindex set debugexceptions
18903 @item set debugexceptions
18904 This boolean value adds debug output concerning exceptions in the
18905 debuggee seen by the debugger.
18907 @kindex set debugmemory
18908 @item set debugmemory
18909 This boolean value adds debug output concerning debuggee memory reads
18910 and writes by the debugger.
18914 This boolean values specifies whether the debuggee is called
18915 via a shell or directly (default value is on).
18919 Displays if the debuggee will be started with a shell.
18924 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
18927 @node Non-debug DLL Symbols
18928 @subsubsection Support for DLLs without Debugging Symbols
18929 @cindex DLLs with no debugging symbols
18930 @cindex Minimal symbols and DLLs
18932 Very often on windows, some of the DLLs that your program relies on do
18933 not include symbolic debugging information (for example,
18934 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
18935 symbols in a DLL, it relies on the minimal amount of symbolic
18936 information contained in the DLL's export table. This section
18937 describes working with such symbols, known internally to @value{GDBN} as
18938 ``minimal symbols''.
18940 Note that before the debugged program has started execution, no DLLs
18941 will have been loaded. The easiest way around this problem is simply to
18942 start the program --- either by setting a breakpoint or letting the
18943 program run once to completion. It is also possible to force
18944 @value{GDBN} to load a particular DLL before starting the executable ---
18945 see the shared library information in @ref{Files}, or the
18946 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
18947 explicitly loading symbols from a DLL with no debugging information will
18948 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
18949 which may adversely affect symbol lookup performance.
18951 @subsubsection DLL Name Prefixes
18953 In keeping with the naming conventions used by the Microsoft debugging
18954 tools, DLL export symbols are made available with a prefix based on the
18955 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
18956 also entered into the symbol table, so @code{CreateFileA} is often
18957 sufficient. In some cases there will be name clashes within a program
18958 (particularly if the executable itself includes full debugging symbols)
18959 necessitating the use of the fully qualified name when referring to the
18960 contents of the DLL. Use single-quotes around the name to avoid the
18961 exclamation mark (``!'') being interpreted as a language operator.
18963 Note that the internal name of the DLL may be all upper-case, even
18964 though the file name of the DLL is lower-case, or vice-versa. Since
18965 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
18966 some confusion. If in doubt, try the @code{info functions} and
18967 @code{info variables} commands or even @code{maint print msymbols}
18968 (@pxref{Symbols}). Here's an example:
18971 (@value{GDBP}) info function CreateFileA
18972 All functions matching regular expression "CreateFileA":
18974 Non-debugging symbols:
18975 0x77e885f4 CreateFileA
18976 0x77e885f4 KERNEL32!CreateFileA
18980 (@value{GDBP}) info function !
18981 All functions matching regular expression "!":
18983 Non-debugging symbols:
18984 0x6100114c cygwin1!__assert
18985 0x61004034 cygwin1!_dll_crt0@@0
18986 0x61004240 cygwin1!dll_crt0(per_process *)
18990 @subsubsection Working with Minimal Symbols
18992 Symbols extracted from a DLL's export table do not contain very much
18993 type information. All that @value{GDBN} can do is guess whether a symbol
18994 refers to a function or variable depending on the linker section that
18995 contains the symbol. Also note that the actual contents of the memory
18996 contained in a DLL are not available unless the program is running. This
18997 means that you cannot examine the contents of a variable or disassemble
18998 a function within a DLL without a running program.
19000 Variables are generally treated as pointers and dereferenced
19001 automatically. For this reason, it is often necessary to prefix a
19002 variable name with the address-of operator (``&'') and provide explicit
19003 type information in the command. Here's an example of the type of
19007 (@value{GDBP}) print 'cygwin1!__argv'
19012 (@value{GDBP}) x 'cygwin1!__argv'
19013 0x10021610: "\230y\""
19016 And two possible solutions:
19019 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
19020 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
19024 (@value{GDBP}) x/2x &'cygwin1!__argv'
19025 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
19026 (@value{GDBP}) x/x 0x10021608
19027 0x10021608: 0x0022fd98
19028 (@value{GDBP}) x/s 0x0022fd98
19029 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
19032 Setting a break point within a DLL is possible even before the program
19033 starts execution. However, under these circumstances, @value{GDBN} can't
19034 examine the initial instructions of the function in order to skip the
19035 function's frame set-up code. You can work around this by using ``*&''
19036 to set the breakpoint at a raw memory address:
19039 (@value{GDBP}) break *&'python22!PyOS_Readline'
19040 Breakpoint 1 at 0x1e04eff0
19043 The author of these extensions is not entirely convinced that setting a
19044 break point within a shared DLL like @file{kernel32.dll} is completely
19048 @subsection Commands Specific to @sc{gnu} Hurd Systems
19049 @cindex @sc{gnu} Hurd debugging
19051 This subsection describes @value{GDBN} commands specific to the
19052 @sc{gnu} Hurd native debugging.
19057 @kindex set signals@r{, Hurd command}
19058 @kindex set sigs@r{, Hurd command}
19059 This command toggles the state of inferior signal interception by
19060 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
19061 affected by this command. @code{sigs} is a shorthand alias for
19066 @kindex show signals@r{, Hurd command}
19067 @kindex show sigs@r{, Hurd command}
19068 Show the current state of intercepting inferior's signals.
19070 @item set signal-thread
19071 @itemx set sigthread
19072 @kindex set signal-thread
19073 @kindex set sigthread
19074 This command tells @value{GDBN} which thread is the @code{libc} signal
19075 thread. That thread is run when a signal is delivered to a running
19076 process. @code{set sigthread} is the shorthand alias of @code{set
19079 @item show signal-thread
19080 @itemx show sigthread
19081 @kindex show signal-thread
19082 @kindex show sigthread
19083 These two commands show which thread will run when the inferior is
19084 delivered a signal.
19087 @kindex set stopped@r{, Hurd command}
19088 This commands tells @value{GDBN} that the inferior process is stopped,
19089 as with the @code{SIGSTOP} signal. The stopped process can be
19090 continued by delivering a signal to it.
19093 @kindex show stopped@r{, Hurd command}
19094 This command shows whether @value{GDBN} thinks the debuggee is
19097 @item set exceptions
19098 @kindex set exceptions@r{, Hurd command}
19099 Use this command to turn off trapping of exceptions in the inferior.
19100 When exception trapping is off, neither breakpoints nor
19101 single-stepping will work. To restore the default, set exception
19104 @item show exceptions
19105 @kindex show exceptions@r{, Hurd command}
19106 Show the current state of trapping exceptions in the inferior.
19108 @item set task pause
19109 @kindex set task@r{, Hurd commands}
19110 @cindex task attributes (@sc{gnu} Hurd)
19111 @cindex pause current task (@sc{gnu} Hurd)
19112 This command toggles task suspension when @value{GDBN} has control.
19113 Setting it to on takes effect immediately, and the task is suspended
19114 whenever @value{GDBN} gets control. Setting it to off will take
19115 effect the next time the inferior is continued. If this option is set
19116 to off, you can use @code{set thread default pause on} or @code{set
19117 thread pause on} (see below) to pause individual threads.
19119 @item show task pause
19120 @kindex show task@r{, Hurd commands}
19121 Show the current state of task suspension.
19123 @item set task detach-suspend-count
19124 @cindex task suspend count
19125 @cindex detach from task, @sc{gnu} Hurd
19126 This command sets the suspend count the task will be left with when
19127 @value{GDBN} detaches from it.
19129 @item show task detach-suspend-count
19130 Show the suspend count the task will be left with when detaching.
19132 @item set task exception-port
19133 @itemx set task excp
19134 @cindex task exception port, @sc{gnu} Hurd
19135 This command sets the task exception port to which @value{GDBN} will
19136 forward exceptions. The argument should be the value of the @dfn{send
19137 rights} of the task. @code{set task excp} is a shorthand alias.
19139 @item set noninvasive
19140 @cindex noninvasive task options
19141 This command switches @value{GDBN} to a mode that is the least
19142 invasive as far as interfering with the inferior is concerned. This
19143 is the same as using @code{set task pause}, @code{set exceptions}, and
19144 @code{set signals} to values opposite to the defaults.
19146 @item info send-rights
19147 @itemx info receive-rights
19148 @itemx info port-rights
19149 @itemx info port-sets
19150 @itemx info dead-names
19153 @cindex send rights, @sc{gnu} Hurd
19154 @cindex receive rights, @sc{gnu} Hurd
19155 @cindex port rights, @sc{gnu} Hurd
19156 @cindex port sets, @sc{gnu} Hurd
19157 @cindex dead names, @sc{gnu} Hurd
19158 These commands display information about, respectively, send rights,
19159 receive rights, port rights, port sets, and dead names of a task.
19160 There are also shorthand aliases: @code{info ports} for @code{info
19161 port-rights} and @code{info psets} for @code{info port-sets}.
19163 @item set thread pause
19164 @kindex set thread@r{, Hurd command}
19165 @cindex thread properties, @sc{gnu} Hurd
19166 @cindex pause current thread (@sc{gnu} Hurd)
19167 This command toggles current thread suspension when @value{GDBN} has
19168 control. Setting it to on takes effect immediately, and the current
19169 thread is suspended whenever @value{GDBN} gets control. Setting it to
19170 off will take effect the next time the inferior is continued.
19171 Normally, this command has no effect, since when @value{GDBN} has
19172 control, the whole task is suspended. However, if you used @code{set
19173 task pause off} (see above), this command comes in handy to suspend
19174 only the current thread.
19176 @item show thread pause
19177 @kindex show thread@r{, Hurd command}
19178 This command shows the state of current thread suspension.
19180 @item set thread run
19181 This command sets whether the current thread is allowed to run.
19183 @item show thread run
19184 Show whether the current thread is allowed to run.
19186 @item set thread detach-suspend-count
19187 @cindex thread suspend count, @sc{gnu} Hurd
19188 @cindex detach from thread, @sc{gnu} Hurd
19189 This command sets the suspend count @value{GDBN} will leave on a
19190 thread when detaching. This number is relative to the suspend count
19191 found by @value{GDBN} when it notices the thread; use @code{set thread
19192 takeover-suspend-count} to force it to an absolute value.
19194 @item show thread detach-suspend-count
19195 Show the suspend count @value{GDBN} will leave on the thread when
19198 @item set thread exception-port
19199 @itemx set thread excp
19200 Set the thread exception port to which to forward exceptions. This
19201 overrides the port set by @code{set task exception-port} (see above).
19202 @code{set thread excp} is the shorthand alias.
19204 @item set thread takeover-suspend-count
19205 Normally, @value{GDBN}'s thread suspend counts are relative to the
19206 value @value{GDBN} finds when it notices each thread. This command
19207 changes the suspend counts to be absolute instead.
19209 @item set thread default
19210 @itemx show thread default
19211 @cindex thread default settings, @sc{gnu} Hurd
19212 Each of the above @code{set thread} commands has a @code{set thread
19213 default} counterpart (e.g., @code{set thread default pause}, @code{set
19214 thread default exception-port}, etc.). The @code{thread default}
19215 variety of commands sets the default thread properties for all
19216 threads; you can then change the properties of individual threads with
19217 the non-default commands.
19222 @subsection QNX Neutrino
19223 @cindex QNX Neutrino
19225 @value{GDBN} provides the following commands specific to the QNX
19229 @item set debug nto-debug
19230 @kindex set debug nto-debug
19231 When set to on, enables debugging messages specific to the QNX
19234 @item show debug nto-debug
19235 @kindex show debug nto-debug
19236 Show the current state of QNX Neutrino messages.
19243 @value{GDBN} provides the following commands specific to the Darwin target:
19246 @item set debug darwin @var{num}
19247 @kindex set debug darwin
19248 When set to a non zero value, enables debugging messages specific to
19249 the Darwin support. Higher values produce more verbose output.
19251 @item show debug darwin
19252 @kindex show debug darwin
19253 Show the current state of Darwin messages.
19255 @item set debug mach-o @var{num}
19256 @kindex set debug mach-o
19257 When set to a non zero value, enables debugging messages while
19258 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
19259 file format used on Darwin for object and executable files.) Higher
19260 values produce more verbose output. This is a command to diagnose
19261 problems internal to @value{GDBN} and should not be needed in normal
19264 @item show debug mach-o
19265 @kindex show debug mach-o
19266 Show the current state of Mach-O file messages.
19268 @item set mach-exceptions on
19269 @itemx set mach-exceptions off
19270 @kindex set mach-exceptions
19271 On Darwin, faults are first reported as a Mach exception and are then
19272 mapped to a Posix signal. Use this command to turn on trapping of
19273 Mach exceptions in the inferior. This might be sometimes useful to
19274 better understand the cause of a fault. The default is off.
19276 @item show mach-exceptions
19277 @kindex show mach-exceptions
19278 Show the current state of exceptions trapping.
19283 @section Embedded Operating Systems
19285 This section describes configurations involving the debugging of
19286 embedded operating systems that are available for several different
19290 * VxWorks:: Using @value{GDBN} with VxWorks
19293 @value{GDBN} includes the ability to debug programs running on
19294 various real-time operating systems.
19297 @subsection Using @value{GDBN} with VxWorks
19303 @kindex target vxworks
19304 @item target vxworks @var{machinename}
19305 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
19306 is the target system's machine name or IP address.
19310 On VxWorks, @code{load} links @var{filename} dynamically on the
19311 current target system as well as adding its symbols in @value{GDBN}.
19313 @value{GDBN} enables developers to spawn and debug tasks running on networked
19314 VxWorks targets from a Unix host. Already-running tasks spawned from
19315 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
19316 both the Unix host and on the VxWorks target. The program
19317 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
19318 installed with the name @code{vxgdb}, to distinguish it from a
19319 @value{GDBN} for debugging programs on the host itself.)
19322 @item VxWorks-timeout @var{args}
19323 @kindex vxworks-timeout
19324 All VxWorks-based targets now support the option @code{vxworks-timeout}.
19325 This option is set by the user, and @var{args} represents the number of
19326 seconds @value{GDBN} waits for responses to rpc's. You might use this if
19327 your VxWorks target is a slow software simulator or is on the far side
19328 of a thin network line.
19331 The following information on connecting to VxWorks was current when
19332 this manual was produced; newer releases of VxWorks may use revised
19335 @findex INCLUDE_RDB
19336 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
19337 to include the remote debugging interface routines in the VxWorks
19338 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
19339 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
19340 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
19341 source debugging task @code{tRdbTask} when VxWorks is booted. For more
19342 information on configuring and remaking VxWorks, see the manufacturer's
19344 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
19346 Once you have included @file{rdb.a} in your VxWorks system image and set
19347 your Unix execution search path to find @value{GDBN}, you are ready to
19348 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
19349 @code{vxgdb}, depending on your installation).
19351 @value{GDBN} comes up showing the prompt:
19358 * VxWorks Connection:: Connecting to VxWorks
19359 * VxWorks Download:: VxWorks download
19360 * VxWorks Attach:: Running tasks
19363 @node VxWorks Connection
19364 @subsubsection Connecting to VxWorks
19366 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
19367 network. To connect to a target whose host name is ``@code{tt}'', type:
19370 (vxgdb) target vxworks tt
19374 @value{GDBN} displays messages like these:
19377 Attaching remote machine across net...
19382 @value{GDBN} then attempts to read the symbol tables of any object modules
19383 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
19384 these files by searching the directories listed in the command search
19385 path (@pxref{Environment, ,Your Program's Environment}); if it fails
19386 to find an object file, it displays a message such as:
19389 prog.o: No such file or directory.
19392 When this happens, add the appropriate directory to the search path with
19393 the @value{GDBN} command @code{path}, and execute the @code{target}
19396 @node VxWorks Download
19397 @subsubsection VxWorks Download
19399 @cindex download to VxWorks
19400 If you have connected to the VxWorks target and you want to debug an
19401 object that has not yet been loaded, you can use the @value{GDBN}
19402 @code{load} command to download a file from Unix to VxWorks
19403 incrementally. The object file given as an argument to the @code{load}
19404 command is actually opened twice: first by the VxWorks target in order
19405 to download the code, then by @value{GDBN} in order to read the symbol
19406 table. This can lead to problems if the current working directories on
19407 the two systems differ. If both systems have NFS mounted the same
19408 filesystems, you can avoid these problems by using absolute paths.
19409 Otherwise, it is simplest to set the working directory on both systems
19410 to the directory in which the object file resides, and then to reference
19411 the file by its name, without any path. For instance, a program
19412 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
19413 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
19414 program, type this on VxWorks:
19417 -> cd "@var{vxpath}/vw/demo/rdb"
19421 Then, in @value{GDBN}, type:
19424 (vxgdb) cd @var{hostpath}/vw/demo/rdb
19425 (vxgdb) load prog.o
19428 @value{GDBN} displays a response similar to this:
19431 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
19434 You can also use the @code{load} command to reload an object module
19435 after editing and recompiling the corresponding source file. Note that
19436 this makes @value{GDBN} delete all currently-defined breakpoints,
19437 auto-displays, and convenience variables, and to clear the value
19438 history. (This is necessary in order to preserve the integrity of
19439 debugger's data structures that reference the target system's symbol
19442 @node VxWorks Attach
19443 @subsubsection Running Tasks
19445 @cindex running VxWorks tasks
19446 You can also attach to an existing task using the @code{attach} command as
19450 (vxgdb) attach @var{task}
19454 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
19455 or suspended when you attach to it. Running tasks are suspended at
19456 the time of attachment.
19458 @node Embedded Processors
19459 @section Embedded Processors
19461 This section goes into details specific to particular embedded
19464 @cindex send command to simulator
19465 Whenever a specific embedded processor has a simulator, @value{GDBN}
19466 allows to send an arbitrary command to the simulator.
19469 @item sim @var{command}
19470 @kindex sim@r{, a command}
19471 Send an arbitrary @var{command} string to the simulator. Consult the
19472 documentation for the specific simulator in use for information about
19473 acceptable commands.
19479 * M32R/D:: Renesas M32R/D
19480 * M68K:: Motorola M68K
19481 * MicroBlaze:: Xilinx MicroBlaze
19482 * MIPS Embedded:: MIPS Embedded
19483 * OpenRISC 1000:: OpenRisc 1000
19484 * PowerPC Embedded:: PowerPC Embedded
19485 * PA:: HP PA Embedded
19486 * Sparclet:: Tsqware Sparclet
19487 * Sparclite:: Fujitsu Sparclite
19488 * Z8000:: Zilog Z8000
19491 * Super-H:: Renesas Super-H
19500 @item target rdi @var{dev}
19501 ARM Angel monitor, via RDI library interface to ADP protocol. You may
19502 use this target to communicate with both boards running the Angel
19503 monitor, or with the EmbeddedICE JTAG debug device.
19506 @item target rdp @var{dev}
19511 @value{GDBN} provides the following ARM-specific commands:
19514 @item set arm disassembler
19516 This commands selects from a list of disassembly styles. The
19517 @code{"std"} style is the standard style.
19519 @item show arm disassembler
19521 Show the current disassembly style.
19523 @item set arm apcs32
19524 @cindex ARM 32-bit mode
19525 This command toggles ARM operation mode between 32-bit and 26-bit.
19527 @item show arm apcs32
19528 Display the current usage of the ARM 32-bit mode.
19530 @item set arm fpu @var{fputype}
19531 This command sets the ARM floating-point unit (FPU) type. The
19532 argument @var{fputype} can be one of these:
19536 Determine the FPU type by querying the OS ABI.
19538 Software FPU, with mixed-endian doubles on little-endian ARM
19541 GCC-compiled FPA co-processor.
19543 Software FPU with pure-endian doubles.
19549 Show the current type of the FPU.
19552 This command forces @value{GDBN} to use the specified ABI.
19555 Show the currently used ABI.
19557 @item set arm fallback-mode (arm|thumb|auto)
19558 @value{GDBN} uses the symbol table, when available, to determine
19559 whether instructions are ARM or Thumb. This command controls
19560 @value{GDBN}'s default behavior when the symbol table is not
19561 available. The default is @samp{auto}, which causes @value{GDBN} to
19562 use the current execution mode (from the @code{T} bit in the @code{CPSR}
19565 @item show arm fallback-mode
19566 Show the current fallback instruction mode.
19568 @item set arm force-mode (arm|thumb|auto)
19569 This command overrides use of the symbol table to determine whether
19570 instructions are ARM or Thumb. The default is @samp{auto}, which
19571 causes @value{GDBN} to use the symbol table and then the setting
19572 of @samp{set arm fallback-mode}.
19574 @item show arm force-mode
19575 Show the current forced instruction mode.
19577 @item set debug arm
19578 Toggle whether to display ARM-specific debugging messages from the ARM
19579 target support subsystem.
19581 @item show debug arm
19582 Show whether ARM-specific debugging messages are enabled.
19585 The following commands are available when an ARM target is debugged
19586 using the RDI interface:
19589 @item rdilogfile @r{[}@var{file}@r{]}
19591 @cindex ADP (Angel Debugger Protocol) logging
19592 Set the filename for the ADP (Angel Debugger Protocol) packet log.
19593 With an argument, sets the log file to the specified @var{file}. With
19594 no argument, show the current log file name. The default log file is
19597 @item rdilogenable @r{[}@var{arg}@r{]}
19598 @kindex rdilogenable
19599 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
19600 enables logging, with an argument 0 or @code{"no"} disables it. With
19601 no arguments displays the current setting. When logging is enabled,
19602 ADP packets exchanged between @value{GDBN} and the RDI target device
19603 are logged to a file.
19605 @item set rdiromatzero
19606 @kindex set rdiromatzero
19607 @cindex ROM at zero address, RDI
19608 Tell @value{GDBN} whether the target has ROM at address 0. If on,
19609 vector catching is disabled, so that zero address can be used. If off
19610 (the default), vector catching is enabled. For this command to take
19611 effect, it needs to be invoked prior to the @code{target rdi} command.
19613 @item show rdiromatzero
19614 @kindex show rdiromatzero
19615 Show the current setting of ROM at zero address.
19617 @item set rdiheartbeat
19618 @kindex set rdiheartbeat
19619 @cindex RDI heartbeat
19620 Enable or disable RDI heartbeat packets. It is not recommended to
19621 turn on this option, since it confuses ARM and EPI JTAG interface, as
19622 well as the Angel monitor.
19624 @item show rdiheartbeat
19625 @kindex show rdiheartbeat
19626 Show the setting of RDI heartbeat packets.
19630 @item target sim @r{[}@var{simargs}@r{]} @dots{}
19631 The @value{GDBN} ARM simulator accepts the following optional arguments.
19634 @item --swi-support=@var{type}
19635 Tell the simulator which SWI interfaces to support.
19636 @var{type} may be a comma separated list of the following values.
19637 The default value is @code{all}.
19650 @subsection Renesas M32R/D and M32R/SDI
19653 @kindex target m32r
19654 @item target m32r @var{dev}
19655 Renesas M32R/D ROM monitor.
19657 @kindex target m32rsdi
19658 @item target m32rsdi @var{dev}
19659 Renesas M32R SDI server, connected via parallel port to the board.
19662 The following @value{GDBN} commands are specific to the M32R monitor:
19665 @item set download-path @var{path}
19666 @kindex set download-path
19667 @cindex find downloadable @sc{srec} files (M32R)
19668 Set the default path for finding downloadable @sc{srec} files.
19670 @item show download-path
19671 @kindex show download-path
19672 Show the default path for downloadable @sc{srec} files.
19674 @item set board-address @var{addr}
19675 @kindex set board-address
19676 @cindex M32-EVA target board address
19677 Set the IP address for the M32R-EVA target board.
19679 @item show board-address
19680 @kindex show board-address
19681 Show the current IP address of the target board.
19683 @item set server-address @var{addr}
19684 @kindex set server-address
19685 @cindex download server address (M32R)
19686 Set the IP address for the download server, which is the @value{GDBN}'s
19689 @item show server-address
19690 @kindex show server-address
19691 Display the IP address of the download server.
19693 @item upload @r{[}@var{file}@r{]}
19694 @kindex upload@r{, M32R}
19695 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
19696 upload capability. If no @var{file} argument is given, the current
19697 executable file is uploaded.
19699 @item tload @r{[}@var{file}@r{]}
19700 @kindex tload@r{, M32R}
19701 Test the @code{upload} command.
19704 The following commands are available for M32R/SDI:
19709 @cindex reset SDI connection, M32R
19710 This command resets the SDI connection.
19714 This command shows the SDI connection status.
19717 @kindex debug_chaos
19718 @cindex M32R/Chaos debugging
19719 Instructs the remote that M32R/Chaos debugging is to be used.
19721 @item use_debug_dma
19722 @kindex use_debug_dma
19723 Instructs the remote to use the DEBUG_DMA method of accessing memory.
19726 @kindex use_mon_code
19727 Instructs the remote to use the MON_CODE method of accessing memory.
19730 @kindex use_ib_break
19731 Instructs the remote to set breakpoints by IB break.
19733 @item use_dbt_break
19734 @kindex use_dbt_break
19735 Instructs the remote to set breakpoints by DBT.
19741 The Motorola m68k configuration includes ColdFire support, and a
19742 target command for the following ROM monitor.
19746 @kindex target dbug
19747 @item target dbug @var{dev}
19748 dBUG ROM monitor for Motorola ColdFire.
19753 @subsection MicroBlaze
19754 @cindex Xilinx MicroBlaze
19755 @cindex XMD, Xilinx Microprocessor Debugger
19757 The MicroBlaze is a soft-core processor supported on various Xilinx
19758 FPGAs, such as Spartan or Virtex series. Boards with these processors
19759 usually have JTAG ports which connect to a host system running the Xilinx
19760 Embedded Development Kit (EDK) or Software Development Kit (SDK).
19761 This host system is used to download the configuration bitstream to
19762 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
19763 communicates with the target board using the JTAG interface and
19764 presents a @code{gdbserver} interface to the board. By default
19765 @code{xmd} uses port @code{1234}. (While it is possible to change
19766 this default port, it requires the use of undocumented @code{xmd}
19767 commands. Contact Xilinx support if you need to do this.)
19769 Use these GDB commands to connect to the MicroBlaze target processor.
19772 @item target remote :1234
19773 Use this command to connect to the target if you are running @value{GDBN}
19774 on the same system as @code{xmd}.
19776 @item target remote @var{xmd-host}:1234
19777 Use this command to connect to the target if it is connected to @code{xmd}
19778 running on a different system named @var{xmd-host}.
19781 Use this command to download a program to the MicroBlaze target.
19783 @item set debug microblaze @var{n}
19784 Enable MicroBlaze-specific debugging messages if non-zero.
19786 @item show debug microblaze @var{n}
19787 Show MicroBlaze-specific debugging level.
19790 @node MIPS Embedded
19791 @subsection @acronym{MIPS} Embedded
19793 @cindex @acronym{MIPS} boards
19794 @value{GDBN} can use the @acronym{MIPS} remote debugging protocol to talk to a
19795 @acronym{MIPS} board attached to a serial line. This is available when
19796 you configure @value{GDBN} with @samp{--target=mips-elf}.
19799 Use these @value{GDBN} commands to specify the connection to your target board:
19802 @item target mips @var{port}
19803 @kindex target mips @var{port}
19804 To run a program on the board, start up @code{@value{GDBP}} with the
19805 name of your program as the argument. To connect to the board, use the
19806 command @samp{target mips @var{port}}, where @var{port} is the name of
19807 the serial port connected to the board. If the program has not already
19808 been downloaded to the board, you may use the @code{load} command to
19809 download it. You can then use all the usual @value{GDBN} commands.
19811 For example, this sequence connects to the target board through a serial
19812 port, and loads and runs a program called @var{prog} through the
19816 host$ @value{GDBP} @var{prog}
19817 @value{GDBN} is free software and @dots{}
19818 (@value{GDBP}) target mips /dev/ttyb
19819 (@value{GDBP}) load @var{prog}
19823 @item target mips @var{hostname}:@var{portnumber}
19824 On some @value{GDBN} host configurations, you can specify a TCP
19825 connection (for instance, to a serial line managed by a terminal
19826 concentrator) instead of a serial port, using the syntax
19827 @samp{@var{hostname}:@var{portnumber}}.
19829 @item target pmon @var{port}
19830 @kindex target pmon @var{port}
19833 @item target ddb @var{port}
19834 @kindex target ddb @var{port}
19835 NEC's DDB variant of PMON for Vr4300.
19837 @item target lsi @var{port}
19838 @kindex target lsi @var{port}
19839 LSI variant of PMON.
19841 @kindex target r3900
19842 @item target r3900 @var{dev}
19843 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
19845 @kindex target array
19846 @item target array @var{dev}
19847 Array Tech LSI33K RAID controller board.
19853 @value{GDBN} also supports these special commands for @acronym{MIPS} targets:
19856 @item set mipsfpu double
19857 @itemx set mipsfpu single
19858 @itemx set mipsfpu none
19859 @itemx set mipsfpu auto
19860 @itemx show mipsfpu
19861 @kindex set mipsfpu
19862 @kindex show mipsfpu
19863 @cindex @acronym{MIPS} remote floating point
19864 @cindex floating point, @acronym{MIPS} remote
19865 If your target board does not support the @acronym{MIPS} floating point
19866 coprocessor, you should use the command @samp{set mipsfpu none} (if you
19867 need this, you may wish to put the command in your @value{GDBN} init
19868 file). This tells @value{GDBN} how to find the return value of
19869 functions which return floating point values. It also allows
19870 @value{GDBN} to avoid saving the floating point registers when calling
19871 functions on the board. If you are using a floating point coprocessor
19872 with only single precision floating point support, as on the @sc{r4650}
19873 processor, use the command @samp{set mipsfpu single}. The default
19874 double precision floating point coprocessor may be selected using
19875 @samp{set mipsfpu double}.
19877 In previous versions the only choices were double precision or no
19878 floating point, so @samp{set mipsfpu on} will select double precision
19879 and @samp{set mipsfpu off} will select no floating point.
19881 As usual, you can inquire about the @code{mipsfpu} variable with
19882 @samp{show mipsfpu}.
19884 @item set timeout @var{seconds}
19885 @itemx set retransmit-timeout @var{seconds}
19886 @itemx show timeout
19887 @itemx show retransmit-timeout
19888 @cindex @code{timeout}, @acronym{MIPS} protocol
19889 @cindex @code{retransmit-timeout}, @acronym{MIPS} protocol
19890 @kindex set timeout
19891 @kindex show timeout
19892 @kindex set retransmit-timeout
19893 @kindex show retransmit-timeout
19894 You can control the timeout used while waiting for a packet, in the @acronym{MIPS}
19895 remote protocol, with the @code{set timeout @var{seconds}} command. The
19896 default is 5 seconds. Similarly, you can control the timeout used while
19897 waiting for an acknowledgment of a packet with the @code{set
19898 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
19899 You can inspect both values with @code{show timeout} and @code{show
19900 retransmit-timeout}. (These commands are @emph{only} available when
19901 @value{GDBN} is configured for @samp{--target=mips-elf}.)
19903 The timeout set by @code{set timeout} does not apply when @value{GDBN}
19904 is waiting for your program to stop. In that case, @value{GDBN} waits
19905 forever because it has no way of knowing how long the program is going
19906 to run before stopping.
19908 @item set syn-garbage-limit @var{num}
19909 @kindex set syn-garbage-limit@r{, @acronym{MIPS} remote}
19910 @cindex synchronize with remote @acronym{MIPS} target
19911 Limit the maximum number of characters @value{GDBN} should ignore when
19912 it tries to synchronize with the remote target. The default is 10
19913 characters. Setting the limit to -1 means there's no limit.
19915 @item show syn-garbage-limit
19916 @kindex show syn-garbage-limit@r{, @acronym{MIPS} remote}
19917 Show the current limit on the number of characters to ignore when
19918 trying to synchronize with the remote system.
19920 @item set monitor-prompt @var{prompt}
19921 @kindex set monitor-prompt@r{, @acronym{MIPS} remote}
19922 @cindex remote monitor prompt
19923 Tell @value{GDBN} to expect the specified @var{prompt} string from the
19924 remote monitor. The default depends on the target:
19934 @item show monitor-prompt
19935 @kindex show monitor-prompt@r{, @acronym{MIPS} remote}
19936 Show the current strings @value{GDBN} expects as the prompt from the
19939 @item set monitor-warnings
19940 @kindex set monitor-warnings@r{, @acronym{MIPS} remote}
19941 Enable or disable monitor warnings about hardware breakpoints. This
19942 has effect only for the @code{lsi} target. When on, @value{GDBN} will
19943 display warning messages whose codes are returned by the @code{lsi}
19944 PMON monitor for breakpoint commands.
19946 @item show monitor-warnings
19947 @kindex show monitor-warnings@r{, @acronym{MIPS} remote}
19948 Show the current setting of printing monitor warnings.
19950 @item pmon @var{command}
19951 @kindex pmon@r{, @acronym{MIPS} remote}
19952 @cindex send PMON command
19953 This command allows sending an arbitrary @var{command} string to the
19954 monitor. The monitor must be in debug mode for this to work.
19957 @node OpenRISC 1000
19958 @subsection OpenRISC 1000
19959 @cindex OpenRISC 1000
19961 @cindex or1k boards
19962 See OR1k Architecture document (@uref{www.opencores.org}) for more information
19963 about platform and commands.
19967 @kindex target jtag
19968 @item target jtag jtag://@var{host}:@var{port}
19970 Connects to remote JTAG server.
19971 JTAG remote server can be either an or1ksim or JTAG server,
19972 connected via parallel port to the board.
19974 Example: @code{target jtag jtag://localhost:9999}
19977 @item or1ksim @var{command}
19978 If connected to @code{or1ksim} OpenRISC 1000 Architectural
19979 Simulator, proprietary commands can be executed.
19981 @kindex info or1k spr
19982 @item info or1k spr
19983 Displays spr groups.
19985 @item info or1k spr @var{group}
19986 @itemx info or1k spr @var{groupno}
19987 Displays register names in selected group.
19989 @item info or1k spr @var{group} @var{register}
19990 @itemx info or1k spr @var{register}
19991 @itemx info or1k spr @var{groupno} @var{registerno}
19992 @itemx info or1k spr @var{registerno}
19993 Shows information about specified spr register.
19996 @item spr @var{group} @var{register} @var{value}
19997 @itemx spr @var{register @var{value}}
19998 @itemx spr @var{groupno} @var{registerno @var{value}}
19999 @itemx spr @var{registerno @var{value}}
20000 Writes @var{value} to specified spr register.
20003 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
20004 It is very similar to @value{GDBN} trace, except it does not interfere with normal
20005 program execution and is thus much faster. Hardware breakpoints/watchpoint
20006 triggers can be set using:
20009 Load effective address/data
20011 Store effective address/data
20013 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
20018 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
20019 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
20021 @code{htrace} commands:
20022 @cindex OpenRISC 1000 htrace
20025 @item hwatch @var{conditional}
20026 Set hardware watchpoint on combination of Load/Store Effective Address(es)
20027 or Data. For example:
20029 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
20031 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
20035 Display information about current HW trace configuration.
20037 @item htrace trigger @var{conditional}
20038 Set starting criteria for HW trace.
20040 @item htrace qualifier @var{conditional}
20041 Set acquisition qualifier for HW trace.
20043 @item htrace stop @var{conditional}
20044 Set HW trace stopping criteria.
20046 @item htrace record [@var{data}]*
20047 Selects the data to be recorded, when qualifier is met and HW trace was
20050 @item htrace enable
20051 @itemx htrace disable
20052 Enables/disables the HW trace.
20054 @item htrace rewind [@var{filename}]
20055 Clears currently recorded trace data.
20057 If filename is specified, new trace file is made and any newly collected data
20058 will be written there.
20060 @item htrace print [@var{start} [@var{len}]]
20061 Prints trace buffer, using current record configuration.
20063 @item htrace mode continuous
20064 Set continuous trace mode.
20066 @item htrace mode suspend
20067 Set suspend trace mode.
20071 @node PowerPC Embedded
20072 @subsection PowerPC Embedded
20074 @cindex DVC register
20075 @value{GDBN} supports using the DVC (Data Value Compare) register to
20076 implement in hardware simple hardware watchpoint conditions of the form:
20079 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
20080 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
20083 The DVC register will be automatically used when @value{GDBN} detects
20084 such pattern in a condition expression, and the created watchpoint uses one
20085 debug register (either the @code{exact-watchpoints} option is on and the
20086 variable is scalar, or the variable has a length of one byte). This feature
20087 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
20090 When running on PowerPC embedded processors, @value{GDBN} automatically uses
20091 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
20092 in which case watchpoints using only one debug register are created when
20093 watching variables of scalar types.
20095 You can create an artificial array to watch an arbitrary memory
20096 region using one of the following commands (@pxref{Expressions}):
20099 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
20100 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
20103 PowerPC embedded processors support masked watchpoints. See the discussion
20104 about the @code{mask} argument in @ref{Set Watchpoints}.
20106 @cindex ranged breakpoint
20107 PowerPC embedded processors support hardware accelerated
20108 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
20109 the inferior whenever it executes an instruction at any address within
20110 the range it specifies. To set a ranged breakpoint in @value{GDBN},
20111 use the @code{break-range} command.
20113 @value{GDBN} provides the following PowerPC-specific commands:
20116 @kindex break-range
20117 @item break-range @var{start-location}, @var{end-location}
20118 Set a breakpoint for an address range.
20119 @var{start-location} and @var{end-location} can specify a function name,
20120 a line number, an offset of lines from the current line or from the start
20121 location, or an address of an instruction (see @ref{Specify Location},
20122 for a list of all the possible ways to specify a @var{location}.)
20123 The breakpoint will stop execution of the inferior whenever it
20124 executes an instruction at any address within the specified range,
20125 (including @var{start-location} and @var{end-location}.)
20127 @kindex set powerpc
20128 @item set powerpc soft-float
20129 @itemx show powerpc soft-float
20130 Force @value{GDBN} to use (or not use) a software floating point calling
20131 convention. By default, @value{GDBN} selects the calling convention based
20132 on the selected architecture and the provided executable file.
20134 @item set powerpc vector-abi
20135 @itemx show powerpc vector-abi
20136 Force @value{GDBN} to use the specified calling convention for vector
20137 arguments and return values. The valid options are @samp{auto};
20138 @samp{generic}, to avoid vector registers even if they are present;
20139 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
20140 registers. By default, @value{GDBN} selects the calling convention
20141 based on the selected architecture and the provided executable file.
20143 @item set powerpc exact-watchpoints
20144 @itemx show powerpc exact-watchpoints
20145 Allow @value{GDBN} to use only one debug register when watching a variable
20146 of scalar type, thus assuming that the variable is accessed through the
20147 address of its first byte.
20149 @kindex target dink32
20150 @item target dink32 @var{dev}
20151 DINK32 ROM monitor.
20153 @kindex target ppcbug
20154 @item target ppcbug @var{dev}
20155 @kindex target ppcbug1
20156 @item target ppcbug1 @var{dev}
20157 PPCBUG ROM monitor for PowerPC.
20160 @item target sds @var{dev}
20161 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
20164 @cindex SDS protocol
20165 The following commands specific to the SDS protocol are supported
20169 @item set sdstimeout @var{nsec}
20170 @kindex set sdstimeout
20171 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
20172 default is 2 seconds.
20174 @item show sdstimeout
20175 @kindex show sdstimeout
20176 Show the current value of the SDS timeout.
20178 @item sds @var{command}
20179 @kindex sds@r{, a command}
20180 Send the specified @var{command} string to the SDS monitor.
20185 @subsection HP PA Embedded
20189 @kindex target op50n
20190 @item target op50n @var{dev}
20191 OP50N monitor, running on an OKI HPPA board.
20193 @kindex target w89k
20194 @item target w89k @var{dev}
20195 W89K monitor, running on a Winbond HPPA board.
20200 @subsection Tsqware Sparclet
20204 @value{GDBN} enables developers to debug tasks running on
20205 Sparclet targets from a Unix host.
20206 @value{GDBN} uses code that runs on
20207 both the Unix host and on the Sparclet target. The program
20208 @code{@value{GDBP}} is installed and executed on the Unix host.
20211 @item remotetimeout @var{args}
20212 @kindex remotetimeout
20213 @value{GDBN} supports the option @code{remotetimeout}.
20214 This option is set by the user, and @var{args} represents the number of
20215 seconds @value{GDBN} waits for responses.
20218 @cindex compiling, on Sparclet
20219 When compiling for debugging, include the options @samp{-g} to get debug
20220 information and @samp{-Ttext} to relocate the program to where you wish to
20221 load it on the target. You may also want to add the options @samp{-n} or
20222 @samp{-N} in order to reduce the size of the sections. Example:
20225 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
20228 You can use @code{objdump} to verify that the addresses are what you intended:
20231 sparclet-aout-objdump --headers --syms prog
20234 @cindex running, on Sparclet
20236 your Unix execution search path to find @value{GDBN}, you are ready to
20237 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
20238 (or @code{sparclet-aout-gdb}, depending on your installation).
20240 @value{GDBN} comes up showing the prompt:
20247 * Sparclet File:: Setting the file to debug
20248 * Sparclet Connection:: Connecting to Sparclet
20249 * Sparclet Download:: Sparclet download
20250 * Sparclet Execution:: Running and debugging
20253 @node Sparclet File
20254 @subsubsection Setting File to Debug
20256 The @value{GDBN} command @code{file} lets you choose with program to debug.
20259 (gdbslet) file prog
20263 @value{GDBN} then attempts to read the symbol table of @file{prog}.
20264 @value{GDBN} locates
20265 the file by searching the directories listed in the command search
20267 If the file was compiled with debug information (option @samp{-g}), source
20268 files will be searched as well.
20269 @value{GDBN} locates
20270 the source files by searching the directories listed in the directory search
20271 path (@pxref{Environment, ,Your Program's Environment}).
20273 to find a file, it displays a message such as:
20276 prog: No such file or directory.
20279 When this happens, add the appropriate directories to the search paths with
20280 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
20281 @code{target} command again.
20283 @node Sparclet Connection
20284 @subsubsection Connecting to Sparclet
20286 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
20287 To connect to a target on serial port ``@code{ttya}'', type:
20290 (gdbslet) target sparclet /dev/ttya
20291 Remote target sparclet connected to /dev/ttya
20292 main () at ../prog.c:3
20296 @value{GDBN} displays messages like these:
20302 @node Sparclet Download
20303 @subsubsection Sparclet Download
20305 @cindex download to Sparclet
20306 Once connected to the Sparclet target,
20307 you can use the @value{GDBN}
20308 @code{load} command to download the file from the host to the target.
20309 The file name and load offset should be given as arguments to the @code{load}
20311 Since the file format is aout, the program must be loaded to the starting
20312 address. You can use @code{objdump} to find out what this value is. The load
20313 offset is an offset which is added to the VMA (virtual memory address)
20314 of each of the file's sections.
20315 For instance, if the program
20316 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
20317 and bss at 0x12010170, in @value{GDBN}, type:
20320 (gdbslet) load prog 0x12010000
20321 Loading section .text, size 0xdb0 vma 0x12010000
20324 If the code is loaded at a different address then what the program was linked
20325 to, you may need to use the @code{section} and @code{add-symbol-file} commands
20326 to tell @value{GDBN} where to map the symbol table.
20328 @node Sparclet Execution
20329 @subsubsection Running and Debugging
20331 @cindex running and debugging Sparclet programs
20332 You can now begin debugging the task using @value{GDBN}'s execution control
20333 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
20334 manual for the list of commands.
20338 Breakpoint 1 at 0x12010000: file prog.c, line 3.
20340 Starting program: prog
20341 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
20342 3 char *symarg = 0;
20344 4 char *execarg = "hello!";
20349 @subsection Fujitsu Sparclite
20353 @kindex target sparclite
20354 @item target sparclite @var{dev}
20355 Fujitsu sparclite boards, used only for the purpose of loading.
20356 You must use an additional command to debug the program.
20357 For example: target remote @var{dev} using @value{GDBN} standard
20363 @subsection Zilog Z8000
20366 @cindex simulator, Z8000
20367 @cindex Zilog Z8000 simulator
20369 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
20372 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
20373 unsegmented variant of the Z8000 architecture) or the Z8001 (the
20374 segmented variant). The simulator recognizes which architecture is
20375 appropriate by inspecting the object code.
20378 @item target sim @var{args}
20380 @kindex target sim@r{, with Z8000}
20381 Debug programs on a simulated CPU. If the simulator supports setup
20382 options, specify them via @var{args}.
20386 After specifying this target, you can debug programs for the simulated
20387 CPU in the same style as programs for your host computer; use the
20388 @code{file} command to load a new program image, the @code{run} command
20389 to run your program, and so on.
20391 As well as making available all the usual machine registers
20392 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
20393 additional items of information as specially named registers:
20398 Counts clock-ticks in the simulator.
20401 Counts instructions run in the simulator.
20404 Execution time in 60ths of a second.
20408 You can refer to these values in @value{GDBN} expressions with the usual
20409 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
20410 conditional breakpoint that suspends only after at least 5000
20411 simulated clock ticks.
20414 @subsection Atmel AVR
20417 When configured for debugging the Atmel AVR, @value{GDBN} supports the
20418 following AVR-specific commands:
20421 @item info io_registers
20422 @kindex info io_registers@r{, AVR}
20423 @cindex I/O registers (Atmel AVR)
20424 This command displays information about the AVR I/O registers. For
20425 each register, @value{GDBN} prints its number and value.
20432 When configured for debugging CRIS, @value{GDBN} provides the
20433 following CRIS-specific commands:
20436 @item set cris-version @var{ver}
20437 @cindex CRIS version
20438 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
20439 The CRIS version affects register names and sizes. This command is useful in
20440 case autodetection of the CRIS version fails.
20442 @item show cris-version
20443 Show the current CRIS version.
20445 @item set cris-dwarf2-cfi
20446 @cindex DWARF-2 CFI and CRIS
20447 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
20448 Change to @samp{off} when using @code{gcc-cris} whose version is below
20451 @item show cris-dwarf2-cfi
20452 Show the current state of using DWARF-2 CFI.
20454 @item set cris-mode @var{mode}
20456 Set the current CRIS mode to @var{mode}. It should only be changed when
20457 debugging in guru mode, in which case it should be set to
20458 @samp{guru} (the default is @samp{normal}).
20460 @item show cris-mode
20461 Show the current CRIS mode.
20465 @subsection Renesas Super-H
20468 For the Renesas Super-H processor, @value{GDBN} provides these
20473 @kindex regs@r{, Super-H}
20474 This command is deprecated, and @code{info all-registers} should be
20477 Show the values of all Super-H registers.
20479 @item set sh calling-convention @var{convention}
20480 @kindex set sh calling-convention
20481 Set the calling-convention used when calling functions from @value{GDBN}.
20482 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
20483 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
20484 convention. If the DWARF-2 information of the called function specifies
20485 that the function follows the Renesas calling convention, the function
20486 is called using the Renesas calling convention. If the calling convention
20487 is set to @samp{renesas}, the Renesas calling convention is always used,
20488 regardless of the DWARF-2 information. This can be used to override the
20489 default of @samp{gcc} if debug information is missing, or the compiler
20490 does not emit the DWARF-2 calling convention entry for a function.
20492 @item show sh calling-convention
20493 @kindex show sh calling-convention
20494 Show the current calling convention setting.
20499 @node Architectures
20500 @section Architectures
20502 This section describes characteristics of architectures that affect
20503 all uses of @value{GDBN} with the architecture, both native and cross.
20509 * HPPA:: HP PA architecture
20510 * SPU:: Cell Broadband Engine SPU architecture
20515 @subsection x86 Architecture-specific Issues
20518 @item set struct-convention @var{mode}
20519 @kindex set struct-convention
20520 @cindex struct return convention
20521 @cindex struct/union returned in registers
20522 Set the convention used by the inferior to return @code{struct}s and
20523 @code{union}s from functions to @var{mode}. Possible values of
20524 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
20525 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
20526 are returned on the stack, while @code{"reg"} means that a
20527 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
20528 be returned in a register.
20530 @item show struct-convention
20531 @kindex show struct-convention
20532 Show the current setting of the convention to return @code{struct}s
20539 See the following section.
20542 @subsection @acronym{MIPS}
20544 @cindex stack on Alpha
20545 @cindex stack on @acronym{MIPS}
20546 @cindex Alpha stack
20547 @cindex @acronym{MIPS} stack
20548 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
20549 sometimes requires @value{GDBN} to search backward in the object code to
20550 find the beginning of a function.
20552 @cindex response time, @acronym{MIPS} debugging
20553 To improve response time (especially for embedded applications, where
20554 @value{GDBN} may be restricted to a slow serial line for this search)
20555 you may want to limit the size of this search, using one of these
20559 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
20560 @item set heuristic-fence-post @var{limit}
20561 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
20562 search for the beginning of a function. A value of @var{0} (the
20563 default) means there is no limit. However, except for @var{0}, the
20564 larger the limit the more bytes @code{heuristic-fence-post} must search
20565 and therefore the longer it takes to run. You should only need to use
20566 this command when debugging a stripped executable.
20568 @item show heuristic-fence-post
20569 Display the current limit.
20573 These commands are available @emph{only} when @value{GDBN} is configured
20574 for debugging programs on Alpha or @acronym{MIPS} processors.
20576 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
20580 @item set mips abi @var{arg}
20581 @kindex set mips abi
20582 @cindex set ABI for @acronym{MIPS}
20583 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
20584 values of @var{arg} are:
20588 The default ABI associated with the current binary (this is the
20598 @item show mips abi
20599 @kindex show mips abi
20600 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
20602 @item set mips compression @var{arg}
20603 @kindex set mips compression
20604 @cindex code compression, @acronym{MIPS}
20605 Tell @value{GDBN} which @acronym{MIPS} compressed
20606 @acronym{ISA, Instruction Set Architecture} encoding is used by the
20607 inferior. @value{GDBN} uses this for code disassembly and other
20608 internal interpretation purposes. This setting is only referred to
20609 when no executable has been associated with the debugging session or
20610 the executable does not provide information about the encoding it uses.
20611 Otherwise this setting is automatically updated from information
20612 provided by the executable.
20614 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
20615 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
20616 executables containing @acronym{MIPS16} code frequently are not
20617 identified as such.
20619 This setting is ``sticky''; that is, it retains its value across
20620 debugging sessions until reset either explicitly with this command or
20621 implicitly from an executable.
20623 The compiler and/or assembler typically add symbol table annotations to
20624 identify functions compiled for the @acronym{MIPS16} or
20625 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
20626 are present, @value{GDBN} uses them in preference to the global
20627 compressed @acronym{ISA} encoding setting.
20629 @item show mips compression
20630 @kindex show mips compression
20631 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
20632 @value{GDBN} to debug the inferior.
20635 @itemx show mipsfpu
20636 @xref{MIPS Embedded, set mipsfpu}.
20638 @item set mips mask-address @var{arg}
20639 @kindex set mips mask-address
20640 @cindex @acronym{MIPS} addresses, masking
20641 This command determines whether the most-significant 32 bits of 64-bit
20642 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
20643 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
20644 setting, which lets @value{GDBN} determine the correct value.
20646 @item show mips mask-address
20647 @kindex show mips mask-address
20648 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
20651 @item set remote-mips64-transfers-32bit-regs
20652 @kindex set remote-mips64-transfers-32bit-regs
20653 This command controls compatibility with 64-bit @acronym{MIPS} targets that
20654 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
20655 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
20656 and 64 bits for other registers, set this option to @samp{on}.
20658 @item show remote-mips64-transfers-32bit-regs
20659 @kindex show remote-mips64-transfers-32bit-regs
20660 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
20662 @item set debug mips
20663 @kindex set debug mips
20664 This command turns on and off debugging messages for the @acronym{MIPS}-specific
20665 target code in @value{GDBN}.
20667 @item show debug mips
20668 @kindex show debug mips
20669 Show the current setting of @acronym{MIPS} debugging messages.
20675 @cindex HPPA support
20677 When @value{GDBN} is debugging the HP PA architecture, it provides the
20678 following special commands:
20681 @item set debug hppa
20682 @kindex set debug hppa
20683 This command determines whether HPPA architecture-specific debugging
20684 messages are to be displayed.
20686 @item show debug hppa
20687 Show whether HPPA debugging messages are displayed.
20689 @item maint print unwind @var{address}
20690 @kindex maint print unwind@r{, HPPA}
20691 This command displays the contents of the unwind table entry at the
20692 given @var{address}.
20698 @subsection Cell Broadband Engine SPU architecture
20699 @cindex Cell Broadband Engine
20702 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
20703 it provides the following special commands:
20706 @item info spu event
20708 Display SPU event facility status. Shows current event mask
20709 and pending event status.
20711 @item info spu signal
20712 Display SPU signal notification facility status. Shows pending
20713 signal-control word and signal notification mode of both signal
20714 notification channels.
20716 @item info spu mailbox
20717 Display SPU mailbox facility status. Shows all pending entries,
20718 in order of processing, in each of the SPU Write Outbound,
20719 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
20722 Display MFC DMA status. Shows all pending commands in the MFC
20723 DMA queue. For each entry, opcode, tag, class IDs, effective
20724 and local store addresses and transfer size are shown.
20726 @item info spu proxydma
20727 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
20728 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
20729 and local store addresses and transfer size are shown.
20733 When @value{GDBN} is debugging a combined PowerPC/SPU application
20734 on the Cell Broadband Engine, it provides in addition the following
20738 @item set spu stop-on-load @var{arg}
20740 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
20741 will give control to the user when a new SPE thread enters its @code{main}
20742 function. The default is @code{off}.
20744 @item show spu stop-on-load
20746 Show whether to stop for new SPE threads.
20748 @item set spu auto-flush-cache @var{arg}
20749 Set whether to automatically flush the software-managed cache. When set to
20750 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
20751 cache to be flushed whenever SPE execution stops. This provides a consistent
20752 view of PowerPC memory that is accessed via the cache. If an application
20753 does not use the software-managed cache, this option has no effect.
20755 @item show spu auto-flush-cache
20756 Show whether to automatically flush the software-managed cache.
20761 @subsection PowerPC
20762 @cindex PowerPC architecture
20764 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
20765 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
20766 numbers stored in the floating point registers. These values must be stored
20767 in two consecutive registers, always starting at an even register like
20768 @code{f0} or @code{f2}.
20770 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
20771 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
20772 @code{f2} and @code{f3} for @code{$dl1} and so on.
20774 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
20775 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
20778 @node Controlling GDB
20779 @chapter Controlling @value{GDBN}
20781 You can alter the way @value{GDBN} interacts with you by using the
20782 @code{set} command. For commands controlling how @value{GDBN} displays
20783 data, see @ref{Print Settings, ,Print Settings}. Other settings are
20788 * Editing:: Command editing
20789 * Command History:: Command history
20790 * Screen Size:: Screen size
20791 * Numbers:: Numbers
20792 * ABI:: Configuring the current ABI
20793 * Auto-loading:: Automatically loading associated files
20794 * Messages/Warnings:: Optional warnings and messages
20795 * Debugging Output:: Optional messages about internal happenings
20796 * Other Misc Settings:: Other Miscellaneous Settings
20804 @value{GDBN} indicates its readiness to read a command by printing a string
20805 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
20806 can change the prompt string with the @code{set prompt} command. For
20807 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
20808 the prompt in one of the @value{GDBN} sessions so that you can always tell
20809 which one you are talking to.
20811 @emph{Note:} @code{set prompt} does not add a space for you after the
20812 prompt you set. This allows you to set a prompt which ends in a space
20813 or a prompt that does not.
20817 @item set prompt @var{newprompt}
20818 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
20820 @kindex show prompt
20822 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
20825 Versions of @value{GDBN} that ship with Python scripting enabled have
20826 prompt extensions. The commands for interacting with these extensions
20830 @kindex set extended-prompt
20831 @item set extended-prompt @var{prompt}
20832 Set an extended prompt that allows for substitutions.
20833 @xref{gdb.prompt}, for a list of escape sequences that can be used for
20834 substitution. Any escape sequences specified as part of the prompt
20835 string are replaced with the corresponding strings each time the prompt
20841 set extended-prompt Current working directory: \w (gdb)
20844 Note that when an extended-prompt is set, it takes control of the
20845 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
20847 @kindex show extended-prompt
20848 @item show extended-prompt
20849 Prints the extended prompt. Any escape sequences specified as part of
20850 the prompt string with @code{set extended-prompt}, are replaced with the
20851 corresponding strings each time the prompt is displayed.
20855 @section Command Editing
20857 @cindex command line editing
20859 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
20860 @sc{gnu} library provides consistent behavior for programs which provide a
20861 command line interface to the user. Advantages are @sc{gnu} Emacs-style
20862 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
20863 substitution, and a storage and recall of command history across
20864 debugging sessions.
20866 You may control the behavior of command line editing in @value{GDBN} with the
20867 command @code{set}.
20870 @kindex set editing
20873 @itemx set editing on
20874 Enable command line editing (enabled by default).
20876 @item set editing off
20877 Disable command line editing.
20879 @kindex show editing
20881 Show whether command line editing is enabled.
20884 @ifset SYSTEM_READLINE
20885 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
20887 @ifclear SYSTEM_READLINE
20888 @xref{Command Line Editing},
20890 for more details about the Readline
20891 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
20892 encouraged to read that chapter.
20894 @node Command History
20895 @section Command History
20896 @cindex command history
20898 @value{GDBN} can keep track of the commands you type during your
20899 debugging sessions, so that you can be certain of precisely what
20900 happened. Use these commands to manage the @value{GDBN} command
20903 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
20904 package, to provide the history facility.
20905 @ifset SYSTEM_READLINE
20906 @xref{Using History Interactively, , , history, GNU History Library},
20908 @ifclear SYSTEM_READLINE
20909 @xref{Using History Interactively},
20911 for the detailed description of the History library.
20913 To issue a command to @value{GDBN} without affecting certain aspects of
20914 the state which is seen by users, prefix it with @samp{server }
20915 (@pxref{Server Prefix}). This
20916 means that this command will not affect the command history, nor will it
20917 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
20918 pressed on a line by itself.
20920 @cindex @code{server}, command prefix
20921 The server prefix does not affect the recording of values into the value
20922 history; to print a value without recording it into the value history,
20923 use the @code{output} command instead of the @code{print} command.
20925 Here is the description of @value{GDBN} commands related to command
20929 @cindex history substitution
20930 @cindex history file
20931 @kindex set history filename
20932 @cindex @env{GDBHISTFILE}, environment variable
20933 @item set history filename @var{fname}
20934 Set the name of the @value{GDBN} command history file to @var{fname}.
20935 This is the file where @value{GDBN} reads an initial command history
20936 list, and where it writes the command history from this session when it
20937 exits. You can access this list through history expansion or through
20938 the history command editing characters listed below. This file defaults
20939 to the value of the environment variable @code{GDBHISTFILE}, or to
20940 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
20943 @cindex save command history
20944 @kindex set history save
20945 @item set history save
20946 @itemx set history save on
20947 Record command history in a file, whose name may be specified with the
20948 @code{set history filename} command. By default, this option is disabled.
20950 @item set history save off
20951 Stop recording command history in a file.
20953 @cindex history size
20954 @kindex set history size
20955 @cindex @env{HISTSIZE}, environment variable
20956 @item set history size @var{size}
20957 Set the number of commands which @value{GDBN} keeps in its history list.
20958 This defaults to the value of the environment variable
20959 @code{HISTSIZE}, or to 256 if this variable is not set.
20962 History expansion assigns special meaning to the character @kbd{!}.
20963 @ifset SYSTEM_READLINE
20964 @xref{Event Designators, , , history, GNU History Library},
20966 @ifclear SYSTEM_READLINE
20967 @xref{Event Designators},
20971 @cindex history expansion, turn on/off
20972 Since @kbd{!} is also the logical not operator in C, history expansion
20973 is off by default. If you decide to enable history expansion with the
20974 @code{set history expansion on} command, you may sometimes need to
20975 follow @kbd{!} (when it is used as logical not, in an expression) with
20976 a space or a tab to prevent it from being expanded. The readline
20977 history facilities do not attempt substitution on the strings
20978 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
20980 The commands to control history expansion are:
20983 @item set history expansion on
20984 @itemx set history expansion
20985 @kindex set history expansion
20986 Enable history expansion. History expansion is off by default.
20988 @item set history expansion off
20989 Disable history expansion.
20992 @kindex show history
20994 @itemx show history filename
20995 @itemx show history save
20996 @itemx show history size
20997 @itemx show history expansion
20998 These commands display the state of the @value{GDBN} history parameters.
20999 @code{show history} by itself displays all four states.
21004 @kindex show commands
21005 @cindex show last commands
21006 @cindex display command history
21007 @item show commands
21008 Display the last ten commands in the command history.
21010 @item show commands @var{n}
21011 Print ten commands centered on command number @var{n}.
21013 @item show commands +
21014 Print ten commands just after the commands last printed.
21018 @section Screen Size
21019 @cindex size of screen
21020 @cindex pauses in output
21022 Certain commands to @value{GDBN} may produce large amounts of
21023 information output to the screen. To help you read all of it,
21024 @value{GDBN} pauses and asks you for input at the end of each page of
21025 output. Type @key{RET} when you want to continue the output, or @kbd{q}
21026 to discard the remaining output. Also, the screen width setting
21027 determines when to wrap lines of output. Depending on what is being
21028 printed, @value{GDBN} tries to break the line at a readable place,
21029 rather than simply letting it overflow onto the following line.
21031 Normally @value{GDBN} knows the size of the screen from the terminal
21032 driver software. For example, on Unix @value{GDBN} uses the termcap data base
21033 together with the value of the @code{TERM} environment variable and the
21034 @code{stty rows} and @code{stty cols} settings. If this is not correct,
21035 you can override it with the @code{set height} and @code{set
21042 @kindex show height
21043 @item set height @var{lpp}
21045 @itemx set width @var{cpl}
21047 These @code{set} commands specify a screen height of @var{lpp} lines and
21048 a screen width of @var{cpl} characters. The associated @code{show}
21049 commands display the current settings.
21051 If you specify a height of zero lines, @value{GDBN} does not pause during
21052 output no matter how long the output is. This is useful if output is to a
21053 file or to an editor buffer.
21055 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
21056 from wrapping its output.
21058 @item set pagination on
21059 @itemx set pagination off
21060 @kindex set pagination
21061 Turn the output pagination on or off; the default is on. Turning
21062 pagination off is the alternative to @code{set height 0}. Note that
21063 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
21064 Options, -batch}) also automatically disables pagination.
21066 @item show pagination
21067 @kindex show pagination
21068 Show the current pagination mode.
21073 @cindex number representation
21074 @cindex entering numbers
21076 You can always enter numbers in octal, decimal, or hexadecimal in
21077 @value{GDBN} by the usual conventions: octal numbers begin with
21078 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
21079 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
21080 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
21081 10; likewise, the default display for numbers---when no particular
21082 format is specified---is base 10. You can change the default base for
21083 both input and output with the commands described below.
21086 @kindex set input-radix
21087 @item set input-radix @var{base}
21088 Set the default base for numeric input. Supported choices
21089 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
21090 specified either unambiguously or using the current input radix; for
21094 set input-radix 012
21095 set input-radix 10.
21096 set input-radix 0xa
21100 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
21101 leaves the input radix unchanged, no matter what it was, since
21102 @samp{10}, being without any leading or trailing signs of its base, is
21103 interpreted in the current radix. Thus, if the current radix is 16,
21104 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
21107 @kindex set output-radix
21108 @item set output-radix @var{base}
21109 Set the default base for numeric display. Supported choices
21110 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
21111 specified either unambiguously or using the current input radix.
21113 @kindex show input-radix
21114 @item show input-radix
21115 Display the current default base for numeric input.
21117 @kindex show output-radix
21118 @item show output-radix
21119 Display the current default base for numeric display.
21121 @item set radix @r{[}@var{base}@r{]}
21125 These commands set and show the default base for both input and output
21126 of numbers. @code{set radix} sets the radix of input and output to
21127 the same base; without an argument, it resets the radix back to its
21128 default value of 10.
21133 @section Configuring the Current ABI
21135 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
21136 application automatically. However, sometimes you need to override its
21137 conclusions. Use these commands to manage @value{GDBN}'s view of the
21144 One @value{GDBN} configuration can debug binaries for multiple operating
21145 system targets, either via remote debugging or native emulation.
21146 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
21147 but you can override its conclusion using the @code{set osabi} command.
21148 One example where this is useful is in debugging of binaries which use
21149 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
21150 not have the same identifying marks that the standard C library for your
21155 Show the OS ABI currently in use.
21158 With no argument, show the list of registered available OS ABI's.
21160 @item set osabi @var{abi}
21161 Set the current OS ABI to @var{abi}.
21164 @cindex float promotion
21166 Generally, the way that an argument of type @code{float} is passed to a
21167 function depends on whether the function is prototyped. For a prototyped
21168 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
21169 according to the architecture's convention for @code{float}. For unprototyped
21170 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
21171 @code{double} and then passed.
21173 Unfortunately, some forms of debug information do not reliably indicate whether
21174 a function is prototyped. If @value{GDBN} calls a function that is not marked
21175 as prototyped, it consults @kbd{set coerce-float-to-double}.
21178 @kindex set coerce-float-to-double
21179 @item set coerce-float-to-double
21180 @itemx set coerce-float-to-double on
21181 Arguments of type @code{float} will be promoted to @code{double} when passed
21182 to an unprototyped function. This is the default setting.
21184 @item set coerce-float-to-double off
21185 Arguments of type @code{float} will be passed directly to unprototyped
21188 @kindex show coerce-float-to-double
21189 @item show coerce-float-to-double
21190 Show the current setting of promoting @code{float} to @code{double}.
21194 @kindex show cp-abi
21195 @value{GDBN} needs to know the ABI used for your program's C@t{++}
21196 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
21197 used to build your application. @value{GDBN} only fully supports
21198 programs with a single C@t{++} ABI; if your program contains code using
21199 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
21200 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
21201 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
21202 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
21203 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
21204 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
21209 Show the C@t{++} ABI currently in use.
21212 With no argument, show the list of supported C@t{++} ABI's.
21214 @item set cp-abi @var{abi}
21215 @itemx set cp-abi auto
21216 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
21220 @section Automatically loading associated files
21221 @cindex auto-loading
21223 @value{GDBN} sometimes reads files with commands and settings automatically,
21224 without being explicitly told so by the user. We call this feature
21225 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
21226 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
21227 results or introduce security risks (e.g., if the file comes from untrusted
21230 Note that loading of these associated files (including the local @file{.gdbinit}
21231 file) requires accordingly configured @code{auto-load safe-path}
21232 (@pxref{Auto-loading safe path}).
21234 For these reasons, @value{GDBN} includes commands and options to let you
21235 control when to auto-load files and which files should be auto-loaded.
21238 @anchor{set auto-load off}
21239 @kindex set auto-load off
21240 @item set auto-load off
21241 Globally disable loading of all auto-loaded files.
21242 You may want to use this command with the @samp{-iex} option
21243 (@pxref{Option -init-eval-command}) such as:
21245 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
21248 Be aware that system init file (@pxref{System-wide configuration})
21249 and init files from your home directory (@pxref{Home Directory Init File})
21250 still get read (as they come from generally trusted directories).
21251 To prevent @value{GDBN} from auto-loading even those init files, use the
21252 @option{-nx} option (@pxref{Mode Options}), in addition to
21253 @code{set auto-load no}.
21255 @anchor{show auto-load}
21256 @kindex show auto-load
21257 @item show auto-load
21258 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
21262 (gdb) show auto-load
21263 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
21264 libthread-db: Auto-loading of inferior specific libthread_db is on.
21265 local-gdbinit: Auto-loading of .gdbinit script from current directory
21267 python-scripts: Auto-loading of Python scripts is on.
21268 safe-path: List of directories from which it is safe to auto-load files
21269 is $debugdir:$datadir/auto-load.
21270 scripts-directory: List of directories from which to load auto-loaded scripts
21271 is $debugdir:$datadir/auto-load.
21274 @anchor{info auto-load}
21275 @kindex info auto-load
21276 @item info auto-load
21277 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
21281 (gdb) info auto-load
21284 Yes /home/user/gdb/gdb-gdb.gdb
21285 libthread-db: No auto-loaded libthread-db.
21286 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
21290 Yes /home/user/gdb/gdb-gdb.py
21294 These are various kinds of files @value{GDBN} can automatically load:
21298 @xref{objfile-gdb.py file}, controlled by @ref{set auto-load python-scripts}.
21300 @xref{objfile-gdb.gdb file}, controlled by @ref{set auto-load gdb-scripts}.
21302 @xref{dotdebug_gdb_scripts section},
21303 controlled by @ref{set auto-load python-scripts}.
21305 @xref{Init File in the Current Directory},
21306 controlled by @ref{set auto-load local-gdbinit}.
21308 @xref{libthread_db.so.1 file}, controlled by @ref{set auto-load libthread-db}.
21311 These are @value{GDBN} control commands for the auto-loading:
21313 @multitable @columnfractions .5 .5
21314 @item @xref{set auto-load off}.
21315 @tab Disable auto-loading globally.
21316 @item @xref{show auto-load}.
21317 @tab Show setting of all kinds of files.
21318 @item @xref{info auto-load}.
21319 @tab Show state of all kinds of files.
21320 @item @xref{set auto-load gdb-scripts}.
21321 @tab Control for @value{GDBN} command scripts.
21322 @item @xref{show auto-load gdb-scripts}.
21323 @tab Show setting of @value{GDBN} command scripts.
21324 @item @xref{info auto-load gdb-scripts}.
21325 @tab Show state of @value{GDBN} command scripts.
21326 @item @xref{set auto-load python-scripts}.
21327 @tab Control for @value{GDBN} Python scripts.
21328 @item @xref{show auto-load python-scripts}.
21329 @tab Show setting of @value{GDBN} Python scripts.
21330 @item @xref{info auto-load python-scripts}.
21331 @tab Show state of @value{GDBN} Python scripts.
21332 @item @xref{set auto-load scripts-directory}.
21333 @tab Control for @value{GDBN} auto-loaded scripts location.
21334 @item @xref{show auto-load scripts-directory}.
21335 @tab Show @value{GDBN} auto-loaded scripts location.
21336 @item @xref{set auto-load local-gdbinit}.
21337 @tab Control for init file in the current directory.
21338 @item @xref{show auto-load local-gdbinit}.
21339 @tab Show setting of init file in the current directory.
21340 @item @xref{info auto-load local-gdbinit}.
21341 @tab Show state of init file in the current directory.
21342 @item @xref{set auto-load libthread-db}.
21343 @tab Control for thread debugging library.
21344 @item @xref{show auto-load libthread-db}.
21345 @tab Show setting of thread debugging library.
21346 @item @xref{info auto-load libthread-db}.
21347 @tab Show state of thread debugging library.
21348 @item @xref{set auto-load safe-path}.
21349 @tab Control directories trusted for automatic loading.
21350 @item @xref{show auto-load safe-path}.
21351 @tab Show directories trusted for automatic loading.
21352 @item @xref{add-auto-load-safe-path}.
21353 @tab Add directory trusted for automatic loading.
21357 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
21358 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
21359 * objfile-gdb.gdb file:: @samp{set/show/info auto-load gdb-script}
21360 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
21361 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
21362 @xref{Python Auto-loading}.
21365 @node Init File in the Current Directory
21366 @subsection Automatically loading init file in the current directory
21367 @cindex auto-loading init file in the current directory
21369 By default, @value{GDBN} reads and executes the canned sequences of commands
21370 from init file (if any) in the current working directory,
21371 see @ref{Init File in the Current Directory during Startup}.
21373 Note that loading of this local @file{.gdbinit} file also requires accordingly
21374 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
21377 @anchor{set auto-load local-gdbinit}
21378 @kindex set auto-load local-gdbinit
21379 @item set auto-load local-gdbinit [on|off]
21380 Enable or disable the auto-loading of canned sequences of commands
21381 (@pxref{Sequences}) found in init file in the current directory.
21383 @anchor{show auto-load local-gdbinit}
21384 @kindex show auto-load local-gdbinit
21385 @item show auto-load local-gdbinit
21386 Show whether auto-loading of canned sequences of commands from init file in the
21387 current directory is enabled or disabled.
21389 @anchor{info auto-load local-gdbinit}
21390 @kindex info auto-load local-gdbinit
21391 @item info auto-load local-gdbinit
21392 Print whether canned sequences of commands from init file in the
21393 current directory have been auto-loaded.
21396 @node libthread_db.so.1 file
21397 @subsection Automatically loading thread debugging library
21398 @cindex auto-loading libthread_db.so.1
21400 This feature is currently present only on @sc{gnu}/Linux native hosts.
21402 @value{GDBN} reads in some cases thread debugging library from places specific
21403 to the inferior (@pxref{set libthread-db-search-path}).
21405 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
21406 without checking this @samp{set auto-load libthread-db} switch as system
21407 libraries have to be trusted in general. In all other cases of
21408 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
21409 auto-load libthread-db} is enabled before trying to open such thread debugging
21412 Note that loading of this debugging library also requires accordingly configured
21413 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
21416 @anchor{set auto-load libthread-db}
21417 @kindex set auto-load libthread-db
21418 @item set auto-load libthread-db [on|off]
21419 Enable or disable the auto-loading of inferior specific thread debugging library.
21421 @anchor{show auto-load libthread-db}
21422 @kindex show auto-load libthread-db
21423 @item show auto-load libthread-db
21424 Show whether auto-loading of inferior specific thread debugging library is
21425 enabled or disabled.
21427 @anchor{info auto-load libthread-db}
21428 @kindex info auto-load libthread-db
21429 @item info auto-load libthread-db
21430 Print the list of all loaded inferior specific thread debugging libraries and
21431 for each such library print list of inferior @var{pid}s using it.
21434 @node objfile-gdb.gdb file
21435 @subsection The @file{@var{objfile}-gdb.gdb} file
21436 @cindex auto-loading @file{@var{objfile}-gdb.gdb}
21438 @value{GDBN} tries to load an @file{@var{objfile}-gdb.gdb} file containing
21439 canned sequences of commands (@pxref{Sequences}), as long as @samp{set
21440 auto-load gdb-scripts} is set to @samp{on}.
21442 Note that loading of this script file also requires accordingly configured
21443 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
21445 For more background refer to the similar Python scripts auto-loading
21446 description (@pxref{objfile-gdb.py file}).
21449 @anchor{set auto-load gdb-scripts}
21450 @kindex set auto-load gdb-scripts
21451 @item set auto-load gdb-scripts [on|off]
21452 Enable or disable the auto-loading of canned sequences of commands scripts.
21454 @anchor{show auto-load gdb-scripts}
21455 @kindex show auto-load gdb-scripts
21456 @item show auto-load gdb-scripts
21457 Show whether auto-loading of canned sequences of commands scripts is enabled or
21460 @anchor{info auto-load gdb-scripts}
21461 @kindex info auto-load gdb-scripts
21462 @cindex print list of auto-loaded canned sequences of commands scripts
21463 @item info auto-load gdb-scripts [@var{regexp}]
21464 Print the list of all canned sequences of commands scripts that @value{GDBN}
21468 If @var{regexp} is supplied only canned sequences of commands scripts with
21469 matching names are printed.
21471 @node Auto-loading safe path
21472 @subsection Security restriction for auto-loading
21473 @cindex auto-loading safe-path
21475 As the files of inferior can come from untrusted source (such as submitted by
21476 an application user) @value{GDBN} does not always load any files automatically.
21477 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
21478 directories trusted for loading files not explicitly requested by user.
21479 Each directory can also be a shell wildcard pattern.
21481 If the path is not set properly you will see a warning and the file will not
21486 Reading symbols from /home/user/gdb/gdb...done.
21487 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
21488 declined by your `auto-load safe-path' set
21489 to "$debugdir:$datadir/auto-load".
21490 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
21491 declined by your `auto-load safe-path' set
21492 to "$debugdir:$datadir/auto-load".
21495 The list of trusted directories is controlled by the following commands:
21498 @anchor{set auto-load safe-path}
21499 @kindex set auto-load safe-path
21500 @item set auto-load safe-path @r{[}@var{directories}@r{]}
21501 Set the list of directories (and their subdirectories) trusted for automatic
21502 loading and execution of scripts. You can also enter a specific trusted file.
21503 Each directory can also be a shell wildcard pattern; wildcards do not match
21504 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
21505 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
21506 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
21507 its default value as specified during @value{GDBN} compilation.
21509 The list of directories uses path separator (@samp{:} on GNU and Unix
21510 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
21511 to the @env{PATH} environment variable.
21513 @anchor{show auto-load safe-path}
21514 @kindex show auto-load safe-path
21515 @item show auto-load safe-path
21516 Show the list of directories trusted for automatic loading and execution of
21519 @anchor{add-auto-load-safe-path}
21520 @kindex add-auto-load-safe-path
21521 @item add-auto-load-safe-path
21522 Add an entry (or list of entries) the list of directories trusted for automatic
21523 loading and execution of scripts. Multiple entries may be delimited by the
21524 host platform path separator in use.
21527 This variable defaults to what @code{--with-auto-load-dir} has been configured
21528 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
21529 substitution applies the same as for @ref{set auto-load scripts-directory}.
21530 The default @code{set auto-load safe-path} value can be also overriden by
21531 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
21533 Setting this variable to @file{/} disables this security protection,
21534 corresponding @value{GDBN} configuration option is
21535 @option{--without-auto-load-safe-path}.
21536 This variable is supposed to be set to the system directories writable by the
21537 system superuser only. Users can add their source directories in init files in
21538 their home directories (@pxref{Home Directory Init File}). See also deprecated
21539 init file in the current directory
21540 (@pxref{Init File in the Current Directory during Startup}).
21542 To force @value{GDBN} to load the files it declined to load in the previous
21543 example, you could use one of the following ways:
21546 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
21547 Specify this trusted directory (or a file) as additional component of the list.
21548 You have to specify also any existing directories displayed by
21549 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
21551 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
21552 Specify this directory as in the previous case but just for a single
21553 @value{GDBN} session.
21555 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
21556 Disable auto-loading safety for a single @value{GDBN} session.
21557 This assumes all the files you debug during this @value{GDBN} session will come
21558 from trusted sources.
21560 @item @kbd{./configure --without-auto-load-safe-path}
21561 During compilation of @value{GDBN} you may disable any auto-loading safety.
21562 This assumes all the files you will ever debug with this @value{GDBN} come from
21566 On the other hand you can also explicitly forbid automatic files loading which
21567 also suppresses any such warning messages:
21570 @item @kbd{gdb -iex "set auto-load no" @dots{}}
21571 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
21573 @item @file{~/.gdbinit}: @samp{set auto-load no}
21574 Disable auto-loading globally for the user
21575 (@pxref{Home Directory Init File}). While it is improbable, you could also
21576 use system init file instead (@pxref{System-wide configuration}).
21579 This setting applies to the file names as entered by user. If no entry matches
21580 @value{GDBN} tries as a last resort to also resolve all the file names into
21581 their canonical form (typically resolving symbolic links) and compare the
21582 entries again. @value{GDBN} already canonicalizes most of the filenames on its
21583 own before starting the comparison so a canonical form of directories is
21584 recommended to be entered.
21586 @node Auto-loading verbose mode
21587 @subsection Displaying files tried for auto-load
21588 @cindex auto-loading verbose mode
21590 For better visibility of all the file locations where you can place scripts to
21591 be auto-loaded with inferior --- or to protect yourself against accidental
21592 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
21593 all the files attempted to be loaded. Both existing and non-existing files may
21596 For example the list of directories from which it is safe to auto-load files
21597 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
21598 may not be too obvious while setting it up.
21601 (gdb) set debug auto-load on
21602 (gdb) file ~/src/t/true
21603 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
21604 for objfile "/tmp/true".
21605 auto-load: Updating directories of "/usr:/opt".
21606 auto-load: Using directory "/usr".
21607 auto-load: Using directory "/opt".
21608 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
21609 by your `auto-load safe-path' set to "/usr:/opt".
21613 @anchor{set debug auto-load}
21614 @kindex set debug auto-load
21615 @item set debug auto-load [on|off]
21616 Set whether to print the filenames attempted to be auto-loaded.
21618 @anchor{show debug auto-load}
21619 @kindex show debug auto-load
21620 @item show debug auto-load
21621 Show whether printing of the filenames attempted to be auto-loaded is turned
21625 @node Messages/Warnings
21626 @section Optional Warnings and Messages
21628 @cindex verbose operation
21629 @cindex optional warnings
21630 By default, @value{GDBN} is silent about its inner workings. If you are
21631 running on a slow machine, you may want to use the @code{set verbose}
21632 command. This makes @value{GDBN} tell you when it does a lengthy
21633 internal operation, so you will not think it has crashed.
21635 Currently, the messages controlled by @code{set verbose} are those
21636 which announce that the symbol table for a source file is being read;
21637 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
21640 @kindex set verbose
21641 @item set verbose on
21642 Enables @value{GDBN} output of certain informational messages.
21644 @item set verbose off
21645 Disables @value{GDBN} output of certain informational messages.
21647 @kindex show verbose
21649 Displays whether @code{set verbose} is on or off.
21652 By default, if @value{GDBN} encounters bugs in the symbol table of an
21653 object file, it is silent; but if you are debugging a compiler, you may
21654 find this information useful (@pxref{Symbol Errors, ,Errors Reading
21659 @kindex set complaints
21660 @item set complaints @var{limit}
21661 Permits @value{GDBN} to output @var{limit} complaints about each type of
21662 unusual symbols before becoming silent about the problem. Set
21663 @var{limit} to zero to suppress all complaints; set it to a large number
21664 to prevent complaints from being suppressed.
21666 @kindex show complaints
21667 @item show complaints
21668 Displays how many symbol complaints @value{GDBN} is permitted to produce.
21672 @anchor{confirmation requests}
21673 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
21674 lot of stupid questions to confirm certain commands. For example, if
21675 you try to run a program which is already running:
21679 The program being debugged has been started already.
21680 Start it from the beginning? (y or n)
21683 If you are willing to unflinchingly face the consequences of your own
21684 commands, you can disable this ``feature'':
21688 @kindex set confirm
21690 @cindex confirmation
21691 @cindex stupid questions
21692 @item set confirm off
21693 Disables confirmation requests. Note that running @value{GDBN} with
21694 the @option{--batch} option (@pxref{Mode Options, -batch}) also
21695 automatically disables confirmation requests.
21697 @item set confirm on
21698 Enables confirmation requests (the default).
21700 @kindex show confirm
21702 Displays state of confirmation requests.
21706 @cindex command tracing
21707 If you need to debug user-defined commands or sourced files you may find it
21708 useful to enable @dfn{command tracing}. In this mode each command will be
21709 printed as it is executed, prefixed with one or more @samp{+} symbols, the
21710 quantity denoting the call depth of each command.
21713 @kindex set trace-commands
21714 @cindex command scripts, debugging
21715 @item set trace-commands on
21716 Enable command tracing.
21717 @item set trace-commands off
21718 Disable command tracing.
21719 @item show trace-commands
21720 Display the current state of command tracing.
21723 @node Debugging Output
21724 @section Optional Messages about Internal Happenings
21725 @cindex optional debugging messages
21727 @value{GDBN} has commands that enable optional debugging messages from
21728 various @value{GDBN} subsystems; normally these commands are of
21729 interest to @value{GDBN} maintainers, or when reporting a bug. This
21730 section documents those commands.
21733 @kindex set exec-done-display
21734 @item set exec-done-display
21735 Turns on or off the notification of asynchronous commands'
21736 completion. When on, @value{GDBN} will print a message when an
21737 asynchronous command finishes its execution. The default is off.
21738 @kindex show exec-done-display
21739 @item show exec-done-display
21740 Displays the current setting of asynchronous command completion
21743 @cindex gdbarch debugging info
21744 @cindex architecture debugging info
21745 @item set debug arch
21746 Turns on or off display of gdbarch debugging info. The default is off
21748 @item show debug arch
21749 Displays the current state of displaying gdbarch debugging info.
21750 @item set debug aix-thread
21751 @cindex AIX threads
21752 Display debugging messages about inner workings of the AIX thread
21754 @item show debug aix-thread
21755 Show the current state of AIX thread debugging info display.
21756 @item set debug check-physname
21758 Check the results of the ``physname'' computation. When reading DWARF
21759 debugging information for C@t{++}, @value{GDBN} attempts to compute
21760 each entity's name. @value{GDBN} can do this computation in two
21761 different ways, depending on exactly what information is present.
21762 When enabled, this setting causes @value{GDBN} to compute the names
21763 both ways and display any discrepancies.
21764 @item show debug check-physname
21765 Show the current state of ``physname'' checking.
21766 @item set debug dwarf2-die
21767 @cindex DWARF2 DIEs
21768 Dump DWARF2 DIEs after they are read in.
21769 The value is the number of nesting levels to print.
21770 A value of zero turns off the display.
21771 @item show debug dwarf2-die
21772 Show the current state of DWARF2 DIE debugging.
21773 @item set debug dwarf2-read
21774 @cindex DWARF2 Reading
21775 Turns on or off display of debugging messages related to reading
21776 DWARF debug info. The default is off.
21777 @item show debug dwarf2-read
21778 Show the current state of DWARF2 reader debugging.
21779 @item set debug displaced
21780 @cindex displaced stepping debugging info
21781 Turns on or off display of @value{GDBN} debugging info for the
21782 displaced stepping support. The default is off.
21783 @item show debug displaced
21784 Displays the current state of displaying @value{GDBN} debugging info
21785 related to displaced stepping.
21786 @item set debug event
21787 @cindex event debugging info
21788 Turns on or off display of @value{GDBN} event debugging info. The
21790 @item show debug event
21791 Displays the current state of displaying @value{GDBN} event debugging
21793 @item set debug expression
21794 @cindex expression debugging info
21795 Turns on or off display of debugging info about @value{GDBN}
21796 expression parsing. The default is off.
21797 @item show debug expression
21798 Displays the current state of displaying debugging info about
21799 @value{GDBN} expression parsing.
21800 @item set debug frame
21801 @cindex frame debugging info
21802 Turns on or off display of @value{GDBN} frame debugging info. The
21804 @item show debug frame
21805 Displays the current state of displaying @value{GDBN} frame debugging
21807 @item set debug gnu-nat
21808 @cindex @sc{gnu}/Hurd debug messages
21809 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
21810 @item show debug gnu-nat
21811 Show the current state of @sc{gnu}/Hurd debugging messages.
21812 @item set debug infrun
21813 @cindex inferior debugging info
21814 Turns on or off display of @value{GDBN} debugging info for running the inferior.
21815 The default is off. @file{infrun.c} contains GDB's runtime state machine used
21816 for implementing operations such as single-stepping the inferior.
21817 @item show debug infrun
21818 Displays the current state of @value{GDBN} inferior debugging.
21819 @item set debug jit
21820 @cindex just-in-time compilation, debugging messages
21821 Turns on or off debugging messages from JIT debug support.
21822 @item show debug jit
21823 Displays the current state of @value{GDBN} JIT debugging.
21824 @item set debug lin-lwp
21825 @cindex @sc{gnu}/Linux LWP debug messages
21826 @cindex Linux lightweight processes
21827 Turns on or off debugging messages from the Linux LWP debug support.
21828 @item show debug lin-lwp
21829 Show the current state of Linux LWP debugging messages.
21830 @item set debug observer
21831 @cindex observer debugging info
21832 Turns on or off display of @value{GDBN} observer debugging. This
21833 includes info such as the notification of observable events.
21834 @item show debug observer
21835 Displays the current state of observer debugging.
21836 @item set debug overload
21837 @cindex C@t{++} overload debugging info
21838 Turns on or off display of @value{GDBN} C@t{++} overload debugging
21839 info. This includes info such as ranking of functions, etc. The default
21841 @item show debug overload
21842 Displays the current state of displaying @value{GDBN} C@t{++} overload
21844 @cindex expression parser, debugging info
21845 @cindex debug expression parser
21846 @item set debug parser
21847 Turns on or off the display of expression parser debugging output.
21848 Internally, this sets the @code{yydebug} variable in the expression
21849 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
21850 details. The default is off.
21851 @item show debug parser
21852 Show the current state of expression parser debugging.
21853 @cindex packets, reporting on stdout
21854 @cindex serial connections, debugging
21855 @cindex debug remote protocol
21856 @cindex remote protocol debugging
21857 @cindex display remote packets
21858 @item set debug remote
21859 Turns on or off display of reports on all packets sent back and forth across
21860 the serial line to the remote machine. The info is printed on the
21861 @value{GDBN} standard output stream. The default is off.
21862 @item show debug remote
21863 Displays the state of display of remote packets.
21864 @item set debug serial
21865 Turns on or off display of @value{GDBN} serial debugging info. The
21867 @item show debug serial
21868 Displays the current state of displaying @value{GDBN} serial debugging
21870 @item set debug solib-frv
21871 @cindex FR-V shared-library debugging
21872 Turns on or off debugging messages for FR-V shared-library code.
21873 @item show debug solib-frv
21874 Display the current state of FR-V shared-library code debugging
21876 @item set debug symtab-create
21877 @cindex symbol table creation
21878 Turns on or off display of debugging messages related to symbol table creation.
21879 The default is off.
21880 @item show debug symtab-create
21881 Show the current state of symbol table creation debugging.
21882 @item set debug target
21883 @cindex target debugging info
21884 Turns on or off display of @value{GDBN} target debugging info. This info
21885 includes what is going on at the target level of GDB, as it happens. The
21886 default is 0. Set it to 1 to track events, and to 2 to also track the
21887 value of large memory transfers. Changes to this flag do not take effect
21888 until the next time you connect to a target or use the @code{run} command.
21889 @item show debug target
21890 Displays the current state of displaying @value{GDBN} target debugging
21892 @item set debug timestamp
21893 @cindex timestampping debugging info
21894 Turns on or off display of timestamps with @value{GDBN} debugging info.
21895 When enabled, seconds and microseconds are displayed before each debugging
21897 @item show debug timestamp
21898 Displays the current state of displaying timestamps with @value{GDBN}
21900 @item set debugvarobj
21901 @cindex variable object debugging info
21902 Turns on or off display of @value{GDBN} variable object debugging
21903 info. The default is off.
21904 @item show debugvarobj
21905 Displays the current state of displaying @value{GDBN} variable object
21907 @item set debug xml
21908 @cindex XML parser debugging
21909 Turns on or off debugging messages for built-in XML parsers.
21910 @item show debug xml
21911 Displays the current state of XML debugging messages.
21914 @node Other Misc Settings
21915 @section Other Miscellaneous Settings
21916 @cindex miscellaneous settings
21919 @kindex set interactive-mode
21920 @item set interactive-mode
21921 If @code{on}, forces @value{GDBN} to assume that GDB was started
21922 in a terminal. In practice, this means that @value{GDBN} should wait
21923 for the user to answer queries generated by commands entered at
21924 the command prompt. If @code{off}, forces @value{GDBN} to operate
21925 in the opposite mode, and it uses the default answers to all queries.
21926 If @code{auto} (the default), @value{GDBN} tries to determine whether
21927 its standard input is a terminal, and works in interactive-mode if it
21928 is, non-interactively otherwise.
21930 In the vast majority of cases, the debugger should be able to guess
21931 correctly which mode should be used. But this setting can be useful
21932 in certain specific cases, such as running a MinGW @value{GDBN}
21933 inside a cygwin window.
21935 @kindex show interactive-mode
21936 @item show interactive-mode
21937 Displays whether the debugger is operating in interactive mode or not.
21940 @node Extending GDB
21941 @chapter Extending @value{GDBN}
21942 @cindex extending GDB
21944 @value{GDBN} provides three mechanisms for extension. The first is based
21945 on composition of @value{GDBN} commands, the second is based on the
21946 Python scripting language, and the third is for defining new aliases of
21949 To facilitate the use of the first two extensions, @value{GDBN} is capable
21950 of evaluating the contents of a file. When doing so, @value{GDBN}
21951 can recognize which scripting language is being used by looking at
21952 the filename extension. Files with an unrecognized filename extension
21953 are always treated as a @value{GDBN} Command Files.
21954 @xref{Command Files,, Command files}.
21956 You can control how @value{GDBN} evaluates these files with the following
21960 @kindex set script-extension
21961 @kindex show script-extension
21962 @item set script-extension off
21963 All scripts are always evaluated as @value{GDBN} Command Files.
21965 @item set script-extension soft
21966 The debugger determines the scripting language based on filename
21967 extension. If this scripting language is supported, @value{GDBN}
21968 evaluates the script using that language. Otherwise, it evaluates
21969 the file as a @value{GDBN} Command File.
21971 @item set script-extension strict
21972 The debugger determines the scripting language based on filename
21973 extension, and evaluates the script using that language. If the
21974 language is not supported, then the evaluation fails.
21976 @item show script-extension
21977 Display the current value of the @code{script-extension} option.
21982 * Sequences:: Canned Sequences of Commands
21983 * Python:: Scripting @value{GDBN} using Python
21984 * Aliases:: Creating new spellings of existing commands
21988 @section Canned Sequences of Commands
21990 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
21991 Command Lists}), @value{GDBN} provides two ways to store sequences of
21992 commands for execution as a unit: user-defined commands and command
21996 * Define:: How to define your own commands
21997 * Hooks:: Hooks for user-defined commands
21998 * Command Files:: How to write scripts of commands to be stored in a file
21999 * Output:: Commands for controlled output
22003 @subsection User-defined Commands
22005 @cindex user-defined command
22006 @cindex arguments, to user-defined commands
22007 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
22008 which you assign a new name as a command. This is done with the
22009 @code{define} command. User commands may accept up to 10 arguments
22010 separated by whitespace. Arguments are accessed within the user command
22011 via @code{$arg0@dots{}$arg9}. A trivial example:
22015 print $arg0 + $arg1 + $arg2
22020 To execute the command use:
22027 This defines the command @code{adder}, which prints the sum of
22028 its three arguments. Note the arguments are text substitutions, so they may
22029 reference variables, use complex expressions, or even perform inferior
22032 @cindex argument count in user-defined commands
22033 @cindex how many arguments (user-defined commands)
22034 In addition, @code{$argc} may be used to find out how many arguments have
22035 been passed. This expands to a number in the range 0@dots{}10.
22040 print $arg0 + $arg1
22043 print $arg0 + $arg1 + $arg2
22051 @item define @var{commandname}
22052 Define a command named @var{commandname}. If there is already a command
22053 by that name, you are asked to confirm that you want to redefine it.
22054 @var{commandname} may be a bare command name consisting of letters,
22055 numbers, dashes, and underscores. It may also start with any predefined
22056 prefix command. For example, @samp{define target my-target} creates
22057 a user-defined @samp{target my-target} command.
22059 The definition of the command is made up of other @value{GDBN} command lines,
22060 which are given following the @code{define} command. The end of these
22061 commands is marked by a line containing @code{end}.
22064 @kindex end@r{ (user-defined commands)}
22065 @item document @var{commandname}
22066 Document the user-defined command @var{commandname}, so that it can be
22067 accessed by @code{help}. The command @var{commandname} must already be
22068 defined. This command reads lines of documentation just as @code{define}
22069 reads the lines of the command definition, ending with @code{end}.
22070 After the @code{document} command is finished, @code{help} on command
22071 @var{commandname} displays the documentation you have written.
22073 You may use the @code{document} command again to change the
22074 documentation of a command. Redefining the command with @code{define}
22075 does not change the documentation.
22077 @kindex dont-repeat
22078 @cindex don't repeat command
22080 Used inside a user-defined command, this tells @value{GDBN} that this
22081 command should not be repeated when the user hits @key{RET}
22082 (@pxref{Command Syntax, repeat last command}).
22084 @kindex help user-defined
22085 @item help user-defined
22086 List all user-defined commands and all python commands defined in class
22087 COMAND_USER. The first line of the documentation or docstring is
22092 @itemx show user @var{commandname}
22093 Display the @value{GDBN} commands used to define @var{commandname} (but
22094 not its documentation). If no @var{commandname} is given, display the
22095 definitions for all user-defined commands.
22096 This does not work for user-defined python commands.
22098 @cindex infinite recursion in user-defined commands
22099 @kindex show max-user-call-depth
22100 @kindex set max-user-call-depth
22101 @item show max-user-call-depth
22102 @itemx set max-user-call-depth
22103 The value of @code{max-user-call-depth} controls how many recursion
22104 levels are allowed in user-defined commands before @value{GDBN} suspects an
22105 infinite recursion and aborts the command.
22106 This does not apply to user-defined python commands.
22109 In addition to the above commands, user-defined commands frequently
22110 use control flow commands, described in @ref{Command Files}.
22112 When user-defined commands are executed, the
22113 commands of the definition are not printed. An error in any command
22114 stops execution of the user-defined command.
22116 If used interactively, commands that would ask for confirmation proceed
22117 without asking when used inside a user-defined command. Many @value{GDBN}
22118 commands that normally print messages to say what they are doing omit the
22119 messages when used in a user-defined command.
22122 @subsection User-defined Command Hooks
22123 @cindex command hooks
22124 @cindex hooks, for commands
22125 @cindex hooks, pre-command
22128 You may define @dfn{hooks}, which are a special kind of user-defined
22129 command. Whenever you run the command @samp{foo}, if the user-defined
22130 command @samp{hook-foo} exists, it is executed (with no arguments)
22131 before that command.
22133 @cindex hooks, post-command
22135 A hook may also be defined which is run after the command you executed.
22136 Whenever you run the command @samp{foo}, if the user-defined command
22137 @samp{hookpost-foo} exists, it is executed (with no arguments) after
22138 that command. Post-execution hooks may exist simultaneously with
22139 pre-execution hooks, for the same command.
22141 It is valid for a hook to call the command which it hooks. If this
22142 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
22144 @c It would be nice if hookpost could be passed a parameter indicating
22145 @c if the command it hooks executed properly or not. FIXME!
22147 @kindex stop@r{, a pseudo-command}
22148 In addition, a pseudo-command, @samp{stop} exists. Defining
22149 (@samp{hook-stop}) makes the associated commands execute every time
22150 execution stops in your program: before breakpoint commands are run,
22151 displays are printed, or the stack frame is printed.
22153 For example, to ignore @code{SIGALRM} signals while
22154 single-stepping, but treat them normally during normal execution,
22159 handle SIGALRM nopass
22163 handle SIGALRM pass
22166 define hook-continue
22167 handle SIGALRM pass
22171 As a further example, to hook at the beginning and end of the @code{echo}
22172 command, and to add extra text to the beginning and end of the message,
22180 define hookpost-echo
22184 (@value{GDBP}) echo Hello World
22185 <<<---Hello World--->>>
22190 You can define a hook for any single-word command in @value{GDBN}, but
22191 not for command aliases; you should define a hook for the basic command
22192 name, e.g.@: @code{backtrace} rather than @code{bt}.
22193 @c FIXME! So how does Joe User discover whether a command is an alias
22195 You can hook a multi-word command by adding @code{hook-} or
22196 @code{hookpost-} to the last word of the command, e.g.@:
22197 @samp{define target hook-remote} to add a hook to @samp{target remote}.
22199 If an error occurs during the execution of your hook, execution of
22200 @value{GDBN} commands stops and @value{GDBN} issues a prompt
22201 (before the command that you actually typed had a chance to run).
22203 If you try to define a hook which does not match any known command, you
22204 get a warning from the @code{define} command.
22206 @node Command Files
22207 @subsection Command Files
22209 @cindex command files
22210 @cindex scripting commands
22211 A command file for @value{GDBN} is a text file made of lines that are
22212 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
22213 also be included. An empty line in a command file does nothing; it
22214 does not mean to repeat the last command, as it would from the
22217 You can request the execution of a command file with the @code{source}
22218 command. Note that the @code{source} command is also used to evaluate
22219 scripts that are not Command Files. The exact behavior can be configured
22220 using the @code{script-extension} setting.
22221 @xref{Extending GDB,, Extending GDB}.
22225 @cindex execute commands from a file
22226 @item source [-s] [-v] @var{filename}
22227 Execute the command file @var{filename}.
22230 The lines in a command file are generally executed sequentially,
22231 unless the order of execution is changed by one of the
22232 @emph{flow-control commands} described below. The commands are not
22233 printed as they are executed. An error in any command terminates
22234 execution of the command file and control is returned to the console.
22236 @value{GDBN} first searches for @var{filename} in the current directory.
22237 If the file is not found there, and @var{filename} does not specify a
22238 directory, then @value{GDBN} also looks for the file on the source search path
22239 (specified with the @samp{directory} command);
22240 except that @file{$cdir} is not searched because the compilation directory
22241 is not relevant to scripts.
22243 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
22244 on the search path even if @var{filename} specifies a directory.
22245 The search is done by appending @var{filename} to each element of the
22246 search path. So, for example, if @var{filename} is @file{mylib/myscript}
22247 and the search path contains @file{/home/user} then @value{GDBN} will
22248 look for the script @file{/home/user/mylib/myscript}.
22249 The search is also done if @var{filename} is an absolute path.
22250 For example, if @var{filename} is @file{/tmp/myscript} and
22251 the search path contains @file{/home/user} then @value{GDBN} will
22252 look for the script @file{/home/user/tmp/myscript}.
22253 For DOS-like systems, if @var{filename} contains a drive specification,
22254 it is stripped before concatenation. For example, if @var{filename} is
22255 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
22256 will look for the script @file{c:/tmp/myscript}.
22258 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
22259 each command as it is executed. The option must be given before
22260 @var{filename}, and is interpreted as part of the filename anywhere else.
22262 Commands that would ask for confirmation if used interactively proceed
22263 without asking when used in a command file. Many @value{GDBN} commands that
22264 normally print messages to say what they are doing omit the messages
22265 when called from command files.
22267 @value{GDBN} also accepts command input from standard input. In this
22268 mode, normal output goes to standard output and error output goes to
22269 standard error. Errors in a command file supplied on standard input do
22270 not terminate execution of the command file---execution continues with
22274 gdb < cmds > log 2>&1
22277 (The syntax above will vary depending on the shell used.) This example
22278 will execute commands from the file @file{cmds}. All output and errors
22279 would be directed to @file{log}.
22281 Since commands stored on command files tend to be more general than
22282 commands typed interactively, they frequently need to deal with
22283 complicated situations, such as different or unexpected values of
22284 variables and symbols, changes in how the program being debugged is
22285 built, etc. @value{GDBN} provides a set of flow-control commands to
22286 deal with these complexities. Using these commands, you can write
22287 complex scripts that loop over data structures, execute commands
22288 conditionally, etc.
22295 This command allows to include in your script conditionally executed
22296 commands. The @code{if} command takes a single argument, which is an
22297 expression to evaluate. It is followed by a series of commands that
22298 are executed only if the expression is true (its value is nonzero).
22299 There can then optionally be an @code{else} line, followed by a series
22300 of commands that are only executed if the expression was false. The
22301 end of the list is marked by a line containing @code{end}.
22305 This command allows to write loops. Its syntax is similar to
22306 @code{if}: the command takes a single argument, which is an expression
22307 to evaluate, and must be followed by the commands to execute, one per
22308 line, terminated by an @code{end}. These commands are called the
22309 @dfn{body} of the loop. The commands in the body of @code{while} are
22310 executed repeatedly as long as the expression evaluates to true.
22314 This command exits the @code{while} loop in whose body it is included.
22315 Execution of the script continues after that @code{while}s @code{end}
22318 @kindex loop_continue
22319 @item loop_continue
22320 This command skips the execution of the rest of the body of commands
22321 in the @code{while} loop in whose body it is included. Execution
22322 branches to the beginning of the @code{while} loop, where it evaluates
22323 the controlling expression.
22325 @kindex end@r{ (if/else/while commands)}
22327 Terminate the block of commands that are the body of @code{if},
22328 @code{else}, or @code{while} flow-control commands.
22333 @subsection Commands for Controlled Output
22335 During the execution of a command file or a user-defined command, normal
22336 @value{GDBN} output is suppressed; the only output that appears is what is
22337 explicitly printed by the commands in the definition. This section
22338 describes three commands useful for generating exactly the output you
22343 @item echo @var{text}
22344 @c I do not consider backslash-space a standard C escape sequence
22345 @c because it is not in ANSI.
22346 Print @var{text}. Nonprinting characters can be included in
22347 @var{text} using C escape sequences, such as @samp{\n} to print a
22348 newline. @strong{No newline is printed unless you specify one.}
22349 In addition to the standard C escape sequences, a backslash followed
22350 by a space stands for a space. This is useful for displaying a
22351 string with spaces at the beginning or the end, since leading and
22352 trailing spaces are otherwise trimmed from all arguments.
22353 To print @samp{@w{ }and foo =@w{ }}, use the command
22354 @samp{echo \@w{ }and foo = \@w{ }}.
22356 A backslash at the end of @var{text} can be used, as in C, to continue
22357 the command onto subsequent lines. For example,
22360 echo This is some text\n\
22361 which is continued\n\
22362 onto several lines.\n
22365 produces the same output as
22368 echo This is some text\n
22369 echo which is continued\n
22370 echo onto several lines.\n
22374 @item output @var{expression}
22375 Print the value of @var{expression} and nothing but that value: no
22376 newlines, no @samp{$@var{nn} = }. The value is not entered in the
22377 value history either. @xref{Expressions, ,Expressions}, for more information
22380 @item output/@var{fmt} @var{expression}
22381 Print the value of @var{expression} in format @var{fmt}. You can use
22382 the same formats as for @code{print}. @xref{Output Formats,,Output
22383 Formats}, for more information.
22386 @item printf @var{template}, @var{expressions}@dots{}
22387 Print the values of one or more @var{expressions} under the control of
22388 the string @var{template}. To print several values, make
22389 @var{expressions} be a comma-separated list of individual expressions,
22390 which may be either numbers or pointers. Their values are printed as
22391 specified by @var{template}, exactly as a C program would do by
22392 executing the code below:
22395 printf (@var{template}, @var{expressions}@dots{});
22398 As in @code{C} @code{printf}, ordinary characters in @var{template}
22399 are printed verbatim, while @dfn{conversion specification} introduced
22400 by the @samp{%} character cause subsequent @var{expressions} to be
22401 evaluated, their values converted and formatted according to type and
22402 style information encoded in the conversion specifications, and then
22405 For example, you can print two values in hex like this:
22408 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
22411 @code{printf} supports all the standard @code{C} conversion
22412 specifications, including the flags and modifiers between the @samp{%}
22413 character and the conversion letter, with the following exceptions:
22417 The argument-ordering modifiers, such as @samp{2$}, are not supported.
22420 The modifier @samp{*} is not supported for specifying precision or
22424 The @samp{'} flag (for separation of digits into groups according to
22425 @code{LC_NUMERIC'}) is not supported.
22428 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
22432 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
22435 The conversion letters @samp{a} and @samp{A} are not supported.
22439 Note that the @samp{ll} type modifier is supported only if the
22440 underlying @code{C} implementation used to build @value{GDBN} supports
22441 the @code{long long int} type, and the @samp{L} type modifier is
22442 supported only if @code{long double} type is available.
22444 As in @code{C}, @code{printf} supports simple backslash-escape
22445 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
22446 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
22447 single character. Octal and hexadecimal escape sequences are not
22450 Additionally, @code{printf} supports conversion specifications for DFP
22451 (@dfn{Decimal Floating Point}) types using the following length modifiers
22452 together with a floating point specifier.
22457 @samp{H} for printing @code{Decimal32} types.
22460 @samp{D} for printing @code{Decimal64} types.
22463 @samp{DD} for printing @code{Decimal128} types.
22466 If the underlying @code{C} implementation used to build @value{GDBN} has
22467 support for the three length modifiers for DFP types, other modifiers
22468 such as width and precision will also be available for @value{GDBN} to use.
22470 In case there is no such @code{C} support, no additional modifiers will be
22471 available and the value will be printed in the standard way.
22473 Here's an example of printing DFP types using the above conversion letters:
22475 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
22479 @item eval @var{template}, @var{expressions}@dots{}
22480 Convert the values of one or more @var{expressions} under the control of
22481 the string @var{template} to a command line, and call it.
22486 @section Scripting @value{GDBN} using Python
22487 @cindex python scripting
22488 @cindex scripting with python
22490 You can script @value{GDBN} using the @uref{http://www.python.org/,
22491 Python programming language}. This feature is available only if
22492 @value{GDBN} was configured using @option{--with-python}.
22494 @cindex python directory
22495 Python scripts used by @value{GDBN} should be installed in
22496 @file{@var{data-directory}/python}, where @var{data-directory} is
22497 the data directory as determined at @value{GDBN} startup (@pxref{Data Files}).
22498 This directory, known as the @dfn{python directory},
22499 is automatically added to the Python Search Path in order to allow
22500 the Python interpreter to locate all scripts installed at this location.
22502 Additionally, @value{GDBN} commands and convenience functions which
22503 are written in Python and are located in the
22504 @file{@var{data-directory}/python/gdb/command} or
22505 @file{@var{data-directory}/python/gdb/function} directories are
22506 automatically imported when @value{GDBN} starts.
22509 * Python Commands:: Accessing Python from @value{GDBN}.
22510 * Python API:: Accessing @value{GDBN} from Python.
22511 * Python Auto-loading:: Automatically loading Python code.
22512 * Python modules:: Python modules provided by @value{GDBN}.
22515 @node Python Commands
22516 @subsection Python Commands
22517 @cindex python commands
22518 @cindex commands to access python
22520 @value{GDBN} provides one command for accessing the Python interpreter,
22521 and one related setting:
22525 @item python @r{[}@var{code}@r{]}
22526 The @code{python} command can be used to evaluate Python code.
22528 If given an argument, the @code{python} command will evaluate the
22529 argument as a Python command. For example:
22532 (@value{GDBP}) python print 23
22536 If you do not provide an argument to @code{python}, it will act as a
22537 multi-line command, like @code{define}. In this case, the Python
22538 script is made up of subsequent command lines, given after the
22539 @code{python} command. This command list is terminated using a line
22540 containing @code{end}. For example:
22543 (@value{GDBP}) python
22545 End with a line saying just "end".
22551 @kindex set python print-stack
22552 @item set python print-stack
22553 By default, @value{GDBN} will print only the message component of a
22554 Python exception when an error occurs in a Python script. This can be
22555 controlled using @code{set python print-stack}: if @code{full}, then
22556 full Python stack printing is enabled; if @code{none}, then Python stack
22557 and message printing is disabled; if @code{message}, the default, only
22558 the message component of the error is printed.
22561 It is also possible to execute a Python script from the @value{GDBN}
22565 @item source @file{script-name}
22566 The script name must end with @samp{.py} and @value{GDBN} must be configured
22567 to recognize the script language based on filename extension using
22568 the @code{script-extension} setting. @xref{Extending GDB, ,Extending GDB}.
22570 @item python execfile ("script-name")
22571 This method is based on the @code{execfile} Python built-in function,
22572 and thus is always available.
22576 @subsection Python API
22578 @cindex programming in python
22580 @cindex python stdout
22581 @cindex python pagination
22582 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
22583 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
22584 A Python program which outputs to one of these streams may have its
22585 output interrupted by the user (@pxref{Screen Size}). In this
22586 situation, a Python @code{KeyboardInterrupt} exception is thrown.
22589 * Basic Python:: Basic Python Functions.
22590 * Exception Handling:: How Python exceptions are translated.
22591 * Values From Inferior:: Python representation of values.
22592 * Types In Python:: Python representation of types.
22593 * Pretty Printing API:: Pretty-printing values.
22594 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
22595 * Writing a Pretty-Printer:: Writing a Pretty-Printer.
22596 * Inferiors In Python:: Python representation of inferiors (processes)
22597 * Events In Python:: Listening for events from @value{GDBN}.
22598 * Threads In Python:: Accessing inferior threads from Python.
22599 * Commands In Python:: Implementing new commands in Python.
22600 * Parameters In Python:: Adding new @value{GDBN} parameters.
22601 * Functions In Python:: Writing new convenience functions.
22602 * Progspaces In Python:: Program spaces.
22603 * Objfiles In Python:: Object files.
22604 * Frames In Python:: Accessing inferior stack frames from Python.
22605 * Blocks In Python:: Accessing frame blocks from Python.
22606 * Symbols In Python:: Python representation of symbols.
22607 * Symbol Tables In Python:: Python representation of symbol tables.
22608 * Breakpoints In Python:: Manipulating breakpoints using Python.
22609 * Finish Breakpoints in Python:: Setting Breakpoints on function return
22611 * Lazy Strings In Python:: Python representation of lazy strings.
22615 @subsubsection Basic Python
22617 @cindex python functions
22618 @cindex python module
22620 @value{GDBN} introduces a new Python module, named @code{gdb}. All
22621 methods and classes added by @value{GDBN} are placed in this module.
22622 @value{GDBN} automatically @code{import}s the @code{gdb} module for
22623 use in all scripts evaluated by the @code{python} command.
22625 @findex gdb.PYTHONDIR
22626 @defvar gdb.PYTHONDIR
22627 A string containing the python directory (@pxref{Python}).
22630 @findex gdb.execute
22631 @defun gdb.execute (command @r{[}, from_tty @r{[}, to_string@r{]]})
22632 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
22633 If a GDB exception happens while @var{command} runs, it is
22634 translated as described in @ref{Exception Handling,,Exception Handling}.
22636 @var{from_tty} specifies whether @value{GDBN} ought to consider this
22637 command as having originated from the user invoking it interactively.
22638 It must be a boolean value. If omitted, it defaults to @code{False}.
22640 By default, any output produced by @var{command} is sent to
22641 @value{GDBN}'s standard output. If the @var{to_string} parameter is
22642 @code{True}, then output will be collected by @code{gdb.execute} and
22643 returned as a string. The default is @code{False}, in which case the
22644 return value is @code{None}. If @var{to_string} is @code{True}, the
22645 @value{GDBN} virtual terminal will be temporarily set to unlimited width
22646 and height, and its pagination will be disabled; @pxref{Screen Size}.
22649 @findex gdb.breakpoints
22650 @defun gdb.breakpoints ()
22651 Return a sequence holding all of @value{GDBN}'s breakpoints.
22652 @xref{Breakpoints In Python}, for more information.
22655 @findex gdb.parameter
22656 @defun gdb.parameter (parameter)
22657 Return the value of a @value{GDBN} parameter. @var{parameter} is a
22658 string naming the parameter to look up; @var{parameter} may contain
22659 spaces if the parameter has a multi-part name. For example,
22660 @samp{print object} is a valid parameter name.
22662 If the named parameter does not exist, this function throws a
22663 @code{gdb.error} (@pxref{Exception Handling}). Otherwise, the
22664 parameter's value is converted to a Python value of the appropriate
22665 type, and returned.
22668 @findex gdb.history
22669 @defun gdb.history (number)
22670 Return a value from @value{GDBN}'s value history (@pxref{Value
22671 History}). @var{number} indicates which history element to return.
22672 If @var{number} is negative, then @value{GDBN} will take its absolute value
22673 and count backward from the last element (i.e., the most recent element) to
22674 find the value to return. If @var{number} is zero, then @value{GDBN} will
22675 return the most recent element. If the element specified by @var{number}
22676 doesn't exist in the value history, a @code{gdb.error} exception will be
22679 If no exception is raised, the return value is always an instance of
22680 @code{gdb.Value} (@pxref{Values From Inferior}).
22683 @findex gdb.parse_and_eval
22684 @defun gdb.parse_and_eval (expression)
22685 Parse @var{expression} as an expression in the current language,
22686 evaluate it, and return the result as a @code{gdb.Value}.
22687 @var{expression} must be a string.
22689 This function can be useful when implementing a new command
22690 (@pxref{Commands In Python}), as it provides a way to parse the
22691 command's argument as an expression. It is also useful simply to
22692 compute values, for example, it is the only way to get the value of a
22693 convenience variable (@pxref{Convenience Vars}) as a @code{gdb.Value}.
22696 @findex gdb.find_pc_line
22697 @defun gdb.find_pc_line (pc)
22698 Return the @code{gdb.Symtab_and_line} object corresponding to the
22699 @var{pc} value. @xref{Symbol Tables In Python}. If an invalid
22700 value of @var{pc} is passed as an argument, then the @code{symtab} and
22701 @code{line} attributes of the returned @code{gdb.Symtab_and_line} object
22702 will be @code{None} and 0 respectively.
22705 @findex gdb.post_event
22706 @defun gdb.post_event (event)
22707 Put @var{event}, a callable object taking no arguments, into
22708 @value{GDBN}'s internal event queue. This callable will be invoked at
22709 some later point, during @value{GDBN}'s event processing. Events
22710 posted using @code{post_event} will be run in the order in which they
22711 were posted; however, there is no way to know when they will be
22712 processed relative to other events inside @value{GDBN}.
22714 @value{GDBN} is not thread-safe. If your Python program uses multiple
22715 threads, you must be careful to only call @value{GDBN}-specific
22716 functions in the main @value{GDBN} thread. @code{post_event} ensures
22720 (@value{GDBP}) python
22724 > def __init__(self, message):
22725 > self.message = message;
22726 > def __call__(self):
22727 > gdb.write(self.message)
22729 >class MyThread1 (threading.Thread):
22731 > gdb.post_event(Writer("Hello "))
22733 >class MyThread2 (threading.Thread):
22735 > gdb.post_event(Writer("World\n"))
22737 >MyThread1().start()
22738 >MyThread2().start()
22740 (@value{GDBP}) Hello World
22745 @defun gdb.write (string @r{[}, stream{]})
22746 Print a string to @value{GDBN}'s paginated output stream. The
22747 optional @var{stream} determines the stream to print to. The default
22748 stream is @value{GDBN}'s standard output stream. Possible stream
22755 @value{GDBN}'s standard output stream.
22760 @value{GDBN}'s standard error stream.
22765 @value{GDBN}'s log stream (@pxref{Logging Output}).
22768 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
22769 call this function and will automatically direct the output to the
22774 @defun gdb.flush ()
22775 Flush the buffer of a @value{GDBN} paginated stream so that the
22776 contents are displayed immediately. @value{GDBN} will flush the
22777 contents of a stream automatically when it encounters a newline in the
22778 buffer. The optional @var{stream} determines the stream to flush. The
22779 default stream is @value{GDBN}'s standard output stream. Possible
22786 @value{GDBN}'s standard output stream.
22791 @value{GDBN}'s standard error stream.
22796 @value{GDBN}'s log stream (@pxref{Logging Output}).
22800 Flushing @code{sys.stdout} or @code{sys.stderr} will automatically
22801 call this function for the relevant stream.
22804 @findex gdb.target_charset
22805 @defun gdb.target_charset ()
22806 Return the name of the current target character set (@pxref{Character
22807 Sets}). This differs from @code{gdb.parameter('target-charset')} in
22808 that @samp{auto} is never returned.
22811 @findex gdb.target_wide_charset
22812 @defun gdb.target_wide_charset ()
22813 Return the name of the current target wide character set
22814 (@pxref{Character Sets}). This differs from
22815 @code{gdb.parameter('target-wide-charset')} in that @samp{auto} is
22819 @findex gdb.solib_name
22820 @defun gdb.solib_name (address)
22821 Return the name of the shared library holding the given @var{address}
22822 as a string, or @code{None}.
22825 @findex gdb.decode_line
22826 @defun gdb.decode_line @r{[}expression@r{]}
22827 Return locations of the line specified by @var{expression}, or of the
22828 current line if no argument was given. This function returns a Python
22829 tuple containing two elements. The first element contains a string
22830 holding any unparsed section of @var{expression} (or @code{None} if
22831 the expression has been fully parsed). The second element contains
22832 either @code{None} or another tuple that contains all the locations
22833 that match the expression represented as @code{gdb.Symtab_and_line}
22834 objects (@pxref{Symbol Tables In Python}). If @var{expression} is
22835 provided, it is decoded the way that @value{GDBN}'s inbuilt
22836 @code{break} or @code{edit} commands do (@pxref{Specify Location}).
22839 @defun gdb.prompt_hook (current_prompt)
22840 @anchor{prompt_hook}
22842 If @var{prompt_hook} is callable, @value{GDBN} will call the method
22843 assigned to this operation before a prompt is displayed by
22846 The parameter @code{current_prompt} contains the current @value{GDBN}
22847 prompt. This method must return a Python string, or @code{None}. If
22848 a string is returned, the @value{GDBN} prompt will be set to that
22849 string. If @code{None} is returned, @value{GDBN} will continue to use
22850 the current prompt.
22852 Some prompts cannot be substituted in @value{GDBN}. Secondary prompts
22853 such as those used by readline for command input, and annotation
22854 related prompts are prohibited from being changed.
22857 @node Exception Handling
22858 @subsubsection Exception Handling
22859 @cindex python exceptions
22860 @cindex exceptions, python
22862 When executing the @code{python} command, Python exceptions
22863 uncaught within the Python code are translated to calls to
22864 @value{GDBN} error-reporting mechanism. If the command that called
22865 @code{python} does not handle the error, @value{GDBN} will
22866 terminate it and print an error message containing the Python
22867 exception name, the associated value, and the Python call stack
22868 backtrace at the point where the exception was raised. Example:
22871 (@value{GDBP}) python print foo
22872 Traceback (most recent call last):
22873 File "<string>", line 1, in <module>
22874 NameError: name 'foo' is not defined
22877 @value{GDBN} errors that happen in @value{GDBN} commands invoked by
22878 Python code are converted to Python exceptions. The type of the
22879 Python exception depends on the error.
22883 This is the base class for most exceptions generated by @value{GDBN}.
22884 It is derived from @code{RuntimeError}, for compatibility with earlier
22885 versions of @value{GDBN}.
22887 If an error occurring in @value{GDBN} does not fit into some more
22888 specific category, then the generated exception will have this type.
22890 @item gdb.MemoryError
22891 This is a subclass of @code{gdb.error} which is thrown when an
22892 operation tried to access invalid memory in the inferior.
22894 @item KeyboardInterrupt
22895 User interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
22896 prompt) is translated to a Python @code{KeyboardInterrupt} exception.
22899 In all cases, your exception handler will see the @value{GDBN} error
22900 message as its value and the Python call stack backtrace at the Python
22901 statement closest to where the @value{GDBN} error occured as the
22904 @findex gdb.GdbError
22905 When implementing @value{GDBN} commands in Python via @code{gdb.Command},
22906 it is useful to be able to throw an exception that doesn't cause a
22907 traceback to be printed. For example, the user may have invoked the
22908 command incorrectly. Use the @code{gdb.GdbError} exception
22909 to handle this case. Example:
22913 >class HelloWorld (gdb.Command):
22914 > """Greet the whole world."""
22915 > def __init__ (self):
22916 > super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
22917 > def invoke (self, args, from_tty):
22918 > argv = gdb.string_to_argv (args)
22919 > if len (argv) != 0:
22920 > raise gdb.GdbError ("hello-world takes no arguments")
22921 > print "Hello, World!"
22924 (gdb) hello-world 42
22925 hello-world takes no arguments
22928 @node Values From Inferior
22929 @subsubsection Values From Inferior
22930 @cindex values from inferior, with Python
22931 @cindex python, working with values from inferior
22933 @cindex @code{gdb.Value}
22934 @value{GDBN} provides values it obtains from the inferior program in
22935 an object of type @code{gdb.Value}. @value{GDBN} uses this object
22936 for its internal bookkeeping of the inferior's values, and for
22937 fetching values when necessary.
22939 Inferior values that are simple scalars can be used directly in
22940 Python expressions that are valid for the value's data type. Here's
22941 an example for an integer or floating-point value @code{some_val}:
22948 As result of this, @code{bar} will also be a @code{gdb.Value} object
22949 whose values are of the same type as those of @code{some_val}.
22951 Inferior values that are structures or instances of some class can
22952 be accessed using the Python @dfn{dictionary syntax}. For example, if
22953 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
22954 can access its @code{foo} element with:
22957 bar = some_val['foo']
22960 Again, @code{bar} will also be a @code{gdb.Value} object.
22962 A @code{gdb.Value} that represents a function can be executed via
22963 inferior function call. Any arguments provided to the call must match
22964 the function's prototype, and must be provided in the order specified
22967 For example, @code{some_val} is a @code{gdb.Value} instance
22968 representing a function that takes two integers as arguments. To
22969 execute this function, call it like so:
22972 result = some_val (10,20)
22975 Any values returned from a function call will be stored as a
22978 The following attributes are provided:
22981 @defvar Value.address
22982 If this object is addressable, this read-only attribute holds a
22983 @code{gdb.Value} object representing the address. Otherwise,
22984 this attribute holds @code{None}.
22987 @cindex optimized out value in Python
22988 @defvar Value.is_optimized_out
22989 This read-only boolean attribute is true if the compiler optimized out
22990 this value, thus it is not available for fetching from the inferior.
22994 The type of this @code{gdb.Value}. The value of this attribute is a
22995 @code{gdb.Type} object (@pxref{Types In Python}).
22998 @defvar Value.dynamic_type
22999 The dynamic type of this @code{gdb.Value}. This uses C@t{++} run-time
23000 type information (@acronym{RTTI}) to determine the dynamic type of the
23001 value. If this value is of class type, it will return the class in
23002 which the value is embedded, if any. If this value is of pointer or
23003 reference to a class type, it will compute the dynamic type of the
23004 referenced object, and return a pointer or reference to that type,
23005 respectively. In all other cases, it will return the value's static
23008 Note that this feature will only work when debugging a C@t{++} program
23009 that includes @acronym{RTTI} for the object in question. Otherwise,
23010 it will just return the static type of the value as in @kbd{ptype foo}
23011 (@pxref{Symbols, ptype}).
23014 @defvar Value.is_lazy
23015 The value of this read-only boolean attribute is @code{True} if this
23016 @code{gdb.Value} has not yet been fetched from the inferior.
23017 @value{GDBN} does not fetch values until necessary, for efficiency.
23021 myval = gdb.parse_and_eval ('somevar')
23024 The value of @code{somevar} is not fetched at this time. It will be
23025 fetched when the value is needed, or when the @code{fetch_lazy}
23030 The following methods are provided:
23033 @defun Value.__init__ (@var{val})
23034 Many Python values can be converted directly to a @code{gdb.Value} via
23035 this object initializer. Specifically:
23038 @item Python boolean
23039 A Python boolean is converted to the boolean type from the current
23042 @item Python integer
23043 A Python integer is converted to the C @code{long} type for the
23044 current architecture.
23047 A Python long is converted to the C @code{long long} type for the
23048 current architecture.
23051 A Python float is converted to the C @code{double} type for the
23052 current architecture.
23054 @item Python string
23055 A Python string is converted to a target string, using the current
23058 @item @code{gdb.Value}
23059 If @code{val} is a @code{gdb.Value}, then a copy of the value is made.
23061 @item @code{gdb.LazyString}
23062 If @code{val} is a @code{gdb.LazyString} (@pxref{Lazy Strings In
23063 Python}), then the lazy string's @code{value} method is called, and
23064 its result is used.
23068 @defun Value.cast (type)
23069 Return a new instance of @code{gdb.Value} that is the result of
23070 casting this instance to the type described by @var{type}, which must
23071 be a @code{gdb.Type} object. If the cast cannot be performed for some
23072 reason, this method throws an exception.
23075 @defun Value.dereference ()
23076 For pointer data types, this method returns a new @code{gdb.Value} object
23077 whose contents is the object pointed to by the pointer. For example, if
23078 @code{foo} is a C pointer to an @code{int}, declared in your C program as
23085 then you can use the corresponding @code{gdb.Value} to access what
23086 @code{foo} points to like this:
23089 bar = foo.dereference ()
23092 The result @code{bar} will be a @code{gdb.Value} object holding the
23093 value pointed to by @code{foo}.
23095 A similar function @code{Value.referenced_value} exists which also
23096 returns @code{gdb.Value} objects corresonding to the values pointed to
23097 by pointer values (and additionally, values referenced by reference
23098 values). However, the behavior of @code{Value.dereference}
23099 differs from @code{Value.referenced_value} by the fact that the
23100 behavior of @code{Value.dereference} is identical to applying the C
23101 unary operator @code{*} on a given value. For example, consider a
23102 reference to a pointer @code{ptrref}, declared in your C@t{++} program
23106 typedef int *intptr;
23110 intptr &ptrref = ptr;
23113 Though @code{ptrref} is a reference value, one can apply the method
23114 @code{Value.dereference} to the @code{gdb.Value} object corresponding
23115 to it and obtain a @code{gdb.Value} which is identical to that
23116 corresponding to @code{val}. However, if you apply the method
23117 @code{Value.referenced_value}, the result would be a @code{gdb.Value}
23118 object identical to that corresponding to @code{ptr}.
23121 py_ptrref = gdb.parse_and_eval ("ptrref")
23122 py_val = py_ptrref.dereference ()
23123 py_ptr = py_ptrref.referenced_value ()
23126 The @code{gdb.Value} object @code{py_val} is identical to that
23127 corresponding to @code{val}, and @code{py_ptr} is identical to that
23128 corresponding to @code{ptr}. In general, @code{Value.dereference} can
23129 be applied whenever the C unary operator @code{*} can be applied
23130 to the corresponding C value. For those cases where applying both
23131 @code{Value.dereference} and @code{Value.referenced_value} is allowed,
23132 the results obtained need not be identical (as we have seen in the above
23133 example). The results are however identical when applied on
23134 @code{gdb.Value} objects corresponding to pointers (@code{gdb.Value}
23135 objects with type code @code{TYPE_CODE_PTR}) in a C/C@t{++} program.
23138 @defun Value.referenced_value ()
23139 For pointer or reference data types, this method returns a new
23140 @code{gdb.Value} object corresponding to the value referenced by the
23141 pointer/reference value. For pointer data types,
23142 @code{Value.dereference} and @code{Value.referenced_value} produce
23143 identical results. The difference between these methods is that
23144 @code{Value.dereference} cannot get the values referenced by reference
23145 values. For example, consider a reference to an @code{int}, declared
23146 in your C@t{++} program as
23154 then applying @code{Value.dereference} to the @code{gdb.Value} object
23155 corresponding to @code{ref} will result in an error, while applying
23156 @code{Value.referenced_value} will result in a @code{gdb.Value} object
23157 identical to that corresponding to @code{val}.
23160 py_ref = gdb.parse_and_eval ("ref")
23161 er_ref = py_ref.dereference () # Results in error
23162 py_val = py_ref.referenced_value () # Returns the referenced value
23165 The @code{gdb.Value} object @code{py_val} is identical to that
23166 corresponding to @code{val}.
23169 @defun Value.dynamic_cast (type)
23170 Like @code{Value.cast}, but works as if the C@t{++} @code{dynamic_cast}
23171 operator were used. Consult a C@t{++} reference for details.
23174 @defun Value.reinterpret_cast (type)
23175 Like @code{Value.cast}, but works as if the C@t{++} @code{reinterpret_cast}
23176 operator were used. Consult a C@t{++} reference for details.
23179 @defun Value.string (@r{[}encoding@r{[}, errors@r{[}, length@r{]]]})
23180 If this @code{gdb.Value} represents a string, then this method
23181 converts the contents to a Python string. Otherwise, this method will
23182 throw an exception.
23184 Strings are recognized in a language-specific way; whether a given
23185 @code{gdb.Value} represents a string is determined by the current
23188 For C-like languages, a value is a string if it is a pointer to or an
23189 array of characters or ints. The string is assumed to be terminated
23190 by a zero of the appropriate width. However if the optional length
23191 argument is given, the string will be converted to that given length,
23192 ignoring any embedded zeros that the string may contain.
23194 If the optional @var{encoding} argument is given, it must be a string
23195 naming the encoding of the string in the @code{gdb.Value}, such as
23196 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
23197 the same encodings as the corresponding argument to Python's
23198 @code{string.decode} method, and the Python codec machinery will be used
23199 to convert the string. If @var{encoding} is not given, or if
23200 @var{encoding} is the empty string, then either the @code{target-charset}
23201 (@pxref{Character Sets}) will be used, or a language-specific encoding
23202 will be used, if the current language is able to supply one.
23204 The optional @var{errors} argument is the same as the corresponding
23205 argument to Python's @code{string.decode} method.
23207 If the optional @var{length} argument is given, the string will be
23208 fetched and converted to the given length.
23211 @defun Value.lazy_string (@r{[}encoding @r{[}, length@r{]]})
23212 If this @code{gdb.Value} represents a string, then this method
23213 converts the contents to a @code{gdb.LazyString} (@pxref{Lazy Strings
23214 In Python}). Otherwise, this method will throw an exception.
23216 If the optional @var{encoding} argument is given, it must be a string
23217 naming the encoding of the @code{gdb.LazyString}. Some examples are:
23218 @samp{ascii}, @samp{iso-8859-6} or @samp{utf-8}. If the
23219 @var{encoding} argument is an encoding that @value{GDBN} does
23220 recognize, @value{GDBN} will raise an error.
23222 When a lazy string is printed, the @value{GDBN} encoding machinery is
23223 used to convert the string during printing. If the optional
23224 @var{encoding} argument is not provided, or is an empty string,
23225 @value{GDBN} will automatically select the encoding most suitable for
23226 the string type. For further information on encoding in @value{GDBN}
23227 please see @ref{Character Sets}.
23229 If the optional @var{length} argument is given, the string will be
23230 fetched and encoded to the length of characters specified. If
23231 the @var{length} argument is not provided, the string will be fetched
23232 and encoded until a null of appropriate width is found.
23235 @defun Value.fetch_lazy ()
23236 If the @code{gdb.Value} object is currently a lazy value
23237 (@code{gdb.Value.is_lazy} is @code{True}), then the value is
23238 fetched from the inferior. Any errors that occur in the process
23239 will produce a Python exception.
23241 If the @code{gdb.Value} object is not a lazy value, this method
23244 This method does not return a value.
23249 @node Types In Python
23250 @subsubsection Types In Python
23251 @cindex types in Python
23252 @cindex Python, working with types
23255 @value{GDBN} represents types from the inferior using the class
23258 The following type-related functions are available in the @code{gdb}
23261 @findex gdb.lookup_type
23262 @defun gdb.lookup_type (name @r{[}, block@r{]})
23263 This function looks up a type by name. @var{name} is the name of the
23264 type to look up. It must be a string.
23266 If @var{block} is given, then @var{name} is looked up in that scope.
23267 Otherwise, it is searched for globally.
23269 Ordinarily, this function will return an instance of @code{gdb.Type}.
23270 If the named type cannot be found, it will throw an exception.
23273 If the type is a structure or class type, or an enum type, the fields
23274 of that type can be accessed using the Python @dfn{dictionary syntax}.
23275 For example, if @code{some_type} is a @code{gdb.Type} instance holding
23276 a structure type, you can access its @code{foo} field with:
23279 bar = some_type['foo']
23282 @code{bar} will be a @code{gdb.Field} object; see below under the
23283 description of the @code{Type.fields} method for a description of the
23284 @code{gdb.Field} class.
23286 An instance of @code{Type} has the following attributes:
23290 The type code for this type. The type code will be one of the
23291 @code{TYPE_CODE_} constants defined below.
23294 @defvar Type.sizeof
23295 The size of this type, in target @code{char} units. Usually, a
23296 target's @code{char} type will be an 8-bit byte. However, on some
23297 unusual platforms, this type may have a different size.
23301 The tag name for this type. The tag name is the name after
23302 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
23303 languages have this concept. If this type has no tag name, then
23304 @code{None} is returned.
23308 The following methods are provided:
23311 @defun Type.fields ()
23312 For structure and union types, this method returns the fields. Range
23313 types have two fields, the minimum and maximum values. Enum types
23314 have one field per enum constant. Function and method types have one
23315 field per parameter. The base types of C@t{++} classes are also
23316 represented as fields. If the type has no fields, or does not fit
23317 into one of these categories, an empty sequence will be returned.
23319 Each field is a @code{gdb.Field} object, with some pre-defined attributes:
23322 This attribute is not available for @code{static} fields (as in
23323 C@t{++} or Java). For non-@code{static} fields, the value is the bit
23324 position of the field. For @code{enum} fields, the value is the
23325 enumeration member's integer representation.
23328 The name of the field, or @code{None} for anonymous fields.
23331 This is @code{True} if the field is artificial, usually meaning that
23332 it was provided by the compiler and not the user. This attribute is
23333 always provided, and is @code{False} if the field is not artificial.
23335 @item is_base_class
23336 This is @code{True} if the field represents a base class of a C@t{++}
23337 structure. This attribute is always provided, and is @code{False}
23338 if the field is not a base class of the type that is the argument of
23339 @code{fields}, or if that type was not a C@t{++} class.
23342 If the field is packed, or is a bitfield, then this will have a
23343 non-zero value, which is the size of the field in bits. Otherwise,
23344 this will be zero; in this case the field's size is given by its type.
23347 The type of the field. This is usually an instance of @code{Type},
23348 but it can be @code{None} in some situations.
23352 @defun Type.array (@var{n1} @r{[}, @var{n2}@r{]})
23353 Return a new @code{gdb.Type} object which represents an array of this
23354 type. If one argument is given, it is the inclusive upper bound of
23355 the array; in this case the lower bound is zero. If two arguments are
23356 given, the first argument is the lower bound of the array, and the
23357 second argument is the upper bound of the array. An array's length
23358 must not be negative, but the bounds can be.
23361 @defun Type.const ()
23362 Return a new @code{gdb.Type} object which represents a
23363 @code{const}-qualified variant of this type.
23366 @defun Type.volatile ()
23367 Return a new @code{gdb.Type} object which represents a
23368 @code{volatile}-qualified variant of this type.
23371 @defun Type.unqualified ()
23372 Return a new @code{gdb.Type} object which represents an unqualified
23373 variant of this type. That is, the result is neither @code{const} nor
23377 @defun Type.range ()
23378 Return a Python @code{Tuple} object that contains two elements: the
23379 low bound of the argument type and the high bound of that type. If
23380 the type does not have a range, @value{GDBN} will raise a
23381 @code{gdb.error} exception (@pxref{Exception Handling}).
23384 @defun Type.reference ()
23385 Return a new @code{gdb.Type} object which represents a reference to this
23389 @defun Type.pointer ()
23390 Return a new @code{gdb.Type} object which represents a pointer to this
23394 @defun Type.strip_typedefs ()
23395 Return a new @code{gdb.Type} that represents the real type,
23396 after removing all layers of typedefs.
23399 @defun Type.target ()
23400 Return a new @code{gdb.Type} object which represents the target type
23403 For a pointer type, the target type is the type of the pointed-to
23404 object. For an array type (meaning C-like arrays), the target type is
23405 the type of the elements of the array. For a function or method type,
23406 the target type is the type of the return value. For a complex type,
23407 the target type is the type of the elements. For a typedef, the
23408 target type is the aliased type.
23410 If the type does not have a target, this method will throw an
23414 @defun Type.template_argument (n @r{[}, block@r{]})
23415 If this @code{gdb.Type} is an instantiation of a template, this will
23416 return a new @code{gdb.Type} which represents the type of the
23417 @var{n}th template argument.
23419 If this @code{gdb.Type} is not a template type, this will throw an
23420 exception. Ordinarily, only C@t{++} code will have template types.
23422 If @var{block} is given, then @var{name} is looked up in that scope.
23423 Otherwise, it is searched for globally.
23428 Each type has a code, which indicates what category this type falls
23429 into. The available type categories are represented by constants
23430 defined in the @code{gdb} module:
23433 @findex TYPE_CODE_PTR
23434 @findex gdb.TYPE_CODE_PTR
23435 @item gdb.TYPE_CODE_PTR
23436 The type is a pointer.
23438 @findex TYPE_CODE_ARRAY
23439 @findex gdb.TYPE_CODE_ARRAY
23440 @item gdb.TYPE_CODE_ARRAY
23441 The type is an array.
23443 @findex TYPE_CODE_STRUCT
23444 @findex gdb.TYPE_CODE_STRUCT
23445 @item gdb.TYPE_CODE_STRUCT
23446 The type is a structure.
23448 @findex TYPE_CODE_UNION
23449 @findex gdb.TYPE_CODE_UNION
23450 @item gdb.TYPE_CODE_UNION
23451 The type is a union.
23453 @findex TYPE_CODE_ENUM
23454 @findex gdb.TYPE_CODE_ENUM
23455 @item gdb.TYPE_CODE_ENUM
23456 The type is an enum.
23458 @findex TYPE_CODE_FLAGS
23459 @findex gdb.TYPE_CODE_FLAGS
23460 @item gdb.TYPE_CODE_FLAGS
23461 A bit flags type, used for things such as status registers.
23463 @findex TYPE_CODE_FUNC
23464 @findex gdb.TYPE_CODE_FUNC
23465 @item gdb.TYPE_CODE_FUNC
23466 The type is a function.
23468 @findex TYPE_CODE_INT
23469 @findex gdb.TYPE_CODE_INT
23470 @item gdb.TYPE_CODE_INT
23471 The type is an integer type.
23473 @findex TYPE_CODE_FLT
23474 @findex gdb.TYPE_CODE_FLT
23475 @item gdb.TYPE_CODE_FLT
23476 A floating point type.
23478 @findex TYPE_CODE_VOID
23479 @findex gdb.TYPE_CODE_VOID
23480 @item gdb.TYPE_CODE_VOID
23481 The special type @code{void}.
23483 @findex TYPE_CODE_SET
23484 @findex gdb.TYPE_CODE_SET
23485 @item gdb.TYPE_CODE_SET
23488 @findex TYPE_CODE_RANGE
23489 @findex gdb.TYPE_CODE_RANGE
23490 @item gdb.TYPE_CODE_RANGE
23491 A range type, that is, an integer type with bounds.
23493 @findex TYPE_CODE_STRING
23494 @findex gdb.TYPE_CODE_STRING
23495 @item gdb.TYPE_CODE_STRING
23496 A string type. Note that this is only used for certain languages with
23497 language-defined string types; C strings are not represented this way.
23499 @findex TYPE_CODE_BITSTRING
23500 @findex gdb.TYPE_CODE_BITSTRING
23501 @item gdb.TYPE_CODE_BITSTRING
23504 @findex TYPE_CODE_ERROR
23505 @findex gdb.TYPE_CODE_ERROR
23506 @item gdb.TYPE_CODE_ERROR
23507 An unknown or erroneous type.
23509 @findex TYPE_CODE_METHOD
23510 @findex gdb.TYPE_CODE_METHOD
23511 @item gdb.TYPE_CODE_METHOD
23512 A method type, as found in C@t{++} or Java.
23514 @findex TYPE_CODE_METHODPTR
23515 @findex gdb.TYPE_CODE_METHODPTR
23516 @item gdb.TYPE_CODE_METHODPTR
23517 A pointer-to-member-function.
23519 @findex TYPE_CODE_MEMBERPTR
23520 @findex gdb.TYPE_CODE_MEMBERPTR
23521 @item gdb.TYPE_CODE_MEMBERPTR
23522 A pointer-to-member.
23524 @findex TYPE_CODE_REF
23525 @findex gdb.TYPE_CODE_REF
23526 @item gdb.TYPE_CODE_REF
23529 @findex TYPE_CODE_CHAR
23530 @findex gdb.TYPE_CODE_CHAR
23531 @item gdb.TYPE_CODE_CHAR
23534 @findex TYPE_CODE_BOOL
23535 @findex gdb.TYPE_CODE_BOOL
23536 @item gdb.TYPE_CODE_BOOL
23539 @findex TYPE_CODE_COMPLEX
23540 @findex gdb.TYPE_CODE_COMPLEX
23541 @item gdb.TYPE_CODE_COMPLEX
23542 A complex float type.
23544 @findex TYPE_CODE_TYPEDEF
23545 @findex gdb.TYPE_CODE_TYPEDEF
23546 @item gdb.TYPE_CODE_TYPEDEF
23547 A typedef to some other type.
23549 @findex TYPE_CODE_NAMESPACE
23550 @findex gdb.TYPE_CODE_NAMESPACE
23551 @item gdb.TYPE_CODE_NAMESPACE
23552 A C@t{++} namespace.
23554 @findex TYPE_CODE_DECFLOAT
23555 @findex gdb.TYPE_CODE_DECFLOAT
23556 @item gdb.TYPE_CODE_DECFLOAT
23557 A decimal floating point type.
23559 @findex TYPE_CODE_INTERNAL_FUNCTION
23560 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
23561 @item gdb.TYPE_CODE_INTERNAL_FUNCTION
23562 A function internal to @value{GDBN}. This is the type used to represent
23563 convenience functions.
23566 Further support for types is provided in the @code{gdb.types}
23567 Python module (@pxref{gdb.types}).
23569 @node Pretty Printing API
23570 @subsubsection Pretty Printing API
23572 An example output is provided (@pxref{Pretty Printing}).
23574 A pretty-printer is just an object that holds a value and implements a
23575 specific interface, defined here.
23577 @defun pretty_printer.children (self)
23578 @value{GDBN} will call this method on a pretty-printer to compute the
23579 children of the pretty-printer's value.
23581 This method must return an object conforming to the Python iterator
23582 protocol. Each item returned by the iterator must be a tuple holding
23583 two elements. The first element is the ``name'' of the child; the
23584 second element is the child's value. The value can be any Python
23585 object which is convertible to a @value{GDBN} value.
23587 This method is optional. If it does not exist, @value{GDBN} will act
23588 as though the value has no children.
23591 @defun pretty_printer.display_hint (self)
23592 The CLI may call this method and use its result to change the
23593 formatting of a value. The result will also be supplied to an MI
23594 consumer as a @samp{displayhint} attribute of the variable being
23597 This method is optional. If it does exist, this method must return a
23600 Some display hints are predefined by @value{GDBN}:
23604 Indicate that the object being printed is ``array-like''. The CLI
23605 uses this to respect parameters such as @code{set print elements} and
23606 @code{set print array}.
23609 Indicate that the object being printed is ``map-like'', and that the
23610 children of this value can be assumed to alternate between keys and
23614 Indicate that the object being printed is ``string-like''. If the
23615 printer's @code{to_string} method returns a Python string of some
23616 kind, then @value{GDBN} will call its internal language-specific
23617 string-printing function to format the string. For the CLI this means
23618 adding quotation marks, possibly escaping some characters, respecting
23619 @code{set print elements}, and the like.
23623 @defun pretty_printer.to_string (self)
23624 @value{GDBN} will call this method to display the string
23625 representation of the value passed to the object's constructor.
23627 When printing from the CLI, if the @code{to_string} method exists,
23628 then @value{GDBN} will prepend its result to the values returned by
23629 @code{children}. Exactly how this formatting is done is dependent on
23630 the display hint, and may change as more hints are added. Also,
23631 depending on the print settings (@pxref{Print Settings}), the CLI may
23632 print just the result of @code{to_string} in a stack trace, omitting
23633 the result of @code{children}.
23635 If this method returns a string, it is printed verbatim.
23637 Otherwise, if this method returns an instance of @code{gdb.Value},
23638 then @value{GDBN} prints this value. This may result in a call to
23639 another pretty-printer.
23641 If instead the method returns a Python value which is convertible to a
23642 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
23643 the resulting value. Again, this may result in a call to another
23644 pretty-printer. Python scalars (integers, floats, and booleans) and
23645 strings are convertible to @code{gdb.Value}; other types are not.
23647 Finally, if this method returns @code{None} then no further operations
23648 are peformed in this method and nothing is printed.
23650 If the result is not one of these types, an exception is raised.
23653 @value{GDBN} provides a function which can be used to look up the
23654 default pretty-printer for a @code{gdb.Value}:
23656 @findex gdb.default_visualizer
23657 @defun gdb.default_visualizer (value)
23658 This function takes a @code{gdb.Value} object as an argument. If a
23659 pretty-printer for this value exists, then it is returned. If no such
23660 printer exists, then this returns @code{None}.
23663 @node Selecting Pretty-Printers
23664 @subsubsection Selecting Pretty-Printers
23666 The Python list @code{gdb.pretty_printers} contains an array of
23667 functions or callable objects that have been registered via addition
23668 as a pretty-printer. Printers in this list are called @code{global}
23669 printers, they're available when debugging all inferiors.
23670 Each @code{gdb.Progspace} contains a @code{pretty_printers} attribute.
23671 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
23674 Each function on these lists is passed a single @code{gdb.Value}
23675 argument and should return a pretty-printer object conforming to the
23676 interface definition above (@pxref{Pretty Printing API}). If a function
23677 cannot create a pretty-printer for the value, it should return
23680 @value{GDBN} first checks the @code{pretty_printers} attribute of each
23681 @code{gdb.Objfile} in the current program space and iteratively calls
23682 each enabled lookup routine in the list for that @code{gdb.Objfile}
23683 until it receives a pretty-printer object.
23684 If no pretty-printer is found in the objfile lists, @value{GDBN} then
23685 searches the pretty-printer list of the current program space,
23686 calling each enabled function until an object is returned.
23687 After these lists have been exhausted, it tries the global
23688 @code{gdb.pretty_printers} list, again calling each enabled function until an
23689 object is returned.
23691 The order in which the objfiles are searched is not specified. For a
23692 given list, functions are always invoked from the head of the list,
23693 and iterated over sequentially until the end of the list, or a printer
23694 object is returned.
23696 For various reasons a pretty-printer may not work.
23697 For example, the underlying data structure may have changed and
23698 the pretty-printer is out of date.
23700 The consequences of a broken pretty-printer are severe enough that
23701 @value{GDBN} provides support for enabling and disabling individual
23702 printers. For example, if @code{print frame-arguments} is on,
23703 a backtrace can become highly illegible if any argument is printed
23704 with a broken printer.
23706 Pretty-printers are enabled and disabled by attaching an @code{enabled}
23707 attribute to the registered function or callable object. If this attribute
23708 is present and its value is @code{False}, the printer is disabled, otherwise
23709 the printer is enabled.
23711 @node Writing a Pretty-Printer
23712 @subsubsection Writing a Pretty-Printer
23713 @cindex writing a pretty-printer
23715 A pretty-printer consists of two parts: a lookup function to detect
23716 if the type is supported, and the printer itself.
23718 Here is an example showing how a @code{std::string} printer might be
23719 written. @xref{Pretty Printing API}, for details on the API this class
23723 class StdStringPrinter(object):
23724 "Print a std::string"
23726 def __init__(self, val):
23729 def to_string(self):
23730 return self.val['_M_dataplus']['_M_p']
23732 def display_hint(self):
23736 And here is an example showing how a lookup function for the printer
23737 example above might be written.
23740 def str_lookup_function(val):
23741 lookup_tag = val.type.tag
23742 if lookup_tag == None:
23744 regex = re.compile("^std::basic_string<char,.*>$")
23745 if regex.match(lookup_tag):
23746 return StdStringPrinter(val)
23750 The example lookup function extracts the value's type, and attempts to
23751 match it to a type that it can pretty-print. If it is a type the
23752 printer can pretty-print, it will return a printer object. If not, it
23753 returns @code{None}.
23755 We recommend that you put your core pretty-printers into a Python
23756 package. If your pretty-printers are for use with a library, we
23757 further recommend embedding a version number into the package name.
23758 This practice will enable @value{GDBN} to load multiple versions of
23759 your pretty-printers at the same time, because they will have
23762 You should write auto-loaded code (@pxref{Python Auto-loading}) such that it
23763 can be evaluated multiple times without changing its meaning. An
23764 ideal auto-load file will consist solely of @code{import}s of your
23765 printer modules, followed by a call to a register pretty-printers with
23766 the current objfile.
23768 Taken as a whole, this approach will scale nicely to multiple
23769 inferiors, each potentially using a different library version.
23770 Embedding a version number in the Python package name will ensure that
23771 @value{GDBN} is able to load both sets of printers simultaneously.
23772 Then, because the search for pretty-printers is done by objfile, and
23773 because your auto-loaded code took care to register your library's
23774 printers with a specific objfile, @value{GDBN} will find the correct
23775 printers for the specific version of the library used by each
23778 To continue the @code{std::string} example (@pxref{Pretty Printing API}),
23779 this code might appear in @code{gdb.libstdcxx.v6}:
23782 def register_printers(objfile):
23783 objfile.pretty_printers.append(str_lookup_function)
23787 And then the corresponding contents of the auto-load file would be:
23790 import gdb.libstdcxx.v6
23791 gdb.libstdcxx.v6.register_printers(gdb.current_objfile())
23794 The previous example illustrates a basic pretty-printer.
23795 There are a few things that can be improved on.
23796 The printer doesn't have a name, making it hard to identify in a
23797 list of installed printers. The lookup function has a name, but
23798 lookup functions can have arbitrary, even identical, names.
23800 Second, the printer only handles one type, whereas a library typically has
23801 several types. One could install a lookup function for each desired type
23802 in the library, but one could also have a single lookup function recognize
23803 several types. The latter is the conventional way this is handled.
23804 If a pretty-printer can handle multiple data types, then its
23805 @dfn{subprinters} are the printers for the individual data types.
23807 The @code{gdb.printing} module provides a formal way of solving these
23808 problems (@pxref{gdb.printing}).
23809 Here is another example that handles multiple types.
23811 These are the types we are going to pretty-print:
23814 struct foo @{ int a, b; @};
23815 struct bar @{ struct foo x, y; @};
23818 Here are the printers:
23822 """Print a foo object."""
23824 def __init__(self, val):
23827 def to_string(self):
23828 return ("a=<" + str(self.val["a"]) +
23829 "> b=<" + str(self.val["b"]) + ">")
23832 """Print a bar object."""
23834 def __init__(self, val):
23837 def to_string(self):
23838 return ("x=<" + str(self.val["x"]) +
23839 "> y=<" + str(self.val["y"]) + ">")
23842 This example doesn't need a lookup function, that is handled by the
23843 @code{gdb.printing} module. Instead a function is provided to build up
23844 the object that handles the lookup.
23847 import gdb.printing
23849 def build_pretty_printer():
23850 pp = gdb.printing.RegexpCollectionPrettyPrinter(
23852 pp.add_printer('foo', '^foo$', fooPrinter)
23853 pp.add_printer('bar', '^bar$', barPrinter)
23857 And here is the autoload support:
23860 import gdb.printing
23862 gdb.printing.register_pretty_printer(
23863 gdb.current_objfile(),
23864 my_library.build_pretty_printer())
23867 Finally, when this printer is loaded into @value{GDBN}, here is the
23868 corresponding output of @samp{info pretty-printer}:
23871 (gdb) info pretty-printer
23878 @node Inferiors In Python
23879 @subsubsection Inferiors In Python
23880 @cindex inferiors in Python
23882 @findex gdb.Inferior
23883 Programs which are being run under @value{GDBN} are called inferiors
23884 (@pxref{Inferiors and Programs}). Python scripts can access
23885 information about and manipulate inferiors controlled by @value{GDBN}
23886 via objects of the @code{gdb.Inferior} class.
23888 The following inferior-related functions are available in the @code{gdb}
23891 @defun gdb.inferiors ()
23892 Return a tuple containing all inferior objects.
23895 @defun gdb.selected_inferior ()
23896 Return an object representing the current inferior.
23899 A @code{gdb.Inferior} object has the following attributes:
23902 @defvar Inferior.num
23903 ID of inferior, as assigned by GDB.
23906 @defvar Inferior.pid
23907 Process ID of the inferior, as assigned by the underlying operating
23911 @defvar Inferior.was_attached
23912 Boolean signaling whether the inferior was created using `attach', or
23913 started by @value{GDBN} itself.
23917 A @code{gdb.Inferior} object has the following methods:
23920 @defun Inferior.is_valid ()
23921 Returns @code{True} if the @code{gdb.Inferior} object is valid,
23922 @code{False} if not. A @code{gdb.Inferior} object will become invalid
23923 if the inferior no longer exists within @value{GDBN}. All other
23924 @code{gdb.Inferior} methods will throw an exception if it is invalid
23925 at the time the method is called.
23928 @defun Inferior.threads ()
23929 This method returns a tuple holding all the threads which are valid
23930 when it is called. If there are no valid threads, the method will
23931 return an empty tuple.
23934 @findex Inferior.read_memory
23935 @defun Inferior.read_memory (address, length)
23936 Read @var{length} bytes of memory from the inferior, starting at
23937 @var{address}. Returns a buffer object, which behaves much like an array
23938 or a string. It can be modified and given to the
23939 @code{Inferior.write_memory} function.
23942 @findex Inferior.write_memory
23943 @defun Inferior.write_memory (address, buffer @r{[}, length@r{]})
23944 Write the contents of @var{buffer} to the inferior, starting at
23945 @var{address}. The @var{buffer} parameter must be a Python object
23946 which supports the buffer protocol, i.e., a string, an array or the
23947 object returned from @code{Inferior.read_memory}. If given, @var{length}
23948 determines the number of bytes from @var{buffer} to be written.
23951 @findex gdb.search_memory
23952 @defun Inferior.search_memory (address, length, pattern)
23953 Search a region of the inferior memory starting at @var{address} with
23954 the given @var{length} using the search pattern supplied in
23955 @var{pattern}. The @var{pattern} parameter must be a Python object
23956 which supports the buffer protocol, i.e., a string, an array or the
23957 object returned from @code{gdb.read_memory}. Returns a Python @code{Long}
23958 containing the address where the pattern was found, or @code{None} if
23959 the pattern could not be found.
23963 @node Events In Python
23964 @subsubsection Events In Python
23965 @cindex inferior events in Python
23967 @value{GDBN} provides a general event facility so that Python code can be
23968 notified of various state changes, particularly changes that occur in
23971 An @dfn{event} is just an object that describes some state change. The
23972 type of the object and its attributes will vary depending on the details
23973 of the change. All the existing events are described below.
23975 In order to be notified of an event, you must register an event handler
23976 with an @dfn{event registry}. An event registry is an object in the
23977 @code{gdb.events} module which dispatches particular events. A registry
23978 provides methods to register and unregister event handlers:
23981 @defun EventRegistry.connect (object)
23982 Add the given callable @var{object} to the registry. This object will be
23983 called when an event corresponding to this registry occurs.
23986 @defun EventRegistry.disconnect (object)
23987 Remove the given @var{object} from the registry. Once removed, the object
23988 will no longer receive notifications of events.
23992 Here is an example:
23995 def exit_handler (event):
23996 print "event type: exit"
23997 print "exit code: %d" % (event.exit_code)
23999 gdb.events.exited.connect (exit_handler)
24002 In the above example we connect our handler @code{exit_handler} to the
24003 registry @code{events.exited}. Once connected, @code{exit_handler} gets
24004 called when the inferior exits. The argument @dfn{event} in this example is
24005 of type @code{gdb.ExitedEvent}. As you can see in the example the
24006 @code{ExitedEvent} object has an attribute which indicates the exit code of
24009 The following is a listing of the event registries that are available and
24010 details of the events they emit:
24015 Emits @code{gdb.ThreadEvent}.
24017 Some events can be thread specific when @value{GDBN} is running in non-stop
24018 mode. When represented in Python, these events all extend
24019 @code{gdb.ThreadEvent}. Note, this event is not emitted directly; instead,
24020 events which are emitted by this or other modules might extend this event.
24021 Examples of these events are @code{gdb.BreakpointEvent} and
24022 @code{gdb.ContinueEvent}.
24025 @defvar ThreadEvent.inferior_thread
24026 In non-stop mode this attribute will be set to the specific thread which was
24027 involved in the emitted event. Otherwise, it will be set to @code{None}.
24031 Emits @code{gdb.ContinueEvent} which extends @code{gdb.ThreadEvent}.
24033 This event indicates that the inferior has been continued after a stop. For
24034 inherited attribute refer to @code{gdb.ThreadEvent} above.
24036 @item events.exited
24037 Emits @code{events.ExitedEvent} which indicates that the inferior has exited.
24038 @code{events.ExitedEvent} has two attributes:
24040 @defvar ExitedEvent.exit_code
24041 An integer representing the exit code, if available, which the inferior
24042 has returned. (The exit code could be unavailable if, for example,
24043 @value{GDBN} detaches from the inferior.) If the exit code is unavailable,
24044 the attribute does not exist.
24046 @defvar ExitedEvent inferior
24047 A reference to the inferior which triggered the @code{exited} event.
24052 Emits @code{gdb.StopEvent} which extends @code{gdb.ThreadEvent}.
24054 Indicates that the inferior has stopped. All events emitted by this registry
24055 extend StopEvent. As a child of @code{gdb.ThreadEvent}, @code{gdb.StopEvent}
24056 will indicate the stopped thread when @value{GDBN} is running in non-stop
24057 mode. Refer to @code{gdb.ThreadEvent} above for more details.
24059 Emits @code{gdb.SignalEvent} which extends @code{gdb.StopEvent}.
24061 This event indicates that the inferior or one of its threads has received as
24062 signal. @code{gdb.SignalEvent} has the following attributes:
24065 @defvar SignalEvent.stop_signal
24066 A string representing the signal received by the inferior. A list of possible
24067 signal values can be obtained by running the command @code{info signals} in
24068 the @value{GDBN} command prompt.
24072 Also emits @code{gdb.BreakpointEvent} which extends @code{gdb.StopEvent}.
24074 @code{gdb.BreakpointEvent} event indicates that one or more breakpoints have
24075 been hit, and has the following attributes:
24078 @defvar BreakpointEvent.breakpoints
24079 A sequence containing references to all the breakpoints (type
24080 @code{gdb.Breakpoint}) that were hit.
24081 @xref{Breakpoints In Python}, for details of the @code{gdb.Breakpoint} object.
24083 @defvar BreakpointEvent.breakpoint
24084 A reference to the first breakpoint that was hit.
24085 This function is maintained for backward compatibility and is now deprecated
24086 in favor of the @code{gdb.BreakpointEvent.breakpoints} attribute.
24090 @item events.new_objfile
24091 Emits @code{gdb.NewObjFileEvent} which indicates that a new object file has
24092 been loaded by @value{GDBN}. @code{gdb.NewObjFileEvent} has one attribute:
24095 @defvar NewObjFileEvent.new_objfile
24096 A reference to the object file (@code{gdb.Objfile}) which has been loaded.
24097 @xref{Objfiles In Python}, for details of the @code{gdb.Objfile} object.
24103 @node Threads In Python
24104 @subsubsection Threads In Python
24105 @cindex threads in python
24107 @findex gdb.InferiorThread
24108 Python scripts can access information about, and manipulate inferior threads
24109 controlled by @value{GDBN}, via objects of the @code{gdb.InferiorThread} class.
24111 The following thread-related functions are available in the @code{gdb}
24114 @findex gdb.selected_thread
24115 @defun gdb.selected_thread ()
24116 This function returns the thread object for the selected thread. If there
24117 is no selected thread, this will return @code{None}.
24120 A @code{gdb.InferiorThread} object has the following attributes:
24123 @defvar InferiorThread.name
24124 The name of the thread. If the user specified a name using
24125 @code{thread name}, then this returns that name. Otherwise, if an
24126 OS-supplied name is available, then it is returned. Otherwise, this
24127 returns @code{None}.
24129 This attribute can be assigned to. The new value must be a string
24130 object, which sets the new name, or @code{None}, which removes any
24131 user-specified thread name.
24134 @defvar InferiorThread.num
24135 ID of the thread, as assigned by GDB.
24138 @defvar InferiorThread.ptid
24139 ID of the thread, as assigned by the operating system. This attribute is a
24140 tuple containing three integers. The first is the Process ID (PID); the second
24141 is the Lightweight Process ID (LWPID), and the third is the Thread ID (TID).
24142 Either the LWPID or TID may be 0, which indicates that the operating system
24143 does not use that identifier.
24147 A @code{gdb.InferiorThread} object has the following methods:
24150 @defun InferiorThread.is_valid ()
24151 Returns @code{True} if the @code{gdb.InferiorThread} object is valid,
24152 @code{False} if not. A @code{gdb.InferiorThread} object will become
24153 invalid if the thread exits, or the inferior that the thread belongs
24154 is deleted. All other @code{gdb.InferiorThread} methods will throw an
24155 exception if it is invalid at the time the method is called.
24158 @defun InferiorThread.switch ()
24159 This changes @value{GDBN}'s currently selected thread to the one represented
24163 @defun InferiorThread.is_stopped ()
24164 Return a Boolean indicating whether the thread is stopped.
24167 @defun InferiorThread.is_running ()
24168 Return a Boolean indicating whether the thread is running.
24171 @defun InferiorThread.is_exited ()
24172 Return a Boolean indicating whether the thread is exited.
24176 @node Commands In Python
24177 @subsubsection Commands In Python
24179 @cindex commands in python
24180 @cindex python commands
24181 You can implement new @value{GDBN} CLI commands in Python. A CLI
24182 command is implemented using an instance of the @code{gdb.Command}
24183 class, most commonly using a subclass.
24185 @defun Command.__init__ (name, @var{command_class} @r{[}, @var{completer_class} @r{[}, @var{prefix}@r{]]})
24186 The object initializer for @code{Command} registers the new command
24187 with @value{GDBN}. This initializer is normally invoked from the
24188 subclass' own @code{__init__} method.
24190 @var{name} is the name of the command. If @var{name} consists of
24191 multiple words, then the initial words are looked for as prefix
24192 commands. In this case, if one of the prefix commands does not exist,
24193 an exception is raised.
24195 There is no support for multi-line commands.
24197 @var{command_class} should be one of the @samp{COMMAND_} constants
24198 defined below. This argument tells @value{GDBN} how to categorize the
24199 new command in the help system.
24201 @var{completer_class} is an optional argument. If given, it should be
24202 one of the @samp{COMPLETE_} constants defined below. This argument
24203 tells @value{GDBN} how to perform completion for this command. If not
24204 given, @value{GDBN} will attempt to complete using the object's
24205 @code{complete} method (see below); if no such method is found, an
24206 error will occur when completion is attempted.
24208 @var{prefix} is an optional argument. If @code{True}, then the new
24209 command is a prefix command; sub-commands of this command may be
24212 The help text for the new command is taken from the Python
24213 documentation string for the command's class, if there is one. If no
24214 documentation string is provided, the default value ``This command is
24215 not documented.'' is used.
24218 @cindex don't repeat Python command
24219 @defun Command.dont_repeat ()
24220 By default, a @value{GDBN} command is repeated when the user enters a
24221 blank line at the command prompt. A command can suppress this
24222 behavior by invoking the @code{dont_repeat} method. This is similar
24223 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
24226 @defun Command.invoke (argument, from_tty)
24227 This method is called by @value{GDBN} when this command is invoked.
24229 @var{argument} is a string. It is the argument to the command, after
24230 leading and trailing whitespace has been stripped.
24232 @var{from_tty} is a boolean argument. When true, this means that the
24233 command was entered by the user at the terminal; when false it means
24234 that the command came from elsewhere.
24236 If this method throws an exception, it is turned into a @value{GDBN}
24237 @code{error} call. Otherwise, the return value is ignored.
24239 @findex gdb.string_to_argv
24240 To break @var{argument} up into an argv-like string use
24241 @code{gdb.string_to_argv}. This function behaves identically to
24242 @value{GDBN}'s internal argument lexer @code{buildargv}.
24243 It is recommended to use this for consistency.
24244 Arguments are separated by spaces and may be quoted.
24248 print gdb.string_to_argv ("1 2\ \\\"3 '4 \"5' \"6 '7\"")
24249 ['1', '2 "3', '4 "5', "6 '7"]
24254 @cindex completion of Python commands
24255 @defun Command.complete (text, word)
24256 This method is called by @value{GDBN} when the user attempts
24257 completion on this command. All forms of completion are handled by
24258 this method, that is, the @key{TAB} and @key{M-?} key bindings
24259 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
24262 The arguments @var{text} and @var{word} are both strings. @var{text}
24263 holds the complete command line up to the cursor's location.
24264 @var{word} holds the last word of the command line; this is computed
24265 using a word-breaking heuristic.
24267 The @code{complete} method can return several values:
24270 If the return value is a sequence, the contents of the sequence are
24271 used as the completions. It is up to @code{complete} to ensure that the
24272 contents actually do complete the word. A zero-length sequence is
24273 allowed, it means that there were no completions available. Only
24274 string elements of the sequence are used; other elements in the
24275 sequence are ignored.
24278 If the return value is one of the @samp{COMPLETE_} constants defined
24279 below, then the corresponding @value{GDBN}-internal completion
24280 function is invoked, and its result is used.
24283 All other results are treated as though there were no available
24288 When a new command is registered, it must be declared as a member of
24289 some general class of commands. This is used to classify top-level
24290 commands in the on-line help system; note that prefix commands are not
24291 listed under their own category but rather that of their top-level
24292 command. The available classifications are represented by constants
24293 defined in the @code{gdb} module:
24296 @findex COMMAND_NONE
24297 @findex gdb.COMMAND_NONE
24298 @item gdb.COMMAND_NONE
24299 The command does not belong to any particular class. A command in
24300 this category will not be displayed in any of the help categories.
24302 @findex COMMAND_RUNNING
24303 @findex gdb.COMMAND_RUNNING
24304 @item gdb.COMMAND_RUNNING
24305 The command is related to running the inferior. For example,
24306 @code{start}, @code{step}, and @code{continue} are in this category.
24307 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
24308 commands in this category.
24310 @findex COMMAND_DATA
24311 @findex gdb.COMMAND_DATA
24312 @item gdb.COMMAND_DATA
24313 The command is related to data or variables. For example,
24314 @code{call}, @code{find}, and @code{print} are in this category. Type
24315 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
24318 @findex COMMAND_STACK
24319 @findex gdb.COMMAND_STACK
24320 @item gdb.COMMAND_STACK
24321 The command has to do with manipulation of the stack. For example,
24322 @code{backtrace}, @code{frame}, and @code{return} are in this
24323 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
24324 list of commands in this category.
24326 @findex COMMAND_FILES
24327 @findex gdb.COMMAND_FILES
24328 @item gdb.COMMAND_FILES
24329 This class is used for file-related commands. For example,
24330 @code{file}, @code{list} and @code{section} are in this category.
24331 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
24332 commands in this category.
24334 @findex COMMAND_SUPPORT
24335 @findex gdb.COMMAND_SUPPORT
24336 @item gdb.COMMAND_SUPPORT
24337 This should be used for ``support facilities'', generally meaning
24338 things that are useful to the user when interacting with @value{GDBN},
24339 but not related to the state of the inferior. For example,
24340 @code{help}, @code{make}, and @code{shell} are in this category. Type
24341 @kbd{help support} at the @value{GDBN} prompt to see a list of
24342 commands in this category.
24344 @findex COMMAND_STATUS
24345 @findex gdb.COMMAND_STATUS
24346 @item gdb.COMMAND_STATUS
24347 The command is an @samp{info}-related command, that is, related to the
24348 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
24349 and @code{show} are in this category. Type @kbd{help status} at the
24350 @value{GDBN} prompt to see a list of commands in this category.
24352 @findex COMMAND_BREAKPOINTS
24353 @findex gdb.COMMAND_BREAKPOINTS
24354 @item gdb.COMMAND_BREAKPOINTS
24355 The command has to do with breakpoints. For example, @code{break},
24356 @code{clear}, and @code{delete} are in this category. Type @kbd{help
24357 breakpoints} at the @value{GDBN} prompt to see a list of commands in
24360 @findex COMMAND_TRACEPOINTS
24361 @findex gdb.COMMAND_TRACEPOINTS
24362 @item gdb.COMMAND_TRACEPOINTS
24363 The command has to do with tracepoints. For example, @code{trace},
24364 @code{actions}, and @code{tfind} are in this category. Type
24365 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
24366 commands in this category.
24368 @findex COMMAND_USER
24369 @findex gdb.COMMAND_USER
24370 @item gdb.COMMAND_USER
24371 The command is a general purpose command for the user, and typically
24372 does not fit in one of the other categories.
24373 Type @kbd{help user-defined} at the @value{GDBN} prompt to see
24374 a list of commands in this category, as well as the list of gdb macros
24375 (@pxref{Sequences}).
24377 @findex COMMAND_OBSCURE
24378 @findex gdb.COMMAND_OBSCURE
24379 @item gdb.COMMAND_OBSCURE
24380 The command is only used in unusual circumstances, or is not of
24381 general interest to users. For example, @code{checkpoint},
24382 @code{fork}, and @code{stop} are in this category. Type @kbd{help
24383 obscure} at the @value{GDBN} prompt to see a list of commands in this
24386 @findex COMMAND_MAINTENANCE
24387 @findex gdb.COMMAND_MAINTENANCE
24388 @item gdb.COMMAND_MAINTENANCE
24389 The command is only useful to @value{GDBN} maintainers. The
24390 @code{maintenance} and @code{flushregs} commands are in this category.
24391 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
24392 commands in this category.
24395 A new command can use a predefined completion function, either by
24396 specifying it via an argument at initialization, or by returning it
24397 from the @code{complete} method. These predefined completion
24398 constants are all defined in the @code{gdb} module:
24401 @findex COMPLETE_NONE
24402 @findex gdb.COMPLETE_NONE
24403 @item gdb.COMPLETE_NONE
24404 This constant means that no completion should be done.
24406 @findex COMPLETE_FILENAME
24407 @findex gdb.COMPLETE_FILENAME
24408 @item gdb.COMPLETE_FILENAME
24409 This constant means that filename completion should be performed.
24411 @findex COMPLETE_LOCATION
24412 @findex gdb.COMPLETE_LOCATION
24413 @item gdb.COMPLETE_LOCATION
24414 This constant means that location completion should be done.
24415 @xref{Specify Location}.
24417 @findex COMPLETE_COMMAND
24418 @findex gdb.COMPLETE_COMMAND
24419 @item gdb.COMPLETE_COMMAND
24420 This constant means that completion should examine @value{GDBN}
24423 @findex COMPLETE_SYMBOL
24424 @findex gdb.COMPLETE_SYMBOL
24425 @item gdb.COMPLETE_SYMBOL
24426 This constant means that completion should be done using symbol names
24430 The following code snippet shows how a trivial CLI command can be
24431 implemented in Python:
24434 class HelloWorld (gdb.Command):
24435 """Greet the whole world."""
24437 def __init__ (self):
24438 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
24440 def invoke (self, arg, from_tty):
24441 print "Hello, World!"
24446 The last line instantiates the class, and is necessary to trigger the
24447 registration of the command with @value{GDBN}. Depending on how the
24448 Python code is read into @value{GDBN}, you may need to import the
24449 @code{gdb} module explicitly.
24451 @node Parameters In Python
24452 @subsubsection Parameters In Python
24454 @cindex parameters in python
24455 @cindex python parameters
24456 @tindex gdb.Parameter
24458 You can implement new @value{GDBN} parameters using Python. A new
24459 parameter is implemented as an instance of the @code{gdb.Parameter}
24462 Parameters are exposed to the user via the @code{set} and
24463 @code{show} commands. @xref{Help}.
24465 There are many parameters that already exist and can be set in
24466 @value{GDBN}. Two examples are: @code{set follow fork} and
24467 @code{set charset}. Setting these parameters influences certain
24468 behavior in @value{GDBN}. Similarly, you can define parameters that
24469 can be used to influence behavior in custom Python scripts and commands.
24471 @defun Parameter.__init__ (name, @var{command-class}, @var{parameter-class} @r{[}, @var{enum-sequence}@r{]})
24472 The object initializer for @code{Parameter} registers the new
24473 parameter with @value{GDBN}. This initializer is normally invoked
24474 from the subclass' own @code{__init__} method.
24476 @var{name} is the name of the new parameter. If @var{name} consists
24477 of multiple words, then the initial words are looked for as prefix
24478 parameters. An example of this can be illustrated with the
24479 @code{set print} set of parameters. If @var{name} is
24480 @code{print foo}, then @code{print} will be searched as the prefix
24481 parameter. In this case the parameter can subsequently be accessed in
24482 @value{GDBN} as @code{set print foo}.
24484 If @var{name} consists of multiple words, and no prefix parameter group
24485 can be found, an exception is raised.
24487 @var{command-class} should be one of the @samp{COMMAND_} constants
24488 (@pxref{Commands In Python}). This argument tells @value{GDBN} how to
24489 categorize the new parameter in the help system.
24491 @var{parameter-class} should be one of the @samp{PARAM_} constants
24492 defined below. This argument tells @value{GDBN} the type of the new
24493 parameter; this information is used for input validation and
24496 If @var{parameter-class} is @code{PARAM_ENUM}, then
24497 @var{enum-sequence} must be a sequence of strings. These strings
24498 represent the possible values for the parameter.
24500 If @var{parameter-class} is not @code{PARAM_ENUM}, then the presence
24501 of a fourth argument will cause an exception to be thrown.
24503 The help text for the new parameter is taken from the Python
24504 documentation string for the parameter's class, if there is one. If
24505 there is no documentation string, a default value is used.
24508 @defvar Parameter.set_doc
24509 If this attribute exists, and is a string, then its value is used as
24510 the help text for this parameter's @code{set} command. The value is
24511 examined when @code{Parameter.__init__} is invoked; subsequent changes
24515 @defvar Parameter.show_doc
24516 If this attribute exists, and is a string, then its value is used as
24517 the help text for this parameter's @code{show} command. The value is
24518 examined when @code{Parameter.__init__} is invoked; subsequent changes
24522 @defvar Parameter.value
24523 The @code{value} attribute holds the underlying value of the
24524 parameter. It can be read and assigned to just as any other
24525 attribute. @value{GDBN} does validation when assignments are made.
24528 There are two methods that should be implemented in any
24529 @code{Parameter} class. These are:
24531 @defun Parameter.get_set_string (self)
24532 @value{GDBN} will call this method when a @var{parameter}'s value has
24533 been changed via the @code{set} API (for example, @kbd{set foo off}).
24534 The @code{value} attribute has already been populated with the new
24535 value and may be used in output. This method must return a string.
24538 @defun Parameter.get_show_string (self, svalue)
24539 @value{GDBN} will call this method when a @var{parameter}'s
24540 @code{show} API has been invoked (for example, @kbd{show foo}). The
24541 argument @code{svalue} receives the string representation of the
24542 current value. This method must return a string.
24545 When a new parameter is defined, its type must be specified. The
24546 available types are represented by constants defined in the @code{gdb}
24550 @findex PARAM_BOOLEAN
24551 @findex gdb.PARAM_BOOLEAN
24552 @item gdb.PARAM_BOOLEAN
24553 The value is a plain boolean. The Python boolean values, @code{True}
24554 and @code{False} are the only valid values.
24556 @findex PARAM_AUTO_BOOLEAN
24557 @findex gdb.PARAM_AUTO_BOOLEAN
24558 @item gdb.PARAM_AUTO_BOOLEAN
24559 The value has three possible states: true, false, and @samp{auto}. In
24560 Python, true and false are represented using boolean constants, and
24561 @samp{auto} is represented using @code{None}.
24563 @findex PARAM_UINTEGER
24564 @findex gdb.PARAM_UINTEGER
24565 @item gdb.PARAM_UINTEGER
24566 The value is an unsigned integer. The value of 0 should be
24567 interpreted to mean ``unlimited''.
24569 @findex PARAM_INTEGER
24570 @findex gdb.PARAM_INTEGER
24571 @item gdb.PARAM_INTEGER
24572 The value is a signed integer. The value of 0 should be interpreted
24573 to mean ``unlimited''.
24575 @findex PARAM_STRING
24576 @findex gdb.PARAM_STRING
24577 @item gdb.PARAM_STRING
24578 The value is a string. When the user modifies the string, any escape
24579 sequences, such as @samp{\t}, @samp{\f}, and octal escapes, are
24580 translated into corresponding characters and encoded into the current
24583 @findex PARAM_STRING_NOESCAPE
24584 @findex gdb.PARAM_STRING_NOESCAPE
24585 @item gdb.PARAM_STRING_NOESCAPE
24586 The value is a string. When the user modifies the string, escapes are
24587 passed through untranslated.
24589 @findex PARAM_OPTIONAL_FILENAME
24590 @findex gdb.PARAM_OPTIONAL_FILENAME
24591 @item gdb.PARAM_OPTIONAL_FILENAME
24592 The value is a either a filename (a string), or @code{None}.
24594 @findex PARAM_FILENAME
24595 @findex gdb.PARAM_FILENAME
24596 @item gdb.PARAM_FILENAME
24597 The value is a filename. This is just like
24598 @code{PARAM_STRING_NOESCAPE}, but uses file names for completion.
24600 @findex PARAM_ZINTEGER
24601 @findex gdb.PARAM_ZINTEGER
24602 @item gdb.PARAM_ZINTEGER
24603 The value is an integer. This is like @code{PARAM_INTEGER}, except 0
24604 is interpreted as itself.
24607 @findex gdb.PARAM_ENUM
24608 @item gdb.PARAM_ENUM
24609 The value is a string, which must be one of a collection string
24610 constants provided when the parameter is created.
24613 @node Functions In Python
24614 @subsubsection Writing new convenience functions
24616 @cindex writing convenience functions
24617 @cindex convenience functions in python
24618 @cindex python convenience functions
24619 @tindex gdb.Function
24621 You can implement new convenience functions (@pxref{Convenience Vars})
24622 in Python. A convenience function is an instance of a subclass of the
24623 class @code{gdb.Function}.
24625 @defun Function.__init__ (name)
24626 The initializer for @code{Function} registers the new function with
24627 @value{GDBN}. The argument @var{name} is the name of the function,
24628 a string. The function will be visible to the user as a convenience
24629 variable of type @code{internal function}, whose name is the same as
24630 the given @var{name}.
24632 The documentation for the new function is taken from the documentation
24633 string for the new class.
24636 @defun Function.invoke (@var{*args})
24637 When a convenience function is evaluated, its arguments are converted
24638 to instances of @code{gdb.Value}, and then the function's
24639 @code{invoke} method is called. Note that @value{GDBN} does not
24640 predetermine the arity of convenience functions. Instead, all
24641 available arguments are passed to @code{invoke}, following the
24642 standard Python calling convention. In particular, a convenience
24643 function can have default values for parameters without ill effect.
24645 The return value of this method is used as its value in the enclosing
24646 expression. If an ordinary Python value is returned, it is converted
24647 to a @code{gdb.Value} following the usual rules.
24650 The following code snippet shows how a trivial convenience function can
24651 be implemented in Python:
24654 class Greet (gdb.Function):
24655 """Return string to greet someone.
24656 Takes a name as argument."""
24658 def __init__ (self):
24659 super (Greet, self).__init__ ("greet")
24661 def invoke (self, name):
24662 return "Hello, %s!" % name.string ()
24667 The last line instantiates the class, and is necessary to trigger the
24668 registration of the function with @value{GDBN}. Depending on how the
24669 Python code is read into @value{GDBN}, you may need to import the
24670 @code{gdb} module explicitly.
24672 @node Progspaces In Python
24673 @subsubsection Program Spaces In Python
24675 @cindex progspaces in python
24676 @tindex gdb.Progspace
24678 A program space, or @dfn{progspace}, represents a symbolic view
24679 of an address space.
24680 It consists of all of the objfiles of the program.
24681 @xref{Objfiles In Python}.
24682 @xref{Inferiors and Programs, program spaces}, for more details
24683 about program spaces.
24685 The following progspace-related functions are available in the
24688 @findex gdb.current_progspace
24689 @defun gdb.current_progspace ()
24690 This function returns the program space of the currently selected inferior.
24691 @xref{Inferiors and Programs}.
24694 @findex gdb.progspaces
24695 @defun gdb.progspaces ()
24696 Return a sequence of all the progspaces currently known to @value{GDBN}.
24699 Each progspace is represented by an instance of the @code{gdb.Progspace}
24702 @defvar Progspace.filename
24703 The file name of the progspace as a string.
24706 @defvar Progspace.pretty_printers
24707 The @code{pretty_printers} attribute is a list of functions. It is
24708 used to look up pretty-printers. A @code{Value} is passed to each
24709 function in order; if the function returns @code{None}, then the
24710 search continues. Otherwise, the return value should be an object
24711 which is used to format the value. @xref{Pretty Printing API}, for more
24715 @node Objfiles In Python
24716 @subsubsection Objfiles In Python
24718 @cindex objfiles in python
24719 @tindex gdb.Objfile
24721 @value{GDBN} loads symbols for an inferior from various
24722 symbol-containing files (@pxref{Files}). These include the primary
24723 executable file, any shared libraries used by the inferior, and any
24724 separate debug info files (@pxref{Separate Debug Files}).
24725 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
24727 The following objfile-related functions are available in the
24730 @findex gdb.current_objfile
24731 @defun gdb.current_objfile ()
24732 When auto-loading a Python script (@pxref{Python Auto-loading}), @value{GDBN}
24733 sets the ``current objfile'' to the corresponding objfile. This
24734 function returns the current objfile. If there is no current objfile,
24735 this function returns @code{None}.
24738 @findex gdb.objfiles
24739 @defun gdb.objfiles ()
24740 Return a sequence of all the objfiles current known to @value{GDBN}.
24741 @xref{Objfiles In Python}.
24744 Each objfile is represented by an instance of the @code{gdb.Objfile}
24747 @defvar Objfile.filename
24748 The file name of the objfile as a string.
24751 @defvar Objfile.pretty_printers
24752 The @code{pretty_printers} attribute is a list of functions. It is
24753 used to look up pretty-printers. A @code{Value} is passed to each
24754 function in order; if the function returns @code{None}, then the
24755 search continues. Otherwise, the return value should be an object
24756 which is used to format the value. @xref{Pretty Printing API}, for more
24760 A @code{gdb.Objfile} object has the following methods:
24762 @defun Objfile.is_valid ()
24763 Returns @code{True} if the @code{gdb.Objfile} object is valid,
24764 @code{False} if not. A @code{gdb.Objfile} object can become invalid
24765 if the object file it refers to is not loaded in @value{GDBN} any
24766 longer. All other @code{gdb.Objfile} methods will throw an exception
24767 if it is invalid at the time the method is called.
24770 @node Frames In Python
24771 @subsubsection Accessing inferior stack frames from Python.
24773 @cindex frames in python
24774 When the debugged program stops, @value{GDBN} is able to analyze its call
24775 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
24776 represents a frame in the stack. A @code{gdb.Frame} object is only valid
24777 while its corresponding frame exists in the inferior's stack. If you try
24778 to use an invalid frame object, @value{GDBN} will throw a @code{gdb.error}
24779 exception (@pxref{Exception Handling}).
24781 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
24785 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
24789 The following frame-related functions are available in the @code{gdb} module:
24791 @findex gdb.selected_frame
24792 @defun gdb.selected_frame ()
24793 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
24796 @findex gdb.newest_frame
24797 @defun gdb.newest_frame ()
24798 Return the newest frame object for the selected thread.
24801 @defun gdb.frame_stop_reason_string (reason)
24802 Return a string explaining the reason why @value{GDBN} stopped unwinding
24803 frames, as expressed by the given @var{reason} code (an integer, see the
24804 @code{unwind_stop_reason} method further down in this section).
24807 A @code{gdb.Frame} object has the following methods:
24810 @defun Frame.is_valid ()
24811 Returns true if the @code{gdb.Frame} object is valid, false if not.
24812 A frame object can become invalid if the frame it refers to doesn't
24813 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
24814 an exception if it is invalid at the time the method is called.
24817 @defun Frame.name ()
24818 Returns the function name of the frame, or @code{None} if it can't be
24822 @defun Frame.type ()
24823 Returns the type of the frame. The value can be one of:
24825 @item gdb.NORMAL_FRAME
24826 An ordinary stack frame.
24828 @item gdb.DUMMY_FRAME
24829 A fake stack frame that was created by @value{GDBN} when performing an
24830 inferior function call.
24832 @item gdb.INLINE_FRAME
24833 A frame representing an inlined function. The function was inlined
24834 into a @code{gdb.NORMAL_FRAME} that is older than this one.
24836 @item gdb.TAILCALL_FRAME
24837 A frame representing a tail call. @xref{Tail Call Frames}.
24839 @item gdb.SIGTRAMP_FRAME
24840 A signal trampoline frame. This is the frame created by the OS when
24841 it calls into a signal handler.
24843 @item gdb.ARCH_FRAME
24844 A fake stack frame representing a cross-architecture call.
24846 @item gdb.SENTINEL_FRAME
24847 This is like @code{gdb.NORMAL_FRAME}, but it is only used for the
24852 @defun Frame.unwind_stop_reason ()
24853 Return an integer representing the reason why it's not possible to find
24854 more frames toward the outermost frame. Use
24855 @code{gdb.frame_stop_reason_string} to convert the value returned by this
24856 function to a string. The value can be one of:
24859 @item gdb.FRAME_UNWIND_NO_REASON
24860 No particular reason (older frames should be available).
24862 @item gdb.FRAME_UNWIND_NULL_ID
24863 The previous frame's analyzer returns an invalid result.
24865 @item gdb.FRAME_UNWIND_OUTERMOST
24866 This frame is the outermost.
24868 @item gdb.FRAME_UNWIND_UNAVAILABLE
24869 Cannot unwind further, because that would require knowing the
24870 values of registers or memory that have not been collected.
24872 @item gdb.FRAME_UNWIND_INNER_ID
24873 This frame ID looks like it ought to belong to a NEXT frame,
24874 but we got it for a PREV frame. Normally, this is a sign of
24875 unwinder failure. It could also indicate stack corruption.
24877 @item gdb.FRAME_UNWIND_SAME_ID
24878 This frame has the same ID as the previous one. That means
24879 that unwinding further would almost certainly give us another
24880 frame with exactly the same ID, so break the chain. Normally,
24881 this is a sign of unwinder failure. It could also indicate
24884 @item gdb.FRAME_UNWIND_NO_SAVED_PC
24885 The frame unwinder did not find any saved PC, but we needed
24886 one to unwind further.
24888 @item gdb.FRAME_UNWIND_FIRST_ERROR
24889 Any stop reason greater or equal to this value indicates some kind
24890 of error. This special value facilitates writing code that tests
24891 for errors in unwinding in a way that will work correctly even if
24892 the list of the other values is modified in future @value{GDBN}
24893 versions. Using it, you could write:
24895 reason = gdb.selected_frame().unwind_stop_reason ()
24896 reason_str = gdb.frame_stop_reason_string (reason)
24897 if reason >= gdb.FRAME_UNWIND_FIRST_ERROR:
24898 print "An error occured: %s" % reason_str
24905 Returns the frame's resume address.
24908 @defun Frame.block ()
24909 Return the frame's code block. @xref{Blocks In Python}.
24912 @defun Frame.function ()
24913 Return the symbol for the function corresponding to this frame.
24914 @xref{Symbols In Python}.
24917 @defun Frame.older ()
24918 Return the frame that called this frame.
24921 @defun Frame.newer ()
24922 Return the frame called by this frame.
24925 @defun Frame.find_sal ()
24926 Return the frame's symtab and line object.
24927 @xref{Symbol Tables In Python}.
24930 @defun Frame.read_var (variable @r{[}, block@r{]})
24931 Return the value of @var{variable} in this frame. If the optional
24932 argument @var{block} is provided, search for the variable from that
24933 block; otherwise start at the frame's current block (which is
24934 determined by the frame's current program counter). @var{variable}
24935 must be a string or a @code{gdb.Symbol} object. @var{block} must be a
24936 @code{gdb.Block} object.
24939 @defun Frame.select ()
24940 Set this frame to be the selected frame. @xref{Stack, ,Examining the
24945 @node Blocks In Python
24946 @subsubsection Accessing frame blocks from Python.
24948 @cindex blocks in python
24951 Within each frame, @value{GDBN} maintains information on each block
24952 stored in that frame. These blocks are organized hierarchically, and
24953 are represented individually in Python as a @code{gdb.Block}.
24954 Please see @ref{Frames In Python}, for a more in-depth discussion on
24955 frames. Furthermore, see @ref{Stack, ,Examining the Stack}, for more
24956 detailed technical information on @value{GDBN}'s book-keeping of the
24959 A @code{gdb.Block} is iterable. The iterator returns the symbols
24960 (@pxref{Symbols In Python}) local to the block. Python programs
24961 should not assume that a specific block object will always contain a
24962 given symbol, since changes in @value{GDBN} features and
24963 infrastructure may cause symbols move across blocks in a symbol
24966 The following block-related functions are available in the @code{gdb}
24969 @findex gdb.block_for_pc
24970 @defun gdb.block_for_pc (pc)
24971 Return the @code{gdb.Block} containing the given @var{pc} value. If the
24972 block cannot be found for the @var{pc} value specified, the function
24973 will return @code{None}.
24976 A @code{gdb.Block} object has the following methods:
24979 @defun Block.is_valid ()
24980 Returns @code{True} if the @code{gdb.Block} object is valid,
24981 @code{False} if not. A block object can become invalid if the block it
24982 refers to doesn't exist anymore in the inferior. All other
24983 @code{gdb.Block} methods will throw an exception if it is invalid at
24984 the time the method is called. The block's validity is also checked
24985 during iteration over symbols of the block.
24989 A @code{gdb.Block} object has the following attributes:
24992 @defvar Block.start
24993 The start address of the block. This attribute is not writable.
24997 The end address of the block. This attribute is not writable.
25000 @defvar Block.function
25001 The name of the block represented as a @code{gdb.Symbol}. If the
25002 block is not named, then this attribute holds @code{None}. This
25003 attribute is not writable.
25006 @defvar Block.superblock
25007 The block containing this block. If this parent block does not exist,
25008 this attribute holds @code{None}. This attribute is not writable.
25011 @defvar Block.global_block
25012 The global block associated with this block. This attribute is not
25016 @defvar Block.static_block
25017 The static block associated with this block. This attribute is not
25021 @defvar Block.is_global
25022 @code{True} if the @code{gdb.Block} object is a global block,
25023 @code{False} if not. This attribute is not
25027 @defvar Block.is_static
25028 @code{True} if the @code{gdb.Block} object is a static block,
25029 @code{False} if not. This attribute is not writable.
25033 @node Symbols In Python
25034 @subsubsection Python representation of Symbols.
25036 @cindex symbols in python
25039 @value{GDBN} represents every variable, function and type as an
25040 entry in a symbol table. @xref{Symbols, ,Examining the Symbol Table}.
25041 Similarly, Python represents these symbols in @value{GDBN} with the
25042 @code{gdb.Symbol} object.
25044 The following symbol-related functions are available in the @code{gdb}
25047 @findex gdb.lookup_symbol
25048 @defun gdb.lookup_symbol (name @r{[}, block @r{[}, domain@r{]]})
25049 This function searches for a symbol by name. The search scope can be
25050 restricted to the parameters defined in the optional domain and block
25053 @var{name} is the name of the symbol. It must be a string. The
25054 optional @var{block} argument restricts the search to symbols visible
25055 in that @var{block}. The @var{block} argument must be a
25056 @code{gdb.Block} object. If omitted, the block for the current frame
25057 is used. The optional @var{domain} argument restricts
25058 the search to the domain type. The @var{domain} argument must be a
25059 domain constant defined in the @code{gdb} module and described later
25062 The result is a tuple of two elements.
25063 The first element is a @code{gdb.Symbol} object or @code{None} if the symbol
25065 If the symbol is found, the second element is @code{True} if the symbol
25066 is a field of a method's object (e.g., @code{this} in C@t{++}),
25067 otherwise it is @code{False}.
25068 If the symbol is not found, the second element is @code{False}.
25071 @findex gdb.lookup_global_symbol
25072 @defun gdb.lookup_global_symbol (name @r{[}, domain@r{]})
25073 This function searches for a global symbol by name.
25074 The search scope can be restricted to by the domain argument.
25076 @var{name} is the name of the symbol. It must be a string.
25077 The optional @var{domain} argument restricts the search to the domain type.
25078 The @var{domain} argument must be a domain constant defined in the @code{gdb}
25079 module and described later in this chapter.
25081 The result is a @code{gdb.Symbol} object or @code{None} if the symbol
25085 A @code{gdb.Symbol} object has the following attributes:
25088 @defvar Symbol.type
25089 The type of the symbol or @code{None} if no type is recorded.
25090 This attribute is represented as a @code{gdb.Type} object.
25091 @xref{Types In Python}. This attribute is not writable.
25094 @defvar Symbol.symtab
25095 The symbol table in which the symbol appears. This attribute is
25096 represented as a @code{gdb.Symtab} object. @xref{Symbol Tables In
25097 Python}. This attribute is not writable.
25100 @defvar Symbol.line
25101 The line number in the source code at which the symbol was defined.
25102 This is an integer.
25105 @defvar Symbol.name
25106 The name of the symbol as a string. This attribute is not writable.
25109 @defvar Symbol.linkage_name
25110 The name of the symbol, as used by the linker (i.e., may be mangled).
25111 This attribute is not writable.
25114 @defvar Symbol.print_name
25115 The name of the symbol in a form suitable for output. This is either
25116 @code{name} or @code{linkage_name}, depending on whether the user
25117 asked @value{GDBN} to display demangled or mangled names.
25120 @defvar Symbol.addr_class
25121 The address class of the symbol. This classifies how to find the value
25122 of a symbol. Each address class is a constant defined in the
25123 @code{gdb} module and described later in this chapter.
25126 @defvar Symbol.needs_frame
25127 This is @code{True} if evaluating this symbol's value requires a frame
25128 (@pxref{Frames In Python}) and @code{False} otherwise. Typically,
25129 local variables will require a frame, but other symbols will not.
25132 @defvar Symbol.is_argument
25133 @code{True} if the symbol is an argument of a function.
25136 @defvar Symbol.is_constant
25137 @code{True} if the symbol is a constant.
25140 @defvar Symbol.is_function
25141 @code{True} if the symbol is a function or a method.
25144 @defvar Symbol.is_variable
25145 @code{True} if the symbol is a variable.
25149 A @code{gdb.Symbol} object has the following methods:
25152 @defun Symbol.is_valid ()
25153 Returns @code{True} if the @code{gdb.Symbol} object is valid,
25154 @code{False} if not. A @code{gdb.Symbol} object can become invalid if
25155 the symbol it refers to does not exist in @value{GDBN} any longer.
25156 All other @code{gdb.Symbol} methods will throw an exception if it is
25157 invalid at the time the method is called.
25160 @defun Symbol.value (@r{[}frame@r{]})
25161 Compute the value of the symbol, as a @code{gdb.Value}. For
25162 functions, this computes the address of the function, cast to the
25163 appropriate type. If the symbol requires a frame in order to compute
25164 its value, then @var{frame} must be given. If @var{frame} is not
25165 given, or if @var{frame} is invalid, then this method will throw an
25170 The available domain categories in @code{gdb.Symbol} are represented
25171 as constants in the @code{gdb} module:
25174 @findex SYMBOL_UNDEF_DOMAIN
25175 @findex gdb.SYMBOL_UNDEF_DOMAIN
25176 @item gdb.SYMBOL_UNDEF_DOMAIN
25177 This is used when a domain has not been discovered or none of the
25178 following domains apply. This usually indicates an error either
25179 in the symbol information or in @value{GDBN}'s handling of symbols.
25180 @findex SYMBOL_VAR_DOMAIN
25181 @findex gdb.SYMBOL_VAR_DOMAIN
25182 @item gdb.SYMBOL_VAR_DOMAIN
25183 This domain contains variables, function names, typedef names and enum
25185 @findex SYMBOL_STRUCT_DOMAIN
25186 @findex gdb.SYMBOL_STRUCT_DOMAIN
25187 @item gdb.SYMBOL_STRUCT_DOMAIN
25188 This domain holds struct, union and enum type names.
25189 @findex SYMBOL_LABEL_DOMAIN
25190 @findex gdb.SYMBOL_LABEL_DOMAIN
25191 @item gdb.SYMBOL_LABEL_DOMAIN
25192 This domain contains names of labels (for gotos).
25193 @findex SYMBOL_VARIABLES_DOMAIN
25194 @findex gdb.SYMBOL_VARIABLES_DOMAIN
25195 @item gdb.SYMBOL_VARIABLES_DOMAIN
25196 This domain holds a subset of the @code{SYMBOLS_VAR_DOMAIN}; it
25197 contains everything minus functions and types.
25198 @findex SYMBOL_FUNCTIONS_DOMAIN
25199 @findex gdb.SYMBOL_FUNCTIONS_DOMAIN
25200 @item gdb.SYMBOL_FUNCTION_DOMAIN
25201 This domain contains all functions.
25202 @findex SYMBOL_TYPES_DOMAIN
25203 @findex gdb.SYMBOL_TYPES_DOMAIN
25204 @item gdb.SYMBOL_TYPES_DOMAIN
25205 This domain contains all types.
25208 The available address class categories in @code{gdb.Symbol} are represented
25209 as constants in the @code{gdb} module:
25212 @findex SYMBOL_LOC_UNDEF
25213 @findex gdb.SYMBOL_LOC_UNDEF
25214 @item gdb.SYMBOL_LOC_UNDEF
25215 If this is returned by address class, it indicates an error either in
25216 the symbol information or in @value{GDBN}'s handling of symbols.
25217 @findex SYMBOL_LOC_CONST
25218 @findex gdb.SYMBOL_LOC_CONST
25219 @item gdb.SYMBOL_LOC_CONST
25220 Value is constant int.
25221 @findex SYMBOL_LOC_STATIC
25222 @findex gdb.SYMBOL_LOC_STATIC
25223 @item gdb.SYMBOL_LOC_STATIC
25224 Value is at a fixed address.
25225 @findex SYMBOL_LOC_REGISTER
25226 @findex gdb.SYMBOL_LOC_REGISTER
25227 @item gdb.SYMBOL_LOC_REGISTER
25228 Value is in a register.
25229 @findex SYMBOL_LOC_ARG
25230 @findex gdb.SYMBOL_LOC_ARG
25231 @item gdb.SYMBOL_LOC_ARG
25232 Value is an argument. This value is at the offset stored within the
25233 symbol inside the frame's argument list.
25234 @findex SYMBOL_LOC_REF_ARG
25235 @findex gdb.SYMBOL_LOC_REF_ARG
25236 @item gdb.SYMBOL_LOC_REF_ARG
25237 Value address is stored in the frame's argument list. Just like
25238 @code{LOC_ARG} except that the value's address is stored at the
25239 offset, not the value itself.
25240 @findex SYMBOL_LOC_REGPARM_ADDR
25241 @findex gdb.SYMBOL_LOC_REGPARM_ADDR
25242 @item gdb.SYMBOL_LOC_REGPARM_ADDR
25243 Value is a specified register. Just like @code{LOC_REGISTER} except
25244 the register holds the address of the argument instead of the argument
25246 @findex SYMBOL_LOC_LOCAL
25247 @findex gdb.SYMBOL_LOC_LOCAL
25248 @item gdb.SYMBOL_LOC_LOCAL
25249 Value is a local variable.
25250 @findex SYMBOL_LOC_TYPEDEF
25251 @findex gdb.SYMBOL_LOC_TYPEDEF
25252 @item gdb.SYMBOL_LOC_TYPEDEF
25253 Value not used. Symbols in the domain @code{SYMBOL_STRUCT_DOMAIN} all
25255 @findex SYMBOL_LOC_BLOCK
25256 @findex gdb.SYMBOL_LOC_BLOCK
25257 @item gdb.SYMBOL_LOC_BLOCK
25259 @findex SYMBOL_LOC_CONST_BYTES
25260 @findex gdb.SYMBOL_LOC_CONST_BYTES
25261 @item gdb.SYMBOL_LOC_CONST_BYTES
25262 Value is a byte-sequence.
25263 @findex SYMBOL_LOC_UNRESOLVED
25264 @findex gdb.SYMBOL_LOC_UNRESOLVED
25265 @item gdb.SYMBOL_LOC_UNRESOLVED
25266 Value is at a fixed address, but the address of the variable has to be
25267 determined from the minimal symbol table whenever the variable is
25269 @findex SYMBOL_LOC_OPTIMIZED_OUT
25270 @findex gdb.SYMBOL_LOC_OPTIMIZED_OUT
25271 @item gdb.SYMBOL_LOC_OPTIMIZED_OUT
25272 The value does not actually exist in the program.
25273 @findex SYMBOL_LOC_COMPUTED
25274 @findex gdb.SYMBOL_LOC_COMPUTED
25275 @item gdb.SYMBOL_LOC_COMPUTED
25276 The value's address is a computed location.
25279 @node Symbol Tables In Python
25280 @subsubsection Symbol table representation in Python.
25282 @cindex symbol tables in python
25284 @tindex gdb.Symtab_and_line
25286 Access to symbol table data maintained by @value{GDBN} on the inferior
25287 is exposed to Python via two objects: @code{gdb.Symtab_and_line} and
25288 @code{gdb.Symtab}. Symbol table and line data for a frame is returned
25289 from the @code{find_sal} method in @code{gdb.Frame} object.
25290 @xref{Frames In Python}.
25292 For more information on @value{GDBN}'s symbol table management, see
25293 @ref{Symbols, ,Examining the Symbol Table}, for more information.
25295 A @code{gdb.Symtab_and_line} object has the following attributes:
25298 @defvar Symtab_and_line.symtab
25299 The symbol table object (@code{gdb.Symtab}) for this frame.
25300 This attribute is not writable.
25303 @defvar Symtab_and_line.pc
25304 Indicates the start of the address range occupied by code for the
25305 current source line. This attribute is not writable.
25308 @defvar Symtab_and_line.last
25309 Indicates the end of the address range occupied by code for the current
25310 source line. This attribute is not writable.
25313 @defvar Symtab_and_line.line
25314 Indicates the current line number for this object. This
25315 attribute is not writable.
25319 A @code{gdb.Symtab_and_line} object has the following methods:
25322 @defun Symtab_and_line.is_valid ()
25323 Returns @code{True} if the @code{gdb.Symtab_and_line} object is valid,
25324 @code{False} if not. A @code{gdb.Symtab_and_line} object can become
25325 invalid if the Symbol table and line object it refers to does not
25326 exist in @value{GDBN} any longer. All other
25327 @code{gdb.Symtab_and_line} methods will throw an exception if it is
25328 invalid at the time the method is called.
25332 A @code{gdb.Symtab} object has the following attributes:
25335 @defvar Symtab.filename
25336 The symbol table's source filename. This attribute is not writable.
25339 @defvar Symtab.objfile
25340 The symbol table's backing object file. @xref{Objfiles In Python}.
25341 This attribute is not writable.
25345 A @code{gdb.Symtab} object has the following methods:
25348 @defun Symtab.is_valid ()
25349 Returns @code{True} if the @code{gdb.Symtab} object is valid,
25350 @code{False} if not. A @code{gdb.Symtab} object can become invalid if
25351 the symbol table it refers to does not exist in @value{GDBN} any
25352 longer. All other @code{gdb.Symtab} methods will throw an exception
25353 if it is invalid at the time the method is called.
25356 @defun Symtab.fullname ()
25357 Return the symbol table's source absolute file name.
25360 @defun Symtab.global_block ()
25361 Return the global block of the underlying symbol table.
25362 @xref{Blocks In Python}.
25365 @defun Symtab.static_block ()
25366 Return the static block of the underlying symbol table.
25367 @xref{Blocks In Python}.
25371 @node Breakpoints In Python
25372 @subsubsection Manipulating breakpoints using Python
25374 @cindex breakpoints in python
25375 @tindex gdb.Breakpoint
25377 Python code can manipulate breakpoints via the @code{gdb.Breakpoint}
25380 @defun Breakpoint.__init__ (spec @r{[}, type @r{[}, wp_class @r{[},internal@r{]]]})
25381 Create a new breakpoint. @var{spec} is a string naming the
25382 location of the breakpoint, or an expression that defines a
25383 watchpoint. The contents can be any location recognized by the
25384 @code{break} command, or in the case of a watchpoint, by the @code{watch}
25385 command. The optional @var{type} denotes the breakpoint to create
25386 from the types defined later in this chapter. This argument can be
25387 either: @code{gdb.BP_BREAKPOINT} or @code{gdb.BP_WATCHPOINT}. @var{type}
25388 defaults to @code{gdb.BP_BREAKPOINT}. The optional @var{internal} argument
25389 allows the breakpoint to become invisible to the user. The breakpoint
25390 will neither be reported when created, nor will it be listed in the
25391 output from @code{info breakpoints} (but will be listed with the
25392 @code{maint info breakpoints} command). The optional @var{wp_class}
25393 argument defines the class of watchpoint to create, if @var{type} is
25394 @code{gdb.BP_WATCHPOINT}. If a watchpoint class is not provided, it is
25395 assumed to be a @code{gdb.WP_WRITE} class.
25398 @defun Breakpoint.stop (self)
25399 The @code{gdb.Breakpoint} class can be sub-classed and, in
25400 particular, you may choose to implement the @code{stop} method.
25401 If this method is defined as a sub-class of @code{gdb.Breakpoint},
25402 it will be called when the inferior reaches any location of a
25403 breakpoint which instantiates that sub-class. If the method returns
25404 @code{True}, the inferior will be stopped at the location of the
25405 breakpoint, otherwise the inferior will continue.
25407 If there are multiple breakpoints at the same location with a
25408 @code{stop} method, each one will be called regardless of the
25409 return status of the previous. This ensures that all @code{stop}
25410 methods have a chance to execute at that location. In this scenario
25411 if one of the methods returns @code{True} but the others return
25412 @code{False}, the inferior will still be stopped.
25414 You should not alter the execution state of the inferior (i.e.@:, step,
25415 next, etc.), alter the current frame context (i.e.@:, change the current
25416 active frame), or alter, add or delete any breakpoint. As a general
25417 rule, you should not alter any data within @value{GDBN} or the inferior
25420 Example @code{stop} implementation:
25423 class MyBreakpoint (gdb.Breakpoint):
25425 inf_val = gdb.parse_and_eval("foo")
25432 The available watchpoint types represented by constants are defined in the
25437 @findex gdb.WP_READ
25439 Read only watchpoint.
25442 @findex gdb.WP_WRITE
25444 Write only watchpoint.
25447 @findex gdb.WP_ACCESS
25448 @item gdb.WP_ACCESS
25449 Read/Write watchpoint.
25452 @defun Breakpoint.is_valid ()
25453 Return @code{True} if this @code{Breakpoint} object is valid,
25454 @code{False} otherwise. A @code{Breakpoint} object can become invalid
25455 if the user deletes the breakpoint. In this case, the object still
25456 exists, but the underlying breakpoint does not. In the cases of
25457 watchpoint scope, the watchpoint remains valid even if execution of the
25458 inferior leaves the scope of that watchpoint.
25461 @defun Breakpoint.delete
25462 Permanently deletes the @value{GDBN} breakpoint. This also
25463 invalidates the Python @code{Breakpoint} object. Any further access
25464 to this object's attributes or methods will raise an error.
25467 @defvar Breakpoint.enabled
25468 This attribute is @code{True} if the breakpoint is enabled, and
25469 @code{False} otherwise. This attribute is writable.
25472 @defvar Breakpoint.silent
25473 This attribute is @code{True} if the breakpoint is silent, and
25474 @code{False} otherwise. This attribute is writable.
25476 Note that a breakpoint can also be silent if it has commands and the
25477 first command is @code{silent}. This is not reported by the
25478 @code{silent} attribute.
25481 @defvar Breakpoint.thread
25482 If the breakpoint is thread-specific, this attribute holds the thread
25483 id. If the breakpoint is not thread-specific, this attribute is
25484 @code{None}. This attribute is writable.
25487 @defvar Breakpoint.task
25488 If the breakpoint is Ada task-specific, this attribute holds the Ada task
25489 id. If the breakpoint is not task-specific (or the underlying
25490 language is not Ada), this attribute is @code{None}. This attribute
25494 @defvar Breakpoint.ignore_count
25495 This attribute holds the ignore count for the breakpoint, an integer.
25496 This attribute is writable.
25499 @defvar Breakpoint.number
25500 This attribute holds the breakpoint's number --- the identifier used by
25501 the user to manipulate the breakpoint. This attribute is not writable.
25504 @defvar Breakpoint.type
25505 This attribute holds the breakpoint's type --- the identifier used to
25506 determine the actual breakpoint type or use-case. This attribute is not
25510 @defvar Breakpoint.visible
25511 This attribute tells whether the breakpoint is visible to the user
25512 when set, or when the @samp{info breakpoints} command is run. This
25513 attribute is not writable.
25516 The available types are represented by constants defined in the @code{gdb}
25520 @findex BP_BREAKPOINT
25521 @findex gdb.BP_BREAKPOINT
25522 @item gdb.BP_BREAKPOINT
25523 Normal code breakpoint.
25525 @findex BP_WATCHPOINT
25526 @findex gdb.BP_WATCHPOINT
25527 @item gdb.BP_WATCHPOINT
25528 Watchpoint breakpoint.
25530 @findex BP_HARDWARE_WATCHPOINT
25531 @findex gdb.BP_HARDWARE_WATCHPOINT
25532 @item gdb.BP_HARDWARE_WATCHPOINT
25533 Hardware assisted watchpoint.
25535 @findex BP_READ_WATCHPOINT
25536 @findex gdb.BP_READ_WATCHPOINT
25537 @item gdb.BP_READ_WATCHPOINT
25538 Hardware assisted read watchpoint.
25540 @findex BP_ACCESS_WATCHPOINT
25541 @findex gdb.BP_ACCESS_WATCHPOINT
25542 @item gdb.BP_ACCESS_WATCHPOINT
25543 Hardware assisted access watchpoint.
25546 @defvar Breakpoint.hit_count
25547 This attribute holds the hit count for the breakpoint, an integer.
25548 This attribute is writable, but currently it can only be set to zero.
25551 @defvar Breakpoint.location
25552 This attribute holds the location of the breakpoint, as specified by
25553 the user. It is a string. If the breakpoint does not have a location
25554 (that is, it is a watchpoint) the attribute's value is @code{None}. This
25555 attribute is not writable.
25558 @defvar Breakpoint.expression
25559 This attribute holds a breakpoint expression, as specified by
25560 the user. It is a string. If the breakpoint does not have an
25561 expression (the breakpoint is not a watchpoint) the attribute's value
25562 is @code{None}. This attribute is not writable.
25565 @defvar Breakpoint.condition
25566 This attribute holds the condition of the breakpoint, as specified by
25567 the user. It is a string. If there is no condition, this attribute's
25568 value is @code{None}. This attribute is writable.
25571 @defvar Breakpoint.commands
25572 This attribute holds the commands attached to the breakpoint. If
25573 there are commands, this attribute's value is a string holding all the
25574 commands, separated by newlines. If there are no commands, this
25575 attribute is @code{None}. This attribute is not writable.
25578 @node Finish Breakpoints in Python
25579 @subsubsection Finish Breakpoints
25581 @cindex python finish breakpoints
25582 @tindex gdb.FinishBreakpoint
25584 A finish breakpoint is a temporary breakpoint set at the return address of
25585 a frame, based on the @code{finish} command. @code{gdb.FinishBreakpoint}
25586 extends @code{gdb.Breakpoint}. The underlying breakpoint will be disabled
25587 and deleted when the execution will run out of the breakpoint scope (i.e.@:
25588 @code{Breakpoint.stop} or @code{FinishBreakpoint.out_of_scope} triggered).
25589 Finish breakpoints are thread specific and must be create with the right
25592 @defun FinishBreakpoint.__init__ (@r{[}frame@r{]} @r{[}, internal@r{]})
25593 Create a finish breakpoint at the return address of the @code{gdb.Frame}
25594 object @var{frame}. If @var{frame} is not provided, this defaults to the
25595 newest frame. The optional @var{internal} argument allows the breakpoint to
25596 become invisible to the user. @xref{Breakpoints In Python}, for further
25597 details about this argument.
25600 @defun FinishBreakpoint.out_of_scope (self)
25601 In some circumstances (e.g.@: @code{longjmp}, C@t{++} exceptions, @value{GDBN}
25602 @code{return} command, @dots{}), a function may not properly terminate, and
25603 thus never hit the finish breakpoint. When @value{GDBN} notices such a
25604 situation, the @code{out_of_scope} callback will be triggered.
25606 You may want to sub-class @code{gdb.FinishBreakpoint} and override this
25610 class MyFinishBreakpoint (gdb.FinishBreakpoint)
25612 print "normal finish"
25615 def out_of_scope ():
25616 print "abnormal finish"
25620 @defvar FinishBreakpoint.return_value
25621 When @value{GDBN} is stopped at a finish breakpoint and the frame
25622 used to build the @code{gdb.FinishBreakpoint} object had debug symbols, this
25623 attribute will contain a @code{gdb.Value} object corresponding to the return
25624 value of the function. The value will be @code{None} if the function return
25625 type is @code{void} or if the return value was not computable. This attribute
25629 @node Lazy Strings In Python
25630 @subsubsection Python representation of lazy strings.
25632 @cindex lazy strings in python
25633 @tindex gdb.LazyString
25635 A @dfn{lazy string} is a string whose contents is not retrieved or
25636 encoded until it is needed.
25638 A @code{gdb.LazyString} is represented in @value{GDBN} as an
25639 @code{address} that points to a region of memory, an @code{encoding}
25640 that will be used to encode that region of memory, and a @code{length}
25641 to delimit the region of memory that represents the string. The
25642 difference between a @code{gdb.LazyString} and a string wrapped within
25643 a @code{gdb.Value} is that a @code{gdb.LazyString} will be treated
25644 differently by @value{GDBN} when printing. A @code{gdb.LazyString} is
25645 retrieved and encoded during printing, while a @code{gdb.Value}
25646 wrapping a string is immediately retrieved and encoded on creation.
25648 A @code{gdb.LazyString} object has the following functions:
25650 @defun LazyString.value ()
25651 Convert the @code{gdb.LazyString} to a @code{gdb.Value}. This value
25652 will point to the string in memory, but will lose all the delayed
25653 retrieval, encoding and handling that @value{GDBN} applies to a
25654 @code{gdb.LazyString}.
25657 @defvar LazyString.address
25658 This attribute holds the address of the string. This attribute is not
25662 @defvar LazyString.length
25663 This attribute holds the length of the string in characters. If the
25664 length is -1, then the string will be fetched and encoded up to the
25665 first null of appropriate width. This attribute is not writable.
25668 @defvar LazyString.encoding
25669 This attribute holds the encoding that will be applied to the string
25670 when the string is printed by @value{GDBN}. If the encoding is not
25671 set, or contains an empty string, then @value{GDBN} will select the
25672 most appropriate encoding when the string is printed. This attribute
25676 @defvar LazyString.type
25677 This attribute holds the type that is represented by the lazy string's
25678 type. For a lazy string this will always be a pointer type. To
25679 resolve this to the lazy string's character type, use the type's
25680 @code{target} method. @xref{Types In Python}. This attribute is not
25684 @node Python Auto-loading
25685 @subsection Python Auto-loading
25686 @cindex Python auto-loading
25688 When a new object file is read (for example, due to the @code{file}
25689 command, or because the inferior has loaded a shared library),
25690 @value{GDBN} will look for Python support scripts in several ways:
25691 @file{@var{objfile}-gdb.py} (@pxref{objfile-gdb.py file})
25692 and @code{.debug_gdb_scripts} section
25693 (@pxref{dotdebug_gdb_scripts section}).
25695 The auto-loading feature is useful for supplying application-specific
25696 debugging commands and scripts.
25698 Auto-loading can be enabled or disabled,
25699 and the list of auto-loaded scripts can be printed.
25702 @anchor{set auto-load python-scripts}
25703 @kindex set auto-load python-scripts
25704 @item set auto-load python-scripts [on|off]
25705 Enable or disable the auto-loading of Python scripts.
25707 @anchor{show auto-load python-scripts}
25708 @kindex show auto-load python-scripts
25709 @item show auto-load python-scripts
25710 Show whether auto-loading of Python scripts is enabled or disabled.
25712 @anchor{info auto-load python-scripts}
25713 @kindex info auto-load python-scripts
25714 @cindex print list of auto-loaded Python scripts
25715 @item info auto-load python-scripts [@var{regexp}]
25716 Print the list of all Python scripts that @value{GDBN} auto-loaded.
25718 Also printed is the list of Python scripts that were mentioned in
25719 the @code{.debug_gdb_scripts} section and were not found
25720 (@pxref{dotdebug_gdb_scripts section}).
25721 This is useful because their names are not printed when @value{GDBN}
25722 tries to load them and fails. There may be many of them, and printing
25723 an error message for each one is problematic.
25725 If @var{regexp} is supplied only Python scripts with matching names are printed.
25730 (gdb) info auto-load python-scripts
25732 Yes py-section-script.py
25733 full name: /tmp/py-section-script.py
25734 No my-foo-pretty-printers.py
25738 When reading an auto-loaded file, @value{GDBN} sets the
25739 @dfn{current objfile}. This is available via the @code{gdb.current_objfile}
25740 function (@pxref{Objfiles In Python}). This can be useful for
25741 registering objfile-specific pretty-printers.
25744 * objfile-gdb.py file:: The @file{@var{objfile}-gdb.py} file
25745 * dotdebug_gdb_scripts section:: The @code{.debug_gdb_scripts} section
25746 * Which flavor to choose?::
25749 @node objfile-gdb.py file
25750 @subsubsection The @file{@var{objfile}-gdb.py} file
25751 @cindex @file{@var{objfile}-gdb.py}
25753 When a new object file is read, @value{GDBN} looks for
25754 a file named @file{@var{objfile}-gdb.py} (we call it @var{script-name} below),
25755 where @var{objfile} is the object file's real name, formed by ensuring
25756 that the file name is absolute, following all symlinks, and resolving
25757 @code{.} and @code{..} components. If this file exists and is
25758 readable, @value{GDBN} will evaluate it as a Python script.
25760 If this file does not exist, then @value{GDBN} will look for
25761 @var{script-name} file in all of the directories as specified below.
25763 Note that loading of this script file also requires accordingly configured
25764 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25767 @anchor{set auto-load scripts-directory}
25768 @kindex set auto-load scripts-directory
25769 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
25770 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
25771 may be delimited by the host platform path separator in use
25772 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
25774 Each entry here needs to be covered also by the security setting
25775 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
25777 @anchor{with-auto-load-dir}
25778 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
25779 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
25780 configuration option @option{--with-auto-load-dir}.
25782 Any reference to @file{$debugdir} will get replaced by
25783 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
25784 reference to @file{$datadir} will get replaced by @var{data-directory} which is
25785 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
25786 @file{$datadir} must be placed as a directory component --- either alone or
25787 delimited by @file{/} or @file{\} directory separators, depending on the host
25790 The list of directories uses path separator (@samp{:} on GNU and Unix
25791 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
25792 to the @env{PATH} environment variable.
25794 @anchor{show auto-load scripts-directory}
25795 @kindex show auto-load scripts-directory
25796 @item show auto-load scripts-directory
25797 Show @value{GDBN} auto-loaded scripts location.
25800 @value{GDBN} does not track which files it has already auto-loaded this way.
25801 @value{GDBN} will load the associated script every time the corresponding
25802 @var{objfile} is opened.
25803 So your @file{-gdb.py} file should be careful to avoid errors if it
25804 is evaluated more than once.
25806 @node dotdebug_gdb_scripts section
25807 @subsubsection The @code{.debug_gdb_scripts} section
25808 @cindex @code{.debug_gdb_scripts} section
25810 For systems using file formats like ELF and COFF,
25811 when @value{GDBN} loads a new object file
25812 it will look for a special section named @samp{.debug_gdb_scripts}.
25813 If this section exists, its contents is a list of names of scripts to load.
25815 @value{GDBN} will look for each specified script file first in the
25816 current directory and then along the source search path
25817 (@pxref{Source Path, ,Specifying Source Directories}),
25818 except that @file{$cdir} is not searched, since the compilation
25819 directory is not relevant to scripts.
25821 Entries can be placed in section @code{.debug_gdb_scripts} with,
25822 for example, this GCC macro:
25825 /* Note: The "MS" section flags are to remove duplicates. */
25826 #define DEFINE_GDB_SCRIPT(script_name) \
25828 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
25830 .asciz \"" script_name "\"\n\
25836 Then one can reference the macro in a header or source file like this:
25839 DEFINE_GDB_SCRIPT ("my-app-scripts.py")
25842 The script name may include directories if desired.
25844 Note that loading of this script file also requires accordingly configured
25845 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25847 If the macro is put in a header, any application or library
25848 using this header will get a reference to the specified script.
25850 @node Which flavor to choose?
25851 @subsubsection Which flavor to choose?
25853 Given the multiple ways of auto-loading Python scripts, it might not always
25854 be clear which one to choose. This section provides some guidance.
25856 Benefits of the @file{-gdb.py} way:
25860 Can be used with file formats that don't support multiple sections.
25863 Ease of finding scripts for public libraries.
25865 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
25866 in the source search path.
25867 For publicly installed libraries, e.g., @file{libstdc++}, there typically
25868 isn't a source directory in which to find the script.
25871 Doesn't require source code additions.
25874 Benefits of the @code{.debug_gdb_scripts} way:
25878 Works with static linking.
25880 Scripts for libraries done the @file{-gdb.py} way require an objfile to
25881 trigger their loading. When an application is statically linked the only
25882 objfile available is the executable, and it is cumbersome to attach all the
25883 scripts from all the input libraries to the executable's @file{-gdb.py} script.
25886 Works with classes that are entirely inlined.
25888 Some classes can be entirely inlined, and thus there may not be an associated
25889 shared library to attach a @file{-gdb.py} script to.
25892 Scripts needn't be copied out of the source tree.
25894 In some circumstances, apps can be built out of large collections of internal
25895 libraries, and the build infrastructure necessary to install the
25896 @file{-gdb.py} scripts in a place where @value{GDBN} can find them is
25897 cumbersome. It may be easier to specify the scripts in the
25898 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
25899 top of the source tree to the source search path.
25902 @node Python modules
25903 @subsection Python modules
25904 @cindex python modules
25906 @value{GDBN} comes with several modules to assist writing Python code.
25909 * gdb.printing:: Building and registering pretty-printers.
25910 * gdb.types:: Utilities for working with types.
25911 * gdb.prompt:: Utilities for prompt value substitution.
25915 @subsubsection gdb.printing
25916 @cindex gdb.printing
25918 This module provides a collection of utilities for working with
25922 @item PrettyPrinter (@var{name}, @var{subprinters}=None)
25923 This class specifies the API that makes @samp{info pretty-printer},
25924 @samp{enable pretty-printer} and @samp{disable pretty-printer} work.
25925 Pretty-printers should generally inherit from this class.
25927 @item SubPrettyPrinter (@var{name})
25928 For printers that handle multiple types, this class specifies the
25929 corresponding API for the subprinters.
25931 @item RegexpCollectionPrettyPrinter (@var{name})
25932 Utility class for handling multiple printers, all recognized via
25933 regular expressions.
25934 @xref{Writing a Pretty-Printer}, for an example.
25936 @item FlagEnumerationPrinter (@var{name})
25937 A pretty-printer which handles printing of @code{enum} values. Unlike
25938 @value{GDBN}'s built-in @code{enum} printing, this printer attempts to
25939 work properly when there is some overlap between the enumeration
25940 constants. @var{name} is the name of the printer and also the name of
25941 the @code{enum} type to look up.
25943 @item register_pretty_printer (@var{obj}, @var{printer}, @var{replace}=False)
25944 Register @var{printer} with the pretty-printer list of @var{obj}.
25945 If @var{replace} is @code{True} then any existing copy of the printer
25946 is replaced. Otherwise a @code{RuntimeError} exception is raised
25947 if a printer with the same name already exists.
25951 @subsubsection gdb.types
25954 This module provides a collection of utilities for working with
25955 @code{gdb.Types} objects.
25958 @item get_basic_type (@var{type})
25959 Return @var{type} with const and volatile qualifiers stripped,
25960 and with typedefs and C@t{++} references converted to the underlying type.
25965 typedef const int const_int;
25967 const_int& foo_ref (foo);
25968 int main () @{ return 0; @}
25975 (gdb) python import gdb.types
25976 (gdb) python foo_ref = gdb.parse_and_eval("foo_ref")
25977 (gdb) python print gdb.types.get_basic_type(foo_ref.type)
25981 @item has_field (@var{type}, @var{field})
25982 Return @code{True} if @var{type}, assumed to be a type with fields
25983 (e.g., a structure or union), has field @var{field}.
25985 @item make_enum_dict (@var{enum_type})
25986 Return a Python @code{dictionary} type produced from @var{enum_type}.
25988 @item deep_items (@var{type})
25989 Returns a Python iterator similar to the standard
25990 @code{gdb.Type.iteritems} method, except that the iterator returned
25991 by @code{deep_items} will recursively traverse anonymous struct or
25992 union fields. For example:
26006 Then in @value{GDBN}:
26008 (@value{GDBP}) python import gdb.types
26009 (@value{GDBP}) python struct_a = gdb.lookup_type("struct A")
26010 (@value{GDBP}) python print struct_a.keys ()
26012 (@value{GDBP}) python print [k for k,v in gdb.types.deep_items(struct_a)]
26013 @{['a', 'b0', 'b1']@}
26019 @subsubsection gdb.prompt
26022 This module provides a method for prompt value-substitution.
26025 @item substitute_prompt (@var{string})
26026 Return @var{string} with escape sequences substituted by values. Some
26027 escape sequences take arguments. You can specify arguments inside
26028 ``@{@}'' immediately following the escape sequence.
26030 The escape sequences you can pass to this function are:
26034 Substitute a backslash.
26036 Substitute an ESC character.
26038 Substitute the selected frame; an argument names a frame parameter.
26040 Substitute a newline.
26042 Substitute a parameter's value; the argument names the parameter.
26044 Substitute a carriage return.
26046 Substitute the selected thread; an argument names a thread parameter.
26048 Substitute the version of GDB.
26050 Substitute the current working directory.
26052 Begin a sequence of non-printing characters. These sequences are
26053 typically used with the ESC character, and are not counted in the string
26054 length. Example: ``\[\e[0;34m\](gdb)\[\e[0m\]'' will return a
26055 blue-colored ``(gdb)'' prompt where the length is five.
26057 End a sequence of non-printing characters.
26063 substitute_prompt (``frame: \f,
26064 print arguments: \p@{print frame-arguments@}'')
26067 @exdent will return the string:
26070 "frame: main, print arguments: scalars"
26075 @section Creating new spellings of existing commands
26076 @cindex aliases for commands
26078 It is often useful to define alternate spellings of existing commands.
26079 For example, if a new @value{GDBN} command defined in Python has
26080 a long name to type, it is handy to have an abbreviated version of it
26081 that involves less typing.
26083 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
26084 of the @samp{step} command even though it is otherwise an ambiguous
26085 abbreviation of other commands like @samp{set} and @samp{show}.
26087 Aliases are also used to provide shortened or more common versions
26088 of multi-word commands. For example, @value{GDBN} provides the
26089 @samp{tty} alias of the @samp{set inferior-tty} command.
26091 You can define a new alias with the @samp{alias} command.
26096 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
26100 @var{ALIAS} specifies the name of the new alias.
26101 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
26104 @var{COMMAND} specifies the name of an existing command
26105 that is being aliased.
26107 The @samp{-a} option specifies that the new alias is an abbreviation
26108 of the command. Abbreviations are not shown in command
26109 lists displayed by the @samp{help} command.
26111 The @samp{--} option specifies the end of options,
26112 and is useful when @var{ALIAS} begins with a dash.
26114 Here is a simple example showing how to make an abbreviation
26115 of a command so that there is less to type.
26116 Suppose you were tired of typing @samp{disas}, the current
26117 shortest unambiguous abbreviation of the @samp{disassemble} command
26118 and you wanted an even shorter version named @samp{di}.
26119 The following will accomplish this.
26122 (gdb) alias -a di = disas
26125 Note that aliases are different from user-defined commands.
26126 With a user-defined command, you also need to write documentation
26127 for it with the @samp{document} command.
26128 An alias automatically picks up the documentation of the existing command.
26130 Here is an example where we make @samp{elms} an abbreviation of
26131 @samp{elements} in the @samp{set print elements} command.
26132 This is to show that you can make an abbreviation of any part
26136 (gdb) alias -a set print elms = set print elements
26137 (gdb) alias -a show print elms = show print elements
26138 (gdb) set p elms 20
26140 Limit on string chars or array elements to print is 200.
26143 Note that if you are defining an alias of a @samp{set} command,
26144 and you want to have an alias for the corresponding @samp{show}
26145 command, then you need to define the latter separately.
26147 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
26148 @var{ALIAS}, just as they are normally.
26151 (gdb) alias -a set pr elms = set p ele
26154 Finally, here is an example showing the creation of a one word
26155 alias for a more complex command.
26156 This creates alias @samp{spe} of the command @samp{set print elements}.
26159 (gdb) alias spe = set print elements
26164 @chapter Command Interpreters
26165 @cindex command interpreters
26167 @value{GDBN} supports multiple command interpreters, and some command
26168 infrastructure to allow users or user interface writers to switch
26169 between interpreters or run commands in other interpreters.
26171 @value{GDBN} currently supports two command interpreters, the console
26172 interpreter (sometimes called the command-line interpreter or @sc{cli})
26173 and the machine interface interpreter (or @sc{gdb/mi}). This manual
26174 describes both of these interfaces in great detail.
26176 By default, @value{GDBN} will start with the console interpreter.
26177 However, the user may choose to start @value{GDBN} with another
26178 interpreter by specifying the @option{-i} or @option{--interpreter}
26179 startup options. Defined interpreters include:
26183 @cindex console interpreter
26184 The traditional console or command-line interpreter. This is the most often
26185 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
26186 @value{GDBN} will use this interpreter.
26189 @cindex mi interpreter
26190 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
26191 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
26192 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
26196 @cindex mi2 interpreter
26197 The current @sc{gdb/mi} interface.
26200 @cindex mi1 interpreter
26201 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
26205 @cindex invoke another interpreter
26206 The interpreter being used by @value{GDBN} may not be dynamically
26207 switched at runtime. Although possible, this could lead to a very
26208 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
26209 enters the command "interpreter-set console" in a console view,
26210 @value{GDBN} would switch to using the console interpreter, rendering
26211 the IDE inoperable!
26213 @kindex interpreter-exec
26214 Although you may only choose a single interpreter at startup, you may execute
26215 commands in any interpreter from the current interpreter using the appropriate
26216 command. If you are running the console interpreter, simply use the
26217 @code{interpreter-exec} command:
26220 interpreter-exec mi "-data-list-register-names"
26223 @sc{gdb/mi} has a similar command, although it is only available in versions of
26224 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
26227 @chapter @value{GDBN} Text User Interface
26229 @cindex Text User Interface
26232 * TUI Overview:: TUI overview
26233 * TUI Keys:: TUI key bindings
26234 * TUI Single Key Mode:: TUI single key mode
26235 * TUI Commands:: TUI-specific commands
26236 * TUI Configuration:: TUI configuration variables
26239 The @value{GDBN} Text User Interface (TUI) is a terminal
26240 interface which uses the @code{curses} library to show the source
26241 file, the assembly output, the program registers and @value{GDBN}
26242 commands in separate text windows. The TUI mode is supported only
26243 on platforms where a suitable version of the @code{curses} library
26246 The TUI mode is enabled by default when you invoke @value{GDBN} as
26247 @samp{@value{GDBP} -tui}.
26248 You can also switch in and out of TUI mode while @value{GDBN} runs by
26249 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
26250 @xref{TUI Keys, ,TUI Key Bindings}.
26253 @section TUI Overview
26255 In TUI mode, @value{GDBN} can display several text windows:
26259 This window is the @value{GDBN} command window with the @value{GDBN}
26260 prompt and the @value{GDBN} output. The @value{GDBN} input is still
26261 managed using readline.
26264 The source window shows the source file of the program. The current
26265 line and active breakpoints are displayed in this window.
26268 The assembly window shows the disassembly output of the program.
26271 This window shows the processor registers. Registers are highlighted
26272 when their values change.
26275 The source and assembly windows show the current program position
26276 by highlighting the current line and marking it with a @samp{>} marker.
26277 Breakpoints are indicated with two markers. The first marker
26278 indicates the breakpoint type:
26282 Breakpoint which was hit at least once.
26285 Breakpoint which was never hit.
26288 Hardware breakpoint which was hit at least once.
26291 Hardware breakpoint which was never hit.
26294 The second marker indicates whether the breakpoint is enabled or not:
26298 Breakpoint is enabled.
26301 Breakpoint is disabled.
26304 The source, assembly and register windows are updated when the current
26305 thread changes, when the frame changes, or when the program counter
26308 These windows are not all visible at the same time. The command
26309 window is always visible. The others can be arranged in several
26320 source and assembly,
26323 source and registers, or
26326 assembly and registers.
26329 A status line above the command window shows the following information:
26333 Indicates the current @value{GDBN} target.
26334 (@pxref{Targets, ,Specifying a Debugging Target}).
26337 Gives the current process or thread number.
26338 When no process is being debugged, this field is set to @code{No process}.
26341 Gives the current function name for the selected frame.
26342 The name is demangled if demangling is turned on (@pxref{Print Settings}).
26343 When there is no symbol corresponding to the current program counter,
26344 the string @code{??} is displayed.
26347 Indicates the current line number for the selected frame.
26348 When the current line number is not known, the string @code{??} is displayed.
26351 Indicates the current program counter address.
26355 @section TUI Key Bindings
26356 @cindex TUI key bindings
26358 The TUI installs several key bindings in the readline keymaps
26359 @ifset SYSTEM_READLINE
26360 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
26362 @ifclear SYSTEM_READLINE
26363 (@pxref{Command Line Editing}).
26365 The following key bindings are installed for both TUI mode and the
26366 @value{GDBN} standard mode.
26375 Enter or leave the TUI mode. When leaving the TUI mode,
26376 the curses window management stops and @value{GDBN} operates using
26377 its standard mode, writing on the terminal directly. When reentering
26378 the TUI mode, control is given back to the curses windows.
26379 The screen is then refreshed.
26383 Use a TUI layout with only one window. The layout will
26384 either be @samp{source} or @samp{assembly}. When the TUI mode
26385 is not active, it will switch to the TUI mode.
26387 Think of this key binding as the Emacs @kbd{C-x 1} binding.
26391 Use a TUI layout with at least two windows. When the current
26392 layout already has two windows, the next layout with two windows is used.
26393 When a new layout is chosen, one window will always be common to the
26394 previous layout and the new one.
26396 Think of it as the Emacs @kbd{C-x 2} binding.
26400 Change the active window. The TUI associates several key bindings
26401 (like scrolling and arrow keys) with the active window. This command
26402 gives the focus to the next TUI window.
26404 Think of it as the Emacs @kbd{C-x o} binding.
26408 Switch in and out of the TUI SingleKey mode that binds single
26409 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
26412 The following key bindings only work in the TUI mode:
26417 Scroll the active window one page up.
26421 Scroll the active window one page down.
26425 Scroll the active window one line up.
26429 Scroll the active window one line down.
26433 Scroll the active window one column left.
26437 Scroll the active window one column right.
26441 Refresh the screen.
26444 Because the arrow keys scroll the active window in the TUI mode, they
26445 are not available for their normal use by readline unless the command
26446 window has the focus. When another window is active, you must use
26447 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
26448 and @kbd{C-f} to control the command window.
26450 @node TUI Single Key Mode
26451 @section TUI Single Key Mode
26452 @cindex TUI single key mode
26454 The TUI also provides a @dfn{SingleKey} mode, which binds several
26455 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
26456 switch into this mode, where the following key bindings are used:
26459 @kindex c @r{(SingleKey TUI key)}
26463 @kindex d @r{(SingleKey TUI key)}
26467 @kindex f @r{(SingleKey TUI key)}
26471 @kindex n @r{(SingleKey TUI key)}
26475 @kindex q @r{(SingleKey TUI key)}
26477 exit the SingleKey mode.
26479 @kindex r @r{(SingleKey TUI key)}
26483 @kindex s @r{(SingleKey TUI key)}
26487 @kindex u @r{(SingleKey TUI key)}
26491 @kindex v @r{(SingleKey TUI key)}
26495 @kindex w @r{(SingleKey TUI key)}
26500 Other keys temporarily switch to the @value{GDBN} command prompt.
26501 The key that was pressed is inserted in the editing buffer so that
26502 it is possible to type most @value{GDBN} commands without interaction
26503 with the TUI SingleKey mode. Once the command is entered the TUI
26504 SingleKey mode is restored. The only way to permanently leave
26505 this mode is by typing @kbd{q} or @kbd{C-x s}.
26509 @section TUI-specific Commands
26510 @cindex TUI commands
26512 The TUI has specific commands to control the text windows.
26513 These commands are always available, even when @value{GDBN} is not in
26514 the TUI mode. When @value{GDBN} is in the standard mode, most
26515 of these commands will automatically switch to the TUI mode.
26517 Note that if @value{GDBN}'s @code{stdout} is not connected to a
26518 terminal, or @value{GDBN} has been started with the machine interface
26519 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
26520 these commands will fail with an error, because it would not be
26521 possible or desirable to enable curses window management.
26526 List and give the size of all displayed windows.
26530 Display the next layout.
26533 Display the previous layout.
26536 Display the source window only.
26539 Display the assembly window only.
26542 Display the source and assembly window.
26545 Display the register window together with the source or assembly window.
26549 Make the next window active for scrolling.
26552 Make the previous window active for scrolling.
26555 Make the source window active for scrolling.
26558 Make the assembly window active for scrolling.
26561 Make the register window active for scrolling.
26564 Make the command window active for scrolling.
26568 Refresh the screen. This is similar to typing @kbd{C-L}.
26570 @item tui reg float
26572 Show the floating point registers in the register window.
26574 @item tui reg general
26575 Show the general registers in the register window.
26578 Show the next register group. The list of register groups as well as
26579 their order is target specific. The predefined register groups are the
26580 following: @code{general}, @code{float}, @code{system}, @code{vector},
26581 @code{all}, @code{save}, @code{restore}.
26583 @item tui reg system
26584 Show the system registers in the register window.
26588 Update the source window and the current execution point.
26590 @item winheight @var{name} +@var{count}
26591 @itemx winheight @var{name} -@var{count}
26593 Change the height of the window @var{name} by @var{count}
26594 lines. Positive counts increase the height, while negative counts
26597 @item tabset @var{nchars}
26599 Set the width of tab stops to be @var{nchars} characters.
26602 @node TUI Configuration
26603 @section TUI Configuration Variables
26604 @cindex TUI configuration variables
26606 Several configuration variables control the appearance of TUI windows.
26609 @item set tui border-kind @var{kind}
26610 @kindex set tui border-kind
26611 Select the border appearance for the source, assembly and register windows.
26612 The possible values are the following:
26615 Use a space character to draw the border.
26618 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
26621 Use the Alternate Character Set to draw the border. The border is
26622 drawn using character line graphics if the terminal supports them.
26625 @item set tui border-mode @var{mode}
26626 @kindex set tui border-mode
26627 @itemx set tui active-border-mode @var{mode}
26628 @kindex set tui active-border-mode
26629 Select the display attributes for the borders of the inactive windows
26630 or the active window. The @var{mode} can be one of the following:
26633 Use normal attributes to display the border.
26639 Use reverse video mode.
26642 Use half bright mode.
26644 @item half-standout
26645 Use half bright and standout mode.
26648 Use extra bright or bold mode.
26650 @item bold-standout
26651 Use extra bright or bold and standout mode.
26656 @chapter Using @value{GDBN} under @sc{gnu} Emacs
26659 @cindex @sc{gnu} Emacs
26660 A special interface allows you to use @sc{gnu} Emacs to view (and
26661 edit) the source files for the program you are debugging with
26664 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
26665 executable file you want to debug as an argument. This command starts
26666 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
26667 created Emacs buffer.
26668 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
26670 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
26675 All ``terminal'' input and output goes through an Emacs buffer, called
26678 This applies both to @value{GDBN} commands and their output, and to the input
26679 and output done by the program you are debugging.
26681 This is useful because it means that you can copy the text of previous
26682 commands and input them again; you can even use parts of the output
26685 All the facilities of Emacs' Shell mode are available for interacting
26686 with your program. In particular, you can send signals the usual
26687 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
26691 @value{GDBN} displays source code through Emacs.
26693 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
26694 source file for that frame and puts an arrow (@samp{=>}) at the
26695 left margin of the current line. Emacs uses a separate buffer for
26696 source display, and splits the screen to show both your @value{GDBN} session
26699 Explicit @value{GDBN} @code{list} or search commands still produce output as
26700 usual, but you probably have no reason to use them from Emacs.
26703 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
26704 a graphical mode, enabled by default, which provides further buffers
26705 that can control the execution and describe the state of your program.
26706 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
26708 If you specify an absolute file name when prompted for the @kbd{M-x
26709 gdb} argument, then Emacs sets your current working directory to where
26710 your program resides. If you only specify the file name, then Emacs
26711 sets your current working directory to the directory associated
26712 with the previous buffer. In this case, @value{GDBN} may find your
26713 program by searching your environment's @code{PATH} variable, but on
26714 some operating systems it might not find the source. So, although the
26715 @value{GDBN} input and output session proceeds normally, the auxiliary
26716 buffer does not display the current source and line of execution.
26718 The initial working directory of @value{GDBN} is printed on the top
26719 line of the GUD buffer and this serves as a default for the commands
26720 that specify files for @value{GDBN} to operate on. @xref{Files,
26721 ,Commands to Specify Files}.
26723 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
26724 need to call @value{GDBN} by a different name (for example, if you
26725 keep several configurations around, with different names) you can
26726 customize the Emacs variable @code{gud-gdb-command-name} to run the
26729 In the GUD buffer, you can use these special Emacs commands in
26730 addition to the standard Shell mode commands:
26734 Describe the features of Emacs' GUD Mode.
26737 Execute to another source line, like the @value{GDBN} @code{step} command; also
26738 update the display window to show the current file and location.
26741 Execute to next source line in this function, skipping all function
26742 calls, like the @value{GDBN} @code{next} command. Then update the display window
26743 to show the current file and location.
26746 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
26747 display window accordingly.
26750 Execute until exit from the selected stack frame, like the @value{GDBN}
26751 @code{finish} command.
26754 Continue execution of your program, like the @value{GDBN} @code{continue}
26758 Go up the number of frames indicated by the numeric argument
26759 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
26760 like the @value{GDBN} @code{up} command.
26763 Go down the number of frames indicated by the numeric argument, like the
26764 @value{GDBN} @code{down} command.
26767 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
26768 tells @value{GDBN} to set a breakpoint on the source line point is on.
26770 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
26771 separate frame which shows a backtrace when the GUD buffer is current.
26772 Move point to any frame in the stack and type @key{RET} to make it
26773 become the current frame and display the associated source in the
26774 source buffer. Alternatively, click @kbd{Mouse-2} to make the
26775 selected frame become the current one. In graphical mode, the
26776 speedbar displays watch expressions.
26778 If you accidentally delete the source-display buffer, an easy way to get
26779 it back is to type the command @code{f} in the @value{GDBN} buffer, to
26780 request a frame display; when you run under Emacs, this recreates
26781 the source buffer if necessary to show you the context of the current
26784 The source files displayed in Emacs are in ordinary Emacs buffers
26785 which are visiting the source files in the usual way. You can edit
26786 the files with these buffers if you wish; but keep in mind that @value{GDBN}
26787 communicates with Emacs in terms of line numbers. If you add or
26788 delete lines from the text, the line numbers that @value{GDBN} knows cease
26789 to correspond properly with the code.
26791 A more detailed description of Emacs' interaction with @value{GDBN} is
26792 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
26795 @c The following dropped because Epoch is nonstandard. Reactivate
26796 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
26798 @kindex Emacs Epoch environment
26802 Version 18 of @sc{gnu} Emacs has a built-in window system
26803 called the @code{epoch}
26804 environment. Users of this environment can use a new command,
26805 @code{inspect} which performs identically to @code{print} except that
26806 each value is printed in its own window.
26811 @chapter The @sc{gdb/mi} Interface
26813 @unnumberedsec Function and Purpose
26815 @cindex @sc{gdb/mi}, its purpose
26816 @sc{gdb/mi} is a line based machine oriented text interface to
26817 @value{GDBN} and is activated by specifying using the
26818 @option{--interpreter} command line option (@pxref{Mode Options}). It
26819 is specifically intended to support the development of systems which
26820 use the debugger as just one small component of a larger system.
26822 This chapter is a specification of the @sc{gdb/mi} interface. It is written
26823 in the form of a reference manual.
26825 Note that @sc{gdb/mi} is still under construction, so some of the
26826 features described below are incomplete and subject to change
26827 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
26829 @unnumberedsec Notation and Terminology
26831 @cindex notational conventions, for @sc{gdb/mi}
26832 This chapter uses the following notation:
26836 @code{|} separates two alternatives.
26839 @code{[ @var{something} ]} indicates that @var{something} is optional:
26840 it may or may not be given.
26843 @code{( @var{group} )*} means that @var{group} inside the parentheses
26844 may repeat zero or more times.
26847 @code{( @var{group} )+} means that @var{group} inside the parentheses
26848 may repeat one or more times.
26851 @code{"@var{string}"} means a literal @var{string}.
26855 @heading Dependencies
26859 * GDB/MI General Design::
26860 * GDB/MI Command Syntax::
26861 * GDB/MI Compatibility with CLI::
26862 * GDB/MI Development and Front Ends::
26863 * GDB/MI Output Records::
26864 * GDB/MI Simple Examples::
26865 * GDB/MI Command Description Format::
26866 * GDB/MI Breakpoint Commands::
26867 * GDB/MI Program Context::
26868 * GDB/MI Thread Commands::
26869 * GDB/MI Ada Tasking Commands::
26870 * GDB/MI Program Execution::
26871 * GDB/MI Stack Manipulation::
26872 * GDB/MI Variable Objects::
26873 * GDB/MI Data Manipulation::
26874 * GDB/MI Tracepoint Commands::
26875 * GDB/MI Symbol Query::
26876 * GDB/MI File Commands::
26878 * GDB/MI Kod Commands::
26879 * GDB/MI Memory Overlay Commands::
26880 * GDB/MI Signal Handling Commands::
26882 * GDB/MI Target Manipulation::
26883 * GDB/MI File Transfer Commands::
26884 * GDB/MI Miscellaneous Commands::
26887 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26888 @node GDB/MI General Design
26889 @section @sc{gdb/mi} General Design
26890 @cindex GDB/MI General Design
26892 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
26893 parts---commands sent to @value{GDBN}, responses to those commands
26894 and notifications. Each command results in exactly one response,
26895 indicating either successful completion of the command, or an error.
26896 For the commands that do not resume the target, the response contains the
26897 requested information. For the commands that resume the target, the
26898 response only indicates whether the target was successfully resumed.
26899 Notifications is the mechanism for reporting changes in the state of the
26900 target, or in @value{GDBN} state, that cannot conveniently be associated with
26901 a command and reported as part of that command response.
26903 The important examples of notifications are:
26907 Exec notifications. These are used to report changes in
26908 target state---when a target is resumed, or stopped. It would not
26909 be feasible to include this information in response of resuming
26910 commands, because one resume commands can result in multiple events in
26911 different threads. Also, quite some time may pass before any event
26912 happens in the target, while a frontend needs to know whether the resuming
26913 command itself was successfully executed.
26916 Console output, and status notifications. Console output
26917 notifications are used to report output of CLI commands, as well as
26918 diagnostics for other commands. Status notifications are used to
26919 report the progress of a long-running operation. Naturally, including
26920 this information in command response would mean no output is produced
26921 until the command is finished, which is undesirable.
26924 General notifications. Commands may have various side effects on
26925 the @value{GDBN} or target state beyond their official purpose. For example,
26926 a command may change the selected thread. Although such changes can
26927 be included in command response, using notification allows for more
26928 orthogonal frontend design.
26932 There's no guarantee that whenever an MI command reports an error,
26933 @value{GDBN} or the target are in any specific state, and especially,
26934 the state is not reverted to the state before the MI command was
26935 processed. Therefore, whenever an MI command results in an error,
26936 we recommend that the frontend refreshes all the information shown in
26937 the user interface.
26941 * Context management::
26942 * Asynchronous and non-stop modes::
26946 @node Context management
26947 @subsection Context management
26949 In most cases when @value{GDBN} accesses the target, this access is
26950 done in context of a specific thread and frame (@pxref{Frames}).
26951 Often, even when accessing global data, the target requires that a thread
26952 be specified. The CLI interface maintains the selected thread and frame,
26953 and supplies them to target on each command. This is convenient,
26954 because a command line user would not want to specify that information
26955 explicitly on each command, and because user interacts with
26956 @value{GDBN} via a single terminal, so no confusion is possible as
26957 to what thread and frame are the current ones.
26959 In the case of MI, the concept of selected thread and frame is less
26960 useful. First, a frontend can easily remember this information
26961 itself. Second, a graphical frontend can have more than one window,
26962 each one used for debugging a different thread, and the frontend might
26963 want to access additional threads for internal purposes. This
26964 increases the risk that by relying on implicitly selected thread, the
26965 frontend may be operating on a wrong one. Therefore, each MI command
26966 should explicitly specify which thread and frame to operate on. To
26967 make it possible, each MI command accepts the @samp{--thread} and
26968 @samp{--frame} options, the value to each is @value{GDBN} identifier
26969 for thread and frame to operate on.
26971 Usually, each top-level window in a frontend allows the user to select
26972 a thread and a frame, and remembers the user selection for further
26973 operations. However, in some cases @value{GDBN} may suggest that the
26974 current thread be changed. For example, when stopping on a breakpoint
26975 it is reasonable to switch to the thread where breakpoint is hit. For
26976 another example, if the user issues the CLI @samp{thread} command via
26977 the frontend, it is desirable to change the frontend's selected thread to the
26978 one specified by user. @value{GDBN} communicates the suggestion to
26979 change current thread using the @samp{=thread-selected} notification.
26980 No such notification is available for the selected frame at the moment.
26982 Note that historically, MI shares the selected thread with CLI, so
26983 frontends used the @code{-thread-select} to execute commands in the
26984 right context. However, getting this to work right is cumbersome. The
26985 simplest way is for frontend to emit @code{-thread-select} command
26986 before every command. This doubles the number of commands that need
26987 to be sent. The alternative approach is to suppress @code{-thread-select}
26988 if the selected thread in @value{GDBN} is supposed to be identical to the
26989 thread the frontend wants to operate on. However, getting this
26990 optimization right can be tricky. In particular, if the frontend
26991 sends several commands to @value{GDBN}, and one of the commands changes the
26992 selected thread, then the behaviour of subsequent commands will
26993 change. So, a frontend should either wait for response from such
26994 problematic commands, or explicitly add @code{-thread-select} for
26995 all subsequent commands. No frontend is known to do this exactly
26996 right, so it is suggested to just always pass the @samp{--thread} and
26997 @samp{--frame} options.
26999 @node Asynchronous and non-stop modes
27000 @subsection Asynchronous command execution and non-stop mode
27002 On some targets, @value{GDBN} is capable of processing MI commands
27003 even while the target is running. This is called @dfn{asynchronous
27004 command execution} (@pxref{Background Execution}). The frontend may
27005 specify a preferrence for asynchronous execution using the
27006 @code{-gdb-set target-async 1} command, which should be emitted before
27007 either running the executable or attaching to the target. After the
27008 frontend has started the executable or attached to the target, it can
27009 find if asynchronous execution is enabled using the
27010 @code{-list-target-features} command.
27012 Even if @value{GDBN} can accept a command while target is running,
27013 many commands that access the target do not work when the target is
27014 running. Therefore, asynchronous command execution is most useful
27015 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
27016 it is possible to examine the state of one thread, while other threads
27019 When a given thread is running, MI commands that try to access the
27020 target in the context of that thread may not work, or may work only on
27021 some targets. In particular, commands that try to operate on thread's
27022 stack will not work, on any target. Commands that read memory, or
27023 modify breakpoints, may work or not work, depending on the target. Note
27024 that even commands that operate on global state, such as @code{print},
27025 @code{set}, and breakpoint commands, still access the target in the
27026 context of a specific thread, so frontend should try to find a
27027 stopped thread and perform the operation on that thread (using the
27028 @samp{--thread} option).
27030 Which commands will work in the context of a running thread is
27031 highly target dependent. However, the two commands
27032 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
27033 to find the state of a thread, will always work.
27035 @node Thread groups
27036 @subsection Thread groups
27037 @value{GDBN} may be used to debug several processes at the same time.
27038 On some platfroms, @value{GDBN} may support debugging of several
27039 hardware systems, each one having several cores with several different
27040 processes running on each core. This section describes the MI
27041 mechanism to support such debugging scenarios.
27043 The key observation is that regardless of the structure of the
27044 target, MI can have a global list of threads, because most commands that
27045 accept the @samp{--thread} option do not need to know what process that
27046 thread belongs to. Therefore, it is not necessary to introduce
27047 neither additional @samp{--process} option, nor an notion of the
27048 current process in the MI interface. The only strictly new feature
27049 that is required is the ability to find how the threads are grouped
27052 To allow the user to discover such grouping, and to support arbitrary
27053 hierarchy of machines/cores/processes, MI introduces the concept of a
27054 @dfn{thread group}. Thread group is a collection of threads and other
27055 thread groups. A thread group always has a string identifier, a type,
27056 and may have additional attributes specific to the type. A new
27057 command, @code{-list-thread-groups}, returns the list of top-level
27058 thread groups, which correspond to processes that @value{GDBN} is
27059 debugging at the moment. By passing an identifier of a thread group
27060 to the @code{-list-thread-groups} command, it is possible to obtain
27061 the members of specific thread group.
27063 To allow the user to easily discover processes, and other objects, he
27064 wishes to debug, a concept of @dfn{available thread group} is
27065 introduced. Available thread group is an thread group that
27066 @value{GDBN} is not debugging, but that can be attached to, using the
27067 @code{-target-attach} command. The list of available top-level thread
27068 groups can be obtained using @samp{-list-thread-groups --available}.
27069 In general, the content of a thread group may be only retrieved only
27070 after attaching to that thread group.
27072 Thread groups are related to inferiors (@pxref{Inferiors and
27073 Programs}). Each inferior corresponds to a thread group of a special
27074 type @samp{process}, and some additional operations are permitted on
27075 such thread groups.
27077 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27078 @node GDB/MI Command Syntax
27079 @section @sc{gdb/mi} Command Syntax
27082 * GDB/MI Input Syntax::
27083 * GDB/MI Output Syntax::
27086 @node GDB/MI Input Syntax
27087 @subsection @sc{gdb/mi} Input Syntax
27089 @cindex input syntax for @sc{gdb/mi}
27090 @cindex @sc{gdb/mi}, input syntax
27092 @item @var{command} @expansion{}
27093 @code{@var{cli-command} | @var{mi-command}}
27095 @item @var{cli-command} @expansion{}
27096 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
27097 @var{cli-command} is any existing @value{GDBN} CLI command.
27099 @item @var{mi-command} @expansion{}
27100 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
27101 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
27103 @item @var{token} @expansion{}
27104 "any sequence of digits"
27106 @item @var{option} @expansion{}
27107 @code{"-" @var{parameter} [ " " @var{parameter} ]}
27109 @item @var{parameter} @expansion{}
27110 @code{@var{non-blank-sequence} | @var{c-string}}
27112 @item @var{operation} @expansion{}
27113 @emph{any of the operations described in this chapter}
27115 @item @var{non-blank-sequence} @expansion{}
27116 @emph{anything, provided it doesn't contain special characters such as
27117 "-", @var{nl}, """ and of course " "}
27119 @item @var{c-string} @expansion{}
27120 @code{""" @var{seven-bit-iso-c-string-content} """}
27122 @item @var{nl} @expansion{}
27131 The CLI commands are still handled by the @sc{mi} interpreter; their
27132 output is described below.
27135 The @code{@var{token}}, when present, is passed back when the command
27139 Some @sc{mi} commands accept optional arguments as part of the parameter
27140 list. Each option is identified by a leading @samp{-} (dash) and may be
27141 followed by an optional argument parameter. Options occur first in the
27142 parameter list and can be delimited from normal parameters using
27143 @samp{--} (this is useful when some parameters begin with a dash).
27150 We want easy access to the existing CLI syntax (for debugging).
27153 We want it to be easy to spot a @sc{mi} operation.
27156 @node GDB/MI Output Syntax
27157 @subsection @sc{gdb/mi} Output Syntax
27159 @cindex output syntax of @sc{gdb/mi}
27160 @cindex @sc{gdb/mi}, output syntax
27161 The output from @sc{gdb/mi} consists of zero or more out-of-band records
27162 followed, optionally, by a single result record. This result record
27163 is for the most recent command. The sequence of output records is
27164 terminated by @samp{(gdb)}.
27166 If an input command was prefixed with a @code{@var{token}} then the
27167 corresponding output for that command will also be prefixed by that same
27171 @item @var{output} @expansion{}
27172 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
27174 @item @var{result-record} @expansion{}
27175 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
27177 @item @var{out-of-band-record} @expansion{}
27178 @code{@var{async-record} | @var{stream-record}}
27180 @item @var{async-record} @expansion{}
27181 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
27183 @item @var{exec-async-output} @expansion{}
27184 @code{[ @var{token} ] "*" @var{async-output}}
27186 @item @var{status-async-output} @expansion{}
27187 @code{[ @var{token} ] "+" @var{async-output}}
27189 @item @var{notify-async-output} @expansion{}
27190 @code{[ @var{token} ] "=" @var{async-output}}
27192 @item @var{async-output} @expansion{}
27193 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
27195 @item @var{result-class} @expansion{}
27196 @code{"done" | "running" | "connected" | "error" | "exit"}
27198 @item @var{async-class} @expansion{}
27199 @code{"stopped" | @var{others}} (where @var{others} will be added
27200 depending on the needs---this is still in development).
27202 @item @var{result} @expansion{}
27203 @code{ @var{variable} "=" @var{value}}
27205 @item @var{variable} @expansion{}
27206 @code{ @var{string} }
27208 @item @var{value} @expansion{}
27209 @code{ @var{const} | @var{tuple} | @var{list} }
27211 @item @var{const} @expansion{}
27212 @code{@var{c-string}}
27214 @item @var{tuple} @expansion{}
27215 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
27217 @item @var{list} @expansion{}
27218 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
27219 @var{result} ( "," @var{result} )* "]" }
27221 @item @var{stream-record} @expansion{}
27222 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
27224 @item @var{console-stream-output} @expansion{}
27225 @code{"~" @var{c-string}}
27227 @item @var{target-stream-output} @expansion{}
27228 @code{"@@" @var{c-string}}
27230 @item @var{log-stream-output} @expansion{}
27231 @code{"&" @var{c-string}}
27233 @item @var{nl} @expansion{}
27236 @item @var{token} @expansion{}
27237 @emph{any sequence of digits}.
27245 All output sequences end in a single line containing a period.
27248 The @code{@var{token}} is from the corresponding request. Note that
27249 for all async output, while the token is allowed by the grammar and
27250 may be output by future versions of @value{GDBN} for select async
27251 output messages, it is generally omitted. Frontends should treat
27252 all async output as reporting general changes in the state of the
27253 target and there should be no need to associate async output to any
27257 @cindex status output in @sc{gdb/mi}
27258 @var{status-async-output} contains on-going status information about the
27259 progress of a slow operation. It can be discarded. All status output is
27260 prefixed by @samp{+}.
27263 @cindex async output in @sc{gdb/mi}
27264 @var{exec-async-output} contains asynchronous state change on the target
27265 (stopped, started, disappeared). All async output is prefixed by
27269 @cindex notify output in @sc{gdb/mi}
27270 @var{notify-async-output} contains supplementary information that the
27271 client should handle (e.g., a new breakpoint information). All notify
27272 output is prefixed by @samp{=}.
27275 @cindex console output in @sc{gdb/mi}
27276 @var{console-stream-output} is output that should be displayed as is in the
27277 console. It is the textual response to a CLI command. All the console
27278 output is prefixed by @samp{~}.
27281 @cindex target output in @sc{gdb/mi}
27282 @var{target-stream-output} is the output produced by the target program.
27283 All the target output is prefixed by @samp{@@}.
27286 @cindex log output in @sc{gdb/mi}
27287 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
27288 instance messages that should be displayed as part of an error log. All
27289 the log output is prefixed by @samp{&}.
27292 @cindex list output in @sc{gdb/mi}
27293 New @sc{gdb/mi} commands should only output @var{lists} containing
27299 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
27300 details about the various output records.
27302 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27303 @node GDB/MI Compatibility with CLI
27304 @section @sc{gdb/mi} Compatibility with CLI
27306 @cindex compatibility, @sc{gdb/mi} and CLI
27307 @cindex @sc{gdb/mi}, compatibility with CLI
27309 For the developers convenience CLI commands can be entered directly,
27310 but there may be some unexpected behaviour. For example, commands
27311 that query the user will behave as if the user replied yes, breakpoint
27312 command lists are not executed and some CLI commands, such as
27313 @code{if}, @code{when} and @code{define}, prompt for further input with
27314 @samp{>}, which is not valid MI output.
27316 This feature may be removed at some stage in the future and it is
27317 recommended that front ends use the @code{-interpreter-exec} command
27318 (@pxref{-interpreter-exec}).
27320 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27321 @node GDB/MI Development and Front Ends
27322 @section @sc{gdb/mi} Development and Front Ends
27323 @cindex @sc{gdb/mi} development
27325 The application which takes the MI output and presents the state of the
27326 program being debugged to the user is called a @dfn{front end}.
27328 Although @sc{gdb/mi} is still incomplete, it is currently being used
27329 by a variety of front ends to @value{GDBN}. This makes it difficult
27330 to introduce new functionality without breaking existing usage. This
27331 section tries to minimize the problems by describing how the protocol
27334 Some changes in MI need not break a carefully designed front end, and
27335 for these the MI version will remain unchanged. The following is a
27336 list of changes that may occur within one level, so front ends should
27337 parse MI output in a way that can handle them:
27341 New MI commands may be added.
27344 New fields may be added to the output of any MI command.
27347 The range of values for fields with specified values, e.g.,
27348 @code{in_scope} (@pxref{-var-update}) may be extended.
27350 @c The format of field's content e.g type prefix, may change so parse it
27351 @c at your own risk. Yes, in general?
27353 @c The order of fields may change? Shouldn't really matter but it might
27354 @c resolve inconsistencies.
27357 If the changes are likely to break front ends, the MI version level
27358 will be increased by one. This will allow the front end to parse the
27359 output according to the MI version. Apart from mi0, new versions of
27360 @value{GDBN} will not support old versions of MI and it will be the
27361 responsibility of the front end to work with the new one.
27363 @c Starting with mi3, add a new command -mi-version that prints the MI
27366 The best way to avoid unexpected changes in MI that might break your front
27367 end is to make your project known to @value{GDBN} developers and
27368 follow development on @email{gdb@@sourceware.org} and
27369 @email{gdb-patches@@sourceware.org}.
27370 @cindex mailing lists
27372 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27373 @node GDB/MI Output Records
27374 @section @sc{gdb/mi} Output Records
27377 * GDB/MI Result Records::
27378 * GDB/MI Stream Records::
27379 * GDB/MI Async Records::
27380 * GDB/MI Frame Information::
27381 * GDB/MI Thread Information::
27382 * GDB/MI Ada Exception Information::
27385 @node GDB/MI Result Records
27386 @subsection @sc{gdb/mi} Result Records
27388 @cindex result records in @sc{gdb/mi}
27389 @cindex @sc{gdb/mi}, result records
27390 In addition to a number of out-of-band notifications, the response to a
27391 @sc{gdb/mi} command includes one of the following result indications:
27395 @item "^done" [ "," @var{results} ]
27396 The synchronous operation was successful, @code{@var{results}} are the return
27401 This result record is equivalent to @samp{^done}. Historically, it
27402 was output instead of @samp{^done} if the command has resumed the
27403 target. This behaviour is maintained for backward compatibility, but
27404 all frontends should treat @samp{^done} and @samp{^running}
27405 identically and rely on the @samp{*running} output record to determine
27406 which threads are resumed.
27410 @value{GDBN} has connected to a remote target.
27412 @item "^error" "," @var{c-string}
27414 The operation failed. The @code{@var{c-string}} contains the corresponding
27419 @value{GDBN} has terminated.
27423 @node GDB/MI Stream Records
27424 @subsection @sc{gdb/mi} Stream Records
27426 @cindex @sc{gdb/mi}, stream records
27427 @cindex stream records in @sc{gdb/mi}
27428 @value{GDBN} internally maintains a number of output streams: the console, the
27429 target, and the log. The output intended for each of these streams is
27430 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
27432 Each stream record begins with a unique @dfn{prefix character} which
27433 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
27434 Syntax}). In addition to the prefix, each stream record contains a
27435 @code{@var{string-output}}. This is either raw text (with an implicit new
27436 line) or a quoted C string (which does not contain an implicit newline).
27439 @item "~" @var{string-output}
27440 The console output stream contains text that should be displayed in the
27441 CLI console window. It contains the textual responses to CLI commands.
27443 @item "@@" @var{string-output}
27444 The target output stream contains any textual output from the running
27445 target. This is only present when GDB's event loop is truly
27446 asynchronous, which is currently only the case for remote targets.
27448 @item "&" @var{string-output}
27449 The log stream contains debugging messages being produced by @value{GDBN}'s
27453 @node GDB/MI Async Records
27454 @subsection @sc{gdb/mi} Async Records
27456 @cindex async records in @sc{gdb/mi}
27457 @cindex @sc{gdb/mi}, async records
27458 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
27459 additional changes that have occurred. Those changes can either be a
27460 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
27461 target activity (e.g., target stopped).
27463 The following is the list of possible async records:
27467 @item *running,thread-id="@var{thread}"
27468 The target is now running. The @var{thread} field tells which
27469 specific thread is now running, and can be @samp{all} if all threads
27470 are running. The frontend should assume that no interaction with a
27471 running thread is possible after this notification is produced.
27472 The frontend should not assume that this notification is output
27473 only once for any command. @value{GDBN} may emit this notification
27474 several times, either for different threads, because it cannot resume
27475 all threads together, or even for a single thread, if the thread must
27476 be stepped though some code before letting it run freely.
27478 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
27479 The target has stopped. The @var{reason} field can have one of the
27483 @item breakpoint-hit
27484 A breakpoint was reached.
27485 @item watchpoint-trigger
27486 A watchpoint was triggered.
27487 @item read-watchpoint-trigger
27488 A read watchpoint was triggered.
27489 @item access-watchpoint-trigger
27490 An access watchpoint was triggered.
27491 @item function-finished
27492 An -exec-finish or similar CLI command was accomplished.
27493 @item location-reached
27494 An -exec-until or similar CLI command was accomplished.
27495 @item watchpoint-scope
27496 A watchpoint has gone out of scope.
27497 @item end-stepping-range
27498 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
27499 similar CLI command was accomplished.
27500 @item exited-signalled
27501 The inferior exited because of a signal.
27503 The inferior exited.
27504 @item exited-normally
27505 The inferior exited normally.
27506 @item signal-received
27507 A signal was received by the inferior.
27509 The inferior has stopped due to a library being loaded or unloaded.
27510 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
27511 set or when a @code{catch load} or @code{catch unload} catchpoint is
27512 in use (@pxref{Set Catchpoints}).
27514 The inferior has forked. This is reported when @code{catch fork}
27515 (@pxref{Set Catchpoints}) has been used.
27517 The inferior has vforked. This is reported in when @code{catch vfork}
27518 (@pxref{Set Catchpoints}) has been used.
27519 @item syscall-entry
27520 The inferior entered a system call. This is reported when @code{catch
27521 syscall} (@pxref{Set Catchpoints}) has been used.
27522 @item syscall-entry
27523 The inferior returned from a system call. This is reported when
27524 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
27526 The inferior called @code{exec}. This is reported when @code{catch exec}
27527 (@pxref{Set Catchpoints}) has been used.
27530 The @var{id} field identifies the thread that directly caused the stop
27531 -- for example by hitting a breakpoint. Depending on whether all-stop
27532 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
27533 stop all threads, or only the thread that directly triggered the stop.
27534 If all threads are stopped, the @var{stopped} field will have the
27535 value of @code{"all"}. Otherwise, the value of the @var{stopped}
27536 field will be a list of thread identifiers. Presently, this list will
27537 always include a single thread, but frontend should be prepared to see
27538 several threads in the list. The @var{core} field reports the
27539 processor core on which the stop event has happened. This field may be absent
27540 if such information is not available.
27542 @item =thread-group-added,id="@var{id}"
27543 @itemx =thread-group-removed,id="@var{id}"
27544 A thread group was either added or removed. The @var{id} field
27545 contains the @value{GDBN} identifier of the thread group. When a thread
27546 group is added, it generally might not be associated with a running
27547 process. When a thread group is removed, its id becomes invalid and
27548 cannot be used in any way.
27550 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
27551 A thread group became associated with a running program,
27552 either because the program was just started or the thread group
27553 was attached to a program. The @var{id} field contains the
27554 @value{GDBN} identifier of the thread group. The @var{pid} field
27555 contains process identifier, specific to the operating system.
27557 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
27558 A thread group is no longer associated with a running program,
27559 either because the program has exited, or because it was detached
27560 from. The @var{id} field contains the @value{GDBN} identifier of the
27561 thread group. @var{code} is the exit code of the inferior; it exists
27562 only when the inferior exited with some code.
27564 @item =thread-created,id="@var{id}",group-id="@var{gid}"
27565 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
27566 A thread either was created, or has exited. The @var{id} field
27567 contains the @value{GDBN} identifier of the thread. The @var{gid}
27568 field identifies the thread group this thread belongs to.
27570 @item =thread-selected,id="@var{id}"
27571 Informs that the selected thread was changed as result of the last
27572 command. This notification is not emitted as result of @code{-thread-select}
27573 command but is emitted whenever an MI command that is not documented
27574 to change the selected thread actually changes it. In particular,
27575 invoking, directly or indirectly (via user-defined command), the CLI
27576 @code{thread} command, will generate this notification.
27578 We suggest that in response to this notification, front ends
27579 highlight the selected thread and cause subsequent commands to apply to
27582 @item =library-loaded,...
27583 Reports that a new library file was loaded by the program. This
27584 notification has 4 fields---@var{id}, @var{target-name},
27585 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
27586 opaque identifier of the library. For remote debugging case,
27587 @var{target-name} and @var{host-name} fields give the name of the
27588 library file on the target, and on the host respectively. For native
27589 debugging, both those fields have the same value. The
27590 @var{symbols-loaded} field is emitted only for backward compatibility
27591 and should not be relied on to convey any useful information. The
27592 @var{thread-group} field, if present, specifies the id of the thread
27593 group in whose context the library was loaded. If the field is
27594 absent, it means the library was loaded in the context of all present
27597 @item =library-unloaded,...
27598 Reports that a library was unloaded by the program. This notification
27599 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
27600 the same meaning as for the @code{=library-loaded} notification.
27601 The @var{thread-group} field, if present, specifies the id of the
27602 thread group in whose context the library was unloaded. If the field is
27603 absent, it means the library was unloaded in the context of all present
27606 @item =breakpoint-created,bkpt=@{...@}
27607 @itemx =breakpoint-modified,bkpt=@{...@}
27608 @itemx =breakpoint-deleted,bkpt=@{...@}
27609 Reports that a breakpoint was created, modified, or deleted,
27610 respectively. Only user-visible breakpoints are reported to the MI
27613 The @var{bkpt} argument is of the same form as returned by the various
27614 breakpoint commands; @xref{GDB/MI Breakpoint Commands}.
27616 Note that if a breakpoint is emitted in the result record of a
27617 command, then it will not also be emitted in an async record.
27621 @node GDB/MI Frame Information
27622 @subsection @sc{gdb/mi} Frame Information
27624 Response from many MI commands includes an information about stack
27625 frame. This information is a tuple that may have the following
27630 The level of the stack frame. The innermost frame has the level of
27631 zero. This field is always present.
27634 The name of the function corresponding to the frame. This field may
27635 be absent if @value{GDBN} is unable to determine the function name.
27638 The code address for the frame. This field is always present.
27641 The name of the source files that correspond to the frame's code
27642 address. This field may be absent.
27645 The source line corresponding to the frames' code address. This field
27649 The name of the binary file (either executable or shared library) the
27650 corresponds to the frame's code address. This field may be absent.
27654 @node GDB/MI Thread Information
27655 @subsection @sc{gdb/mi} Thread Information
27657 Whenever @value{GDBN} has to report an information about a thread, it
27658 uses a tuple with the following fields:
27662 The numeric id assigned to the thread by @value{GDBN}. This field is
27666 Target-specific string identifying the thread. This field is always present.
27669 Additional information about the thread provided by the target.
27670 It is supposed to be human-readable and not interpreted by the
27671 frontend. This field is optional.
27674 Either @samp{stopped} or @samp{running}, depending on whether the
27675 thread is presently running. This field is always present.
27678 The value of this field is an integer number of the processor core the
27679 thread was last seen on. This field is optional.
27682 @node GDB/MI Ada Exception Information
27683 @subsection @sc{gdb/mi} Ada Exception Information
27685 Whenever a @code{*stopped} record is emitted because the program
27686 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
27687 @value{GDBN} provides the name of the exception that was raised via
27688 the @code{exception-name} field.
27690 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27691 @node GDB/MI Simple Examples
27692 @section Simple Examples of @sc{gdb/mi} Interaction
27693 @cindex @sc{gdb/mi}, simple examples
27695 This subsection presents several simple examples of interaction using
27696 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
27697 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
27698 the output received from @sc{gdb/mi}.
27700 Note the line breaks shown in the examples are here only for
27701 readability, they don't appear in the real output.
27703 @subheading Setting a Breakpoint
27705 Setting a breakpoint generates synchronous output which contains detailed
27706 information of the breakpoint.
27709 -> -break-insert main
27710 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
27711 enabled="y",addr="0x08048564",func="main",file="myprog.c",
27712 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
27716 @subheading Program Execution
27718 Program execution generates asynchronous records and MI gives the
27719 reason that execution stopped.
27725 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
27726 frame=@{addr="0x08048564",func="main",
27727 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
27728 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
27733 <- *stopped,reason="exited-normally"
27737 @subheading Quitting @value{GDBN}
27739 Quitting @value{GDBN} just prints the result class @samp{^exit}.
27747 Please note that @samp{^exit} is printed immediately, but it might
27748 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
27749 performs necessary cleanups, including killing programs being debugged
27750 or disconnecting from debug hardware, so the frontend should wait till
27751 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
27752 fails to exit in reasonable time.
27754 @subheading A Bad Command
27756 Here's what happens if you pass a non-existent command:
27760 <- ^error,msg="Undefined MI command: rubbish"
27765 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27766 @node GDB/MI Command Description Format
27767 @section @sc{gdb/mi} Command Description Format
27769 The remaining sections describe blocks of commands. Each block of
27770 commands is laid out in a fashion similar to this section.
27772 @subheading Motivation
27774 The motivation for this collection of commands.
27776 @subheading Introduction
27778 A brief introduction to this collection of commands as a whole.
27780 @subheading Commands
27782 For each command in the block, the following is described:
27784 @subsubheading Synopsis
27787 -command @var{args}@dots{}
27790 @subsubheading Result
27792 @subsubheading @value{GDBN} Command
27794 The corresponding @value{GDBN} CLI command(s), if any.
27796 @subsubheading Example
27798 Example(s) formatted for readability. Some of the described commands have
27799 not been implemented yet and these are labeled N.A.@: (not available).
27802 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27803 @node GDB/MI Breakpoint Commands
27804 @section @sc{gdb/mi} Breakpoint Commands
27806 @cindex breakpoint commands for @sc{gdb/mi}
27807 @cindex @sc{gdb/mi}, breakpoint commands
27808 This section documents @sc{gdb/mi} commands for manipulating
27811 @subheading The @code{-break-after} Command
27812 @findex -break-after
27814 @subsubheading Synopsis
27817 -break-after @var{number} @var{count}
27820 The breakpoint number @var{number} is not in effect until it has been
27821 hit @var{count} times. To see how this is reflected in the output of
27822 the @samp{-break-list} command, see the description of the
27823 @samp{-break-list} command below.
27825 @subsubheading @value{GDBN} Command
27827 The corresponding @value{GDBN} command is @samp{ignore}.
27829 @subsubheading Example
27834 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
27835 enabled="y",addr="0x000100d0",func="main",file="hello.c",
27836 fullname="/home/foo/hello.c",line="5",times="0"@}
27843 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27844 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27845 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27846 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27847 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27848 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27849 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27850 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27851 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27852 line="5",times="0",ignore="3"@}]@}
27857 @subheading The @code{-break-catch} Command
27858 @findex -break-catch
27861 @subheading The @code{-break-commands} Command
27862 @findex -break-commands
27864 @subsubheading Synopsis
27867 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
27870 Specifies the CLI commands that should be executed when breakpoint
27871 @var{number} is hit. The parameters @var{command1} to @var{commandN}
27872 are the commands. If no command is specified, any previously-set
27873 commands are cleared. @xref{Break Commands}. Typical use of this
27874 functionality is tracing a program, that is, printing of values of
27875 some variables whenever breakpoint is hit and then continuing.
27877 @subsubheading @value{GDBN} Command
27879 The corresponding @value{GDBN} command is @samp{commands}.
27881 @subsubheading Example
27886 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
27887 enabled="y",addr="0x000100d0",func="main",file="hello.c",
27888 fullname="/home/foo/hello.c",line="5",times="0"@}
27890 -break-commands 1 "print v" "continue"
27895 @subheading The @code{-break-condition} Command
27896 @findex -break-condition
27898 @subsubheading Synopsis
27901 -break-condition @var{number} @var{expr}
27904 Breakpoint @var{number} will stop the program only if the condition in
27905 @var{expr} is true. The condition becomes part of the
27906 @samp{-break-list} output (see the description of the @samp{-break-list}
27909 @subsubheading @value{GDBN} Command
27911 The corresponding @value{GDBN} command is @samp{condition}.
27913 @subsubheading Example
27917 -break-condition 1 1
27921 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27922 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27923 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27924 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27925 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27926 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27927 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27928 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27929 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27930 line="5",cond="1",times="0",ignore="3"@}]@}
27934 @subheading The @code{-break-delete} Command
27935 @findex -break-delete
27937 @subsubheading Synopsis
27940 -break-delete ( @var{breakpoint} )+
27943 Delete the breakpoint(s) whose number(s) are specified in the argument
27944 list. This is obviously reflected in the breakpoint list.
27946 @subsubheading @value{GDBN} Command
27948 The corresponding @value{GDBN} command is @samp{delete}.
27950 @subsubheading Example
27958 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
27959 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27960 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27961 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27962 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27963 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27964 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27969 @subheading The @code{-break-disable} Command
27970 @findex -break-disable
27972 @subsubheading Synopsis
27975 -break-disable ( @var{breakpoint} )+
27978 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
27979 break list is now set to @samp{n} for the named @var{breakpoint}(s).
27981 @subsubheading @value{GDBN} Command
27983 The corresponding @value{GDBN} command is @samp{disable}.
27985 @subsubheading Example
27993 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27994 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27995 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27996 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27997 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27998 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27999 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28000 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
28001 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
28002 line="5",times="0"@}]@}
28006 @subheading The @code{-break-enable} Command
28007 @findex -break-enable
28009 @subsubheading Synopsis
28012 -break-enable ( @var{breakpoint} )+
28015 Enable (previously disabled) @var{breakpoint}(s).
28017 @subsubheading @value{GDBN} Command
28019 The corresponding @value{GDBN} command is @samp{enable}.
28021 @subsubheading Example
28029 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
28030 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28031 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28032 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28033 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28034 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28035 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28036 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
28037 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
28038 line="5",times="0"@}]@}
28042 @subheading The @code{-break-info} Command
28043 @findex -break-info
28045 @subsubheading Synopsis
28048 -break-info @var{breakpoint}
28052 Get information about a single breakpoint.
28054 @subsubheading @value{GDBN} Command
28056 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
28058 @subsubheading Example
28061 @subheading The @code{-break-insert} Command
28062 @findex -break-insert
28064 @subsubheading Synopsis
28067 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
28068 [ -c @var{condition} ] [ -i @var{ignore-count} ]
28069 [ -p @var{thread-id} ] [ @var{location} ]
28073 If specified, @var{location}, can be one of:
28080 @item filename:linenum
28081 @item filename:function
28085 The possible optional parameters of this command are:
28089 Insert a temporary breakpoint.
28091 Insert a hardware breakpoint.
28093 If @var{location} cannot be parsed (for example if it
28094 refers to unknown files or functions), create a pending
28095 breakpoint. Without this flag, @value{GDBN} will report
28096 an error, and won't create a breakpoint, if @var{location}
28099 Create a disabled breakpoint.
28101 Create a tracepoint. @xref{Tracepoints}. When this parameter
28102 is used together with @samp{-h}, a fast tracepoint is created.
28103 @item -c @var{condition}
28104 Make the breakpoint conditional on @var{condition}.
28105 @item -i @var{ignore-count}
28106 Initialize the @var{ignore-count}.
28107 @item -p @var{thread-id}
28108 Restrict the breakpoint to the specified @var{thread-id}.
28111 @subsubheading Result
28113 The result is in the form:
28116 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
28117 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
28118 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
28119 times="@var{times}"@}
28123 where @var{number} is the @value{GDBN} number for this breakpoint,
28124 @var{funcname} is the name of the function where the breakpoint was
28125 inserted, @var{filename} is the name of the source file which contains
28126 this function, @var{lineno} is the source line number within that file
28127 and @var{times} the number of times that the breakpoint has been hit
28128 (always 0 for -break-insert but may be greater for -break-info or -break-list
28129 which use the same output).
28131 Note: this format is open to change.
28132 @c An out-of-band breakpoint instead of part of the result?
28134 @subsubheading @value{GDBN} Command
28136 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
28137 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
28139 @subsubheading Example
28144 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
28145 fullname="/home/foo/recursive2.c,line="4",times="0"@}
28147 -break-insert -t foo
28148 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
28149 fullname="/home/foo/recursive2.c,line="11",times="0"@}
28152 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
28153 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28154 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28155 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28156 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28157 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28158 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28159 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28160 addr="0x0001072c", func="main",file="recursive2.c",
28161 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
28162 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
28163 addr="0x00010774",func="foo",file="recursive2.c",
28164 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
28166 @c -break-insert -r foo.*
28167 @c ~int foo(int, int);
28168 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
28169 @c "fullname="/home/foo/recursive2.c",line="11",times="0"@}
28173 @subheading The @code{-break-list} Command
28174 @findex -break-list
28176 @subsubheading Synopsis
28182 Displays the list of inserted breakpoints, showing the following fields:
28186 number of the breakpoint
28188 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
28190 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
28193 is the breakpoint enabled or no: @samp{y} or @samp{n}
28195 memory location at which the breakpoint is set
28197 logical location of the breakpoint, expressed by function name, file
28200 number of times the breakpoint has been hit
28203 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
28204 @code{body} field is an empty list.
28206 @subsubheading @value{GDBN} Command
28208 The corresponding @value{GDBN} command is @samp{info break}.
28210 @subsubheading Example
28215 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
28216 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28217 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28218 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28219 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28220 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28221 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28222 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28223 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
28224 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
28225 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
28226 line="13",times="0"@}]@}
28230 Here's an example of the result when there are no breakpoints:
28235 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
28236 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28237 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28238 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28239 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28240 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28241 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28246 @subheading The @code{-break-passcount} Command
28247 @findex -break-passcount
28249 @subsubheading Synopsis
28252 -break-passcount @var{tracepoint-number} @var{passcount}
28255 Set the passcount for tracepoint @var{tracepoint-number} to
28256 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
28257 is not a tracepoint, error is emitted. This corresponds to CLI
28258 command @samp{passcount}.
28260 @subheading The @code{-break-watch} Command
28261 @findex -break-watch
28263 @subsubheading Synopsis
28266 -break-watch [ -a | -r ]
28269 Create a watchpoint. With the @samp{-a} option it will create an
28270 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
28271 read from or on a write to the memory location. With the @samp{-r}
28272 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
28273 trigger only when the memory location is accessed for reading. Without
28274 either of the options, the watchpoint created is a regular watchpoint,
28275 i.e., it will trigger when the memory location is accessed for writing.
28276 @xref{Set Watchpoints, , Setting Watchpoints}.
28278 Note that @samp{-break-list} will report a single list of watchpoints and
28279 breakpoints inserted.
28281 @subsubheading @value{GDBN} Command
28283 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
28286 @subsubheading Example
28288 Setting a watchpoint on a variable in the @code{main} function:
28293 ^done,wpt=@{number="2",exp="x"@}
28298 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
28299 value=@{old="-268439212",new="55"@},
28300 frame=@{func="main",args=[],file="recursive2.c",
28301 fullname="/home/foo/bar/recursive2.c",line="5"@}
28305 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
28306 the program execution twice: first for the variable changing value, then
28307 for the watchpoint going out of scope.
28312 ^done,wpt=@{number="5",exp="C"@}
28317 *stopped,reason="watchpoint-trigger",
28318 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
28319 frame=@{func="callee4",args=[],
28320 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28321 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
28326 *stopped,reason="watchpoint-scope",wpnum="5",
28327 frame=@{func="callee3",args=[@{name="strarg",
28328 value="0x11940 \"A string argument.\""@}],
28329 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28330 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
28334 Listing breakpoints and watchpoints, at different points in the program
28335 execution. Note that once the watchpoint goes out of scope, it is
28341 ^done,wpt=@{number="2",exp="C"@}
28344 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
28345 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28346 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28347 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28348 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28349 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28350 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28351 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28352 addr="0x00010734",func="callee4",
28353 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28354 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
28355 bkpt=@{number="2",type="watchpoint",disp="keep",
28356 enabled="y",addr="",what="C",times="0"@}]@}
28361 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
28362 value=@{old="-276895068",new="3"@},
28363 frame=@{func="callee4",args=[],
28364 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28365 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
28368 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
28369 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28370 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28371 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28372 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28373 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28374 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28375 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28376 addr="0x00010734",func="callee4",
28377 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28378 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
28379 bkpt=@{number="2",type="watchpoint",disp="keep",
28380 enabled="y",addr="",what="C",times="-5"@}]@}
28384 ^done,reason="watchpoint-scope",wpnum="2",
28385 frame=@{func="callee3",args=[@{name="strarg",
28386 value="0x11940 \"A string argument.\""@}],
28387 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28388 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
28391 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
28392 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28393 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28394 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28395 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28396 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28397 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28398 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28399 addr="0x00010734",func="callee4",
28400 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28401 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
28406 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28407 @node GDB/MI Program Context
28408 @section @sc{gdb/mi} Program Context
28410 @subheading The @code{-exec-arguments} Command
28411 @findex -exec-arguments
28414 @subsubheading Synopsis
28417 -exec-arguments @var{args}
28420 Set the inferior program arguments, to be used in the next
28423 @subsubheading @value{GDBN} Command
28425 The corresponding @value{GDBN} command is @samp{set args}.
28427 @subsubheading Example
28431 -exec-arguments -v word
28438 @subheading The @code{-exec-show-arguments} Command
28439 @findex -exec-show-arguments
28441 @subsubheading Synopsis
28444 -exec-show-arguments
28447 Print the arguments of the program.
28449 @subsubheading @value{GDBN} Command
28451 The corresponding @value{GDBN} command is @samp{show args}.
28453 @subsubheading Example
28458 @subheading The @code{-environment-cd} Command
28459 @findex -environment-cd
28461 @subsubheading Synopsis
28464 -environment-cd @var{pathdir}
28467 Set @value{GDBN}'s working directory.
28469 @subsubheading @value{GDBN} Command
28471 The corresponding @value{GDBN} command is @samp{cd}.
28473 @subsubheading Example
28477 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
28483 @subheading The @code{-environment-directory} Command
28484 @findex -environment-directory
28486 @subsubheading Synopsis
28489 -environment-directory [ -r ] [ @var{pathdir} ]+
28492 Add directories @var{pathdir} to beginning of search path for source files.
28493 If the @samp{-r} option is used, the search path is reset to the default
28494 search path. If directories @var{pathdir} are supplied in addition to the
28495 @samp{-r} option, the search path is first reset and then addition
28497 Multiple directories may be specified, separated by blanks. Specifying
28498 multiple directories in a single command
28499 results in the directories added to the beginning of the
28500 search path in the same order they were presented in the command.
28501 If blanks are needed as
28502 part of a directory name, double-quotes should be used around
28503 the name. In the command output, the path will show up separated
28504 by the system directory-separator character. The directory-separator
28505 character must not be used
28506 in any directory name.
28507 If no directories are specified, the current search path is displayed.
28509 @subsubheading @value{GDBN} Command
28511 The corresponding @value{GDBN} command is @samp{dir}.
28513 @subsubheading Example
28517 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
28518 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
28520 -environment-directory ""
28521 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
28523 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
28524 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
28526 -environment-directory -r
28527 ^done,source-path="$cdir:$cwd"
28532 @subheading The @code{-environment-path} Command
28533 @findex -environment-path
28535 @subsubheading Synopsis
28538 -environment-path [ -r ] [ @var{pathdir} ]+
28541 Add directories @var{pathdir} to beginning of search path for object files.
28542 If the @samp{-r} option is used, the search path is reset to the original
28543 search path that existed at gdb start-up. If directories @var{pathdir} are
28544 supplied in addition to the
28545 @samp{-r} option, the search path is first reset and then addition
28547 Multiple directories may be specified, separated by blanks. Specifying
28548 multiple directories in a single command
28549 results in the directories added to the beginning of the
28550 search path in the same order they were presented in the command.
28551 If blanks are needed as
28552 part of a directory name, double-quotes should be used around
28553 the name. In the command output, the path will show up separated
28554 by the system directory-separator character. The directory-separator
28555 character must not be used
28556 in any directory name.
28557 If no directories are specified, the current path is displayed.
28560 @subsubheading @value{GDBN} Command
28562 The corresponding @value{GDBN} command is @samp{path}.
28564 @subsubheading Example
28569 ^done,path="/usr/bin"
28571 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
28572 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
28574 -environment-path -r /usr/local/bin
28575 ^done,path="/usr/local/bin:/usr/bin"
28580 @subheading The @code{-environment-pwd} Command
28581 @findex -environment-pwd
28583 @subsubheading Synopsis
28589 Show the current working directory.
28591 @subsubheading @value{GDBN} Command
28593 The corresponding @value{GDBN} command is @samp{pwd}.
28595 @subsubheading Example
28600 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
28604 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28605 @node GDB/MI Thread Commands
28606 @section @sc{gdb/mi} Thread Commands
28609 @subheading The @code{-thread-info} Command
28610 @findex -thread-info
28612 @subsubheading Synopsis
28615 -thread-info [ @var{thread-id} ]
28618 Reports information about either a specific thread, if
28619 the @var{thread-id} parameter is present, or about all
28620 threads. When printing information about all threads,
28621 also reports the current thread.
28623 @subsubheading @value{GDBN} Command
28625 The @samp{info thread} command prints the same information
28628 @subsubheading Result
28630 The result is a list of threads. The following attributes are
28631 defined for a given thread:
28635 This field exists only for the current thread. It has the value @samp{*}.
28638 The identifier that @value{GDBN} uses to refer to the thread.
28641 The identifier that the target uses to refer to the thread.
28644 Extra information about the thread, in a target-specific format. This
28648 The name of the thread. If the user specified a name using the
28649 @code{thread name} command, then this name is given. Otherwise, if
28650 @value{GDBN} can extract the thread name from the target, then that
28651 name is given. If @value{GDBN} cannot find the thread name, then this
28655 The stack frame currently executing in the thread.
28658 The thread's state. The @samp{state} field may have the following
28663 The thread is stopped. Frame information is available for stopped
28667 The thread is running. There's no frame information for running
28673 If @value{GDBN} can find the CPU core on which this thread is running,
28674 then this field is the core identifier. This field is optional.
28678 @subsubheading Example
28683 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
28684 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
28685 args=[]@},state="running"@},
28686 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
28687 frame=@{level="0",addr="0x0804891f",func="foo",
28688 args=[@{name="i",value="10"@}],
28689 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},
28690 state="running"@}],
28691 current-thread-id="1"
28695 @subheading The @code{-thread-list-ids} Command
28696 @findex -thread-list-ids
28698 @subsubheading Synopsis
28704 Produces a list of the currently known @value{GDBN} thread ids. At the
28705 end of the list it also prints the total number of such threads.
28707 This command is retained for historical reasons, the
28708 @code{-thread-info} command should be used instead.
28710 @subsubheading @value{GDBN} Command
28712 Part of @samp{info threads} supplies the same information.
28714 @subsubheading Example
28719 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
28720 current-thread-id="1",number-of-threads="3"
28725 @subheading The @code{-thread-select} Command
28726 @findex -thread-select
28728 @subsubheading Synopsis
28731 -thread-select @var{threadnum}
28734 Make @var{threadnum} the current thread. It prints the number of the new
28735 current thread, and the topmost frame for that thread.
28737 This command is deprecated in favor of explicitly using the
28738 @samp{--thread} option to each command.
28740 @subsubheading @value{GDBN} Command
28742 The corresponding @value{GDBN} command is @samp{thread}.
28744 @subsubheading Example
28751 *stopped,reason="end-stepping-range",thread-id="2",line="187",
28752 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
28756 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
28757 number-of-threads="3"
28760 ^done,new-thread-id="3",
28761 frame=@{level="0",func="vprintf",
28762 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
28763 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
28767 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28768 @node GDB/MI Ada Tasking Commands
28769 @section @sc{gdb/mi} Ada Tasking Commands
28771 @subheading The @code{-ada-task-info} Command
28772 @findex -ada-task-info
28774 @subsubheading Synopsis
28777 -ada-task-info [ @var{task-id} ]
28780 Reports information about either a specific Ada task, if the
28781 @var{task-id} parameter is present, or about all Ada tasks.
28783 @subsubheading @value{GDBN} Command
28785 The @samp{info tasks} command prints the same information
28786 about all Ada tasks (@pxref{Ada Tasks}).
28788 @subsubheading Result
28790 The result is a table of Ada tasks. The following columns are
28791 defined for each Ada task:
28795 This field exists only for the current thread. It has the value @samp{*}.
28798 The identifier that @value{GDBN} uses to refer to the Ada task.
28801 The identifier that the target uses to refer to the Ada task.
28804 The identifier of the thread corresponding to the Ada task.
28806 This field should always exist, as Ada tasks are always implemented
28807 on top of a thread. But if @value{GDBN} cannot find this corresponding
28808 thread for any reason, the field is omitted.
28811 This field exists only when the task was created by another task.
28812 In this case, it provides the ID of the parent task.
28815 The base priority of the task.
28818 The current state of the task. For a detailed description of the
28819 possible states, see @ref{Ada Tasks}.
28822 The name of the task.
28826 @subsubheading Example
28830 ^done,tasks=@{nr_rows="3",nr_cols="8",
28831 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
28832 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
28833 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
28834 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
28835 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
28836 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
28837 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
28838 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
28839 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
28840 state="Child Termination Wait",name="main_task"@}]@}
28844 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28845 @node GDB/MI Program Execution
28846 @section @sc{gdb/mi} Program Execution
28848 These are the asynchronous commands which generate the out-of-band
28849 record @samp{*stopped}. Currently @value{GDBN} only really executes
28850 asynchronously with remote targets and this interaction is mimicked in
28853 @subheading The @code{-exec-continue} Command
28854 @findex -exec-continue
28856 @subsubheading Synopsis
28859 -exec-continue [--reverse] [--all|--thread-group N]
28862 Resumes the execution of the inferior program, which will continue
28863 to execute until it reaches a debugger stop event. If the
28864 @samp{--reverse} option is specified, execution resumes in reverse until
28865 it reaches a stop event. Stop events may include
28868 breakpoints or watchpoints
28870 signals or exceptions
28872 the end of the process (or its beginning under @samp{--reverse})
28874 the end or beginning of a replay log if one is being used.
28876 In all-stop mode (@pxref{All-Stop
28877 Mode}), may resume only one thread, or all threads, depending on the
28878 value of the @samp{scheduler-locking} variable. If @samp{--all} is
28879 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
28880 ignored in all-stop mode. If the @samp{--thread-group} options is
28881 specified, then all threads in that thread group are resumed.
28883 @subsubheading @value{GDBN} Command
28885 The corresponding @value{GDBN} corresponding is @samp{continue}.
28887 @subsubheading Example
28894 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
28895 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
28901 @subheading The @code{-exec-finish} Command
28902 @findex -exec-finish
28904 @subsubheading Synopsis
28907 -exec-finish [--reverse]
28910 Resumes the execution of the inferior program until the current
28911 function is exited. Displays the results returned by the function.
28912 If the @samp{--reverse} option is specified, resumes the reverse
28913 execution of the inferior program until the point where current
28914 function was called.
28916 @subsubheading @value{GDBN} Command
28918 The corresponding @value{GDBN} command is @samp{finish}.
28920 @subsubheading Example
28922 Function returning @code{void}.
28929 *stopped,reason="function-finished",frame=@{func="main",args=[],
28930 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
28934 Function returning other than @code{void}. The name of the internal
28935 @value{GDBN} variable storing the result is printed, together with the
28942 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
28943 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
28944 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28945 gdb-result-var="$1",return-value="0"
28950 @subheading The @code{-exec-interrupt} Command
28951 @findex -exec-interrupt
28953 @subsubheading Synopsis
28956 -exec-interrupt [--all|--thread-group N]
28959 Interrupts the background execution of the target. Note how the token
28960 associated with the stop message is the one for the execution command
28961 that has been interrupted. The token for the interrupt itself only
28962 appears in the @samp{^done} output. If the user is trying to
28963 interrupt a non-running program, an error message will be printed.
28965 Note that when asynchronous execution is enabled, this command is
28966 asynchronous just like other execution commands. That is, first the
28967 @samp{^done} response will be printed, and the target stop will be
28968 reported after that using the @samp{*stopped} notification.
28970 In non-stop mode, only the context thread is interrupted by default.
28971 All threads (in all inferiors) will be interrupted if the
28972 @samp{--all} option is specified. If the @samp{--thread-group}
28973 option is specified, all threads in that group will be interrupted.
28975 @subsubheading @value{GDBN} Command
28977 The corresponding @value{GDBN} command is @samp{interrupt}.
28979 @subsubheading Example
28990 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
28991 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
28992 fullname="/home/foo/bar/try.c",line="13"@}
28997 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
29001 @subheading The @code{-exec-jump} Command
29004 @subsubheading Synopsis
29007 -exec-jump @var{location}
29010 Resumes execution of the inferior program at the location specified by
29011 parameter. @xref{Specify Location}, for a description of the
29012 different forms of @var{location}.
29014 @subsubheading @value{GDBN} Command
29016 The corresponding @value{GDBN} command is @samp{jump}.
29018 @subsubheading Example
29021 -exec-jump foo.c:10
29022 *running,thread-id="all"
29027 @subheading The @code{-exec-next} Command
29030 @subsubheading Synopsis
29033 -exec-next [--reverse]
29036 Resumes execution of the inferior program, stopping when the beginning
29037 of the next source line is reached.
29039 If the @samp{--reverse} option is specified, resumes reverse execution
29040 of the inferior program, stopping at the beginning of the previous
29041 source line. If you issue this command on the first line of a
29042 function, it will take you back to the caller of that function, to the
29043 source line where the function was called.
29046 @subsubheading @value{GDBN} Command
29048 The corresponding @value{GDBN} command is @samp{next}.
29050 @subsubheading Example
29056 *stopped,reason="end-stepping-range",line="8",file="hello.c"
29061 @subheading The @code{-exec-next-instruction} Command
29062 @findex -exec-next-instruction
29064 @subsubheading Synopsis
29067 -exec-next-instruction [--reverse]
29070 Executes one machine instruction. If the instruction is a function
29071 call, continues until the function returns. If the program stops at an
29072 instruction in the middle of a source line, the address will be
29075 If the @samp{--reverse} option is specified, resumes reverse execution
29076 of the inferior program, stopping at the previous instruction. If the
29077 previously executed instruction was a return from another function,
29078 it will continue to execute in reverse until the call to that function
29079 (from the current stack frame) is reached.
29081 @subsubheading @value{GDBN} Command
29083 The corresponding @value{GDBN} command is @samp{nexti}.
29085 @subsubheading Example
29089 -exec-next-instruction
29093 *stopped,reason="end-stepping-range",
29094 addr="0x000100d4",line="5",file="hello.c"
29099 @subheading The @code{-exec-return} Command
29100 @findex -exec-return
29102 @subsubheading Synopsis
29108 Makes current function return immediately. Doesn't execute the inferior.
29109 Displays the new current frame.
29111 @subsubheading @value{GDBN} Command
29113 The corresponding @value{GDBN} command is @samp{return}.
29115 @subsubheading Example
29119 200-break-insert callee4
29120 200^done,bkpt=@{number="1",addr="0x00010734",
29121 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
29126 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
29127 frame=@{func="callee4",args=[],
29128 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29129 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
29135 111^done,frame=@{level="0",func="callee3",
29136 args=[@{name="strarg",
29137 value="0x11940 \"A string argument.\""@}],
29138 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29139 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
29144 @subheading The @code{-exec-run} Command
29147 @subsubheading Synopsis
29150 -exec-run [--all | --thread-group N]
29153 Starts execution of the inferior from the beginning. The inferior
29154 executes until either a breakpoint is encountered or the program
29155 exits. In the latter case the output will include an exit code, if
29156 the program has exited exceptionally.
29158 When no option is specified, the current inferior is started. If the
29159 @samp{--thread-group} option is specified, it should refer to a thread
29160 group of type @samp{process}, and that thread group will be started.
29161 If the @samp{--all} option is specified, then all inferiors will be started.
29163 @subsubheading @value{GDBN} Command
29165 The corresponding @value{GDBN} command is @samp{run}.
29167 @subsubheading Examples
29172 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
29177 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
29178 frame=@{func="main",args=[],file="recursive2.c",
29179 fullname="/home/foo/bar/recursive2.c",line="4"@}
29184 Program exited normally:
29192 *stopped,reason="exited-normally"
29197 Program exited exceptionally:
29205 *stopped,reason="exited",exit-code="01"
29209 Another way the program can terminate is if it receives a signal such as
29210 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
29214 *stopped,reason="exited-signalled",signal-name="SIGINT",
29215 signal-meaning="Interrupt"
29219 @c @subheading -exec-signal
29222 @subheading The @code{-exec-step} Command
29225 @subsubheading Synopsis
29228 -exec-step [--reverse]
29231 Resumes execution of the inferior program, stopping when the beginning
29232 of the next source line is reached, if the next source line is not a
29233 function call. If it is, stop at the first instruction of the called
29234 function. If the @samp{--reverse} option is specified, resumes reverse
29235 execution of the inferior program, stopping at the beginning of the
29236 previously executed source line.
29238 @subsubheading @value{GDBN} Command
29240 The corresponding @value{GDBN} command is @samp{step}.
29242 @subsubheading Example
29244 Stepping into a function:
29250 *stopped,reason="end-stepping-range",
29251 frame=@{func="foo",args=[@{name="a",value="10"@},
29252 @{name="b",value="0"@}],file="recursive2.c",
29253 fullname="/home/foo/bar/recursive2.c",line="11"@}
29263 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
29268 @subheading The @code{-exec-step-instruction} Command
29269 @findex -exec-step-instruction
29271 @subsubheading Synopsis
29274 -exec-step-instruction [--reverse]
29277 Resumes the inferior which executes one machine instruction. If the
29278 @samp{--reverse} option is specified, resumes reverse execution of the
29279 inferior program, stopping at the previously executed instruction.
29280 The output, once @value{GDBN} has stopped, will vary depending on
29281 whether we have stopped in the middle of a source line or not. In the
29282 former case, the address at which the program stopped will be printed
29285 @subsubheading @value{GDBN} Command
29287 The corresponding @value{GDBN} command is @samp{stepi}.
29289 @subsubheading Example
29293 -exec-step-instruction
29297 *stopped,reason="end-stepping-range",
29298 frame=@{func="foo",args=[],file="try.c",
29299 fullname="/home/foo/bar/try.c",line="10"@}
29301 -exec-step-instruction
29305 *stopped,reason="end-stepping-range",
29306 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
29307 fullname="/home/foo/bar/try.c",line="10"@}
29312 @subheading The @code{-exec-until} Command
29313 @findex -exec-until
29315 @subsubheading Synopsis
29318 -exec-until [ @var{location} ]
29321 Executes the inferior until the @var{location} specified in the
29322 argument is reached. If there is no argument, the inferior executes
29323 until a source line greater than the current one is reached. The
29324 reason for stopping in this case will be @samp{location-reached}.
29326 @subsubheading @value{GDBN} Command
29328 The corresponding @value{GDBN} command is @samp{until}.
29330 @subsubheading Example
29334 -exec-until recursive2.c:6
29338 *stopped,reason="location-reached",frame=@{func="main",args=[],
29339 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
29344 @subheading -file-clear
29345 Is this going away????
29348 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29349 @node GDB/MI Stack Manipulation
29350 @section @sc{gdb/mi} Stack Manipulation Commands
29353 @subheading The @code{-stack-info-frame} Command
29354 @findex -stack-info-frame
29356 @subsubheading Synopsis
29362 Get info on the selected frame.
29364 @subsubheading @value{GDBN} Command
29366 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
29367 (without arguments).
29369 @subsubheading Example
29374 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
29375 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29376 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
29380 @subheading The @code{-stack-info-depth} Command
29381 @findex -stack-info-depth
29383 @subsubheading Synopsis
29386 -stack-info-depth [ @var{max-depth} ]
29389 Return the depth of the stack. If the integer argument @var{max-depth}
29390 is specified, do not count beyond @var{max-depth} frames.
29392 @subsubheading @value{GDBN} Command
29394 There's no equivalent @value{GDBN} command.
29396 @subsubheading Example
29398 For a stack with frame levels 0 through 11:
29405 -stack-info-depth 4
29408 -stack-info-depth 12
29411 -stack-info-depth 11
29414 -stack-info-depth 13
29419 @subheading The @code{-stack-list-arguments} Command
29420 @findex -stack-list-arguments
29422 @subsubheading Synopsis
29425 -stack-list-arguments @var{print-values}
29426 [ @var{low-frame} @var{high-frame} ]
29429 Display a list of the arguments for the frames between @var{low-frame}
29430 and @var{high-frame} (inclusive). If @var{low-frame} and
29431 @var{high-frame} are not provided, list the arguments for the whole
29432 call stack. If the two arguments are equal, show the single frame
29433 at the corresponding level. It is an error if @var{low-frame} is
29434 larger than the actual number of frames. On the other hand,
29435 @var{high-frame} may be larger than the actual number of frames, in
29436 which case only existing frames will be returned.
29438 If @var{print-values} is 0 or @code{--no-values}, print only the names of
29439 the variables; if it is 1 or @code{--all-values}, print also their
29440 values; and if it is 2 or @code{--simple-values}, print the name,
29441 type and value for simple data types, and the name and type for arrays,
29442 structures and unions.
29444 Use of this command to obtain arguments in a single frame is
29445 deprecated in favor of the @samp{-stack-list-variables} command.
29447 @subsubheading @value{GDBN} Command
29449 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
29450 @samp{gdb_get_args} command which partially overlaps with the
29451 functionality of @samp{-stack-list-arguments}.
29453 @subsubheading Example
29460 frame=@{level="0",addr="0x00010734",func="callee4",
29461 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29462 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
29463 frame=@{level="1",addr="0x0001076c",func="callee3",
29464 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29465 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
29466 frame=@{level="2",addr="0x0001078c",func="callee2",
29467 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29468 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
29469 frame=@{level="3",addr="0x000107b4",func="callee1",
29470 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29471 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
29472 frame=@{level="4",addr="0x000107e0",func="main",
29473 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29474 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
29476 -stack-list-arguments 0
29479 frame=@{level="0",args=[]@},
29480 frame=@{level="1",args=[name="strarg"]@},
29481 frame=@{level="2",args=[name="intarg",name="strarg"]@},
29482 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
29483 frame=@{level="4",args=[]@}]
29485 -stack-list-arguments 1
29488 frame=@{level="0",args=[]@},
29490 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
29491 frame=@{level="2",args=[
29492 @{name="intarg",value="2"@},
29493 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
29494 @{frame=@{level="3",args=[
29495 @{name="intarg",value="2"@},
29496 @{name="strarg",value="0x11940 \"A string argument.\""@},
29497 @{name="fltarg",value="3.5"@}]@},
29498 frame=@{level="4",args=[]@}]
29500 -stack-list-arguments 0 2 2
29501 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
29503 -stack-list-arguments 1 2 2
29504 ^done,stack-args=[frame=@{level="2",
29505 args=[@{name="intarg",value="2"@},
29506 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
29510 @c @subheading -stack-list-exception-handlers
29513 @subheading The @code{-stack-list-frames} Command
29514 @findex -stack-list-frames
29516 @subsubheading Synopsis
29519 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
29522 List the frames currently on the stack. For each frame it displays the
29527 The frame number, 0 being the topmost frame, i.e., the innermost function.
29529 The @code{$pc} value for that frame.
29533 File name of the source file where the function lives.
29534 @item @var{fullname}
29535 The full file name of the source file where the function lives.
29537 Line number corresponding to the @code{$pc}.
29539 The shared library where this function is defined. This is only given
29540 if the frame's function is not known.
29543 If invoked without arguments, this command prints a backtrace for the
29544 whole stack. If given two integer arguments, it shows the frames whose
29545 levels are between the two arguments (inclusive). If the two arguments
29546 are equal, it shows the single frame at the corresponding level. It is
29547 an error if @var{low-frame} is larger than the actual number of
29548 frames. On the other hand, @var{high-frame} may be larger than the
29549 actual number of frames, in which case only existing frames will be returned.
29551 @subsubheading @value{GDBN} Command
29553 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
29555 @subsubheading Example
29557 Full stack backtrace:
29563 [frame=@{level="0",addr="0x0001076c",func="foo",
29564 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
29565 frame=@{level="1",addr="0x000107a4",func="foo",
29566 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29567 frame=@{level="2",addr="0x000107a4",func="foo",
29568 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29569 frame=@{level="3",addr="0x000107a4",func="foo",
29570 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29571 frame=@{level="4",addr="0x000107a4",func="foo",
29572 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29573 frame=@{level="5",addr="0x000107a4",func="foo",
29574 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29575 frame=@{level="6",addr="0x000107a4",func="foo",
29576 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29577 frame=@{level="7",addr="0x000107a4",func="foo",
29578 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29579 frame=@{level="8",addr="0x000107a4",func="foo",
29580 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29581 frame=@{level="9",addr="0x000107a4",func="foo",
29582 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29583 frame=@{level="10",addr="0x000107a4",func="foo",
29584 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29585 frame=@{level="11",addr="0x00010738",func="main",
29586 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
29590 Show frames between @var{low_frame} and @var{high_frame}:
29594 -stack-list-frames 3 5
29596 [frame=@{level="3",addr="0x000107a4",func="foo",
29597 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29598 frame=@{level="4",addr="0x000107a4",func="foo",
29599 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29600 frame=@{level="5",addr="0x000107a4",func="foo",
29601 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
29605 Show a single frame:
29609 -stack-list-frames 3 3
29611 [frame=@{level="3",addr="0x000107a4",func="foo",
29612 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
29617 @subheading The @code{-stack-list-locals} Command
29618 @findex -stack-list-locals
29620 @subsubheading Synopsis
29623 -stack-list-locals @var{print-values}
29626 Display the local variable names for the selected frame. If
29627 @var{print-values} is 0 or @code{--no-values}, print only the names of
29628 the variables; if it is 1 or @code{--all-values}, print also their
29629 values; and if it is 2 or @code{--simple-values}, print the name,
29630 type and value for simple data types, and the name and type for arrays,
29631 structures and unions. In this last case, a frontend can immediately
29632 display the value of simple data types and create variable objects for
29633 other data types when the user wishes to explore their values in
29636 This command is deprecated in favor of the
29637 @samp{-stack-list-variables} command.
29639 @subsubheading @value{GDBN} Command
29641 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
29643 @subsubheading Example
29647 -stack-list-locals 0
29648 ^done,locals=[name="A",name="B",name="C"]
29650 -stack-list-locals --all-values
29651 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
29652 @{name="C",value="@{1, 2, 3@}"@}]
29653 -stack-list-locals --simple-values
29654 ^done,locals=[@{name="A",type="int",value="1"@},
29655 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
29659 @subheading The @code{-stack-list-variables} Command
29660 @findex -stack-list-variables
29662 @subsubheading Synopsis
29665 -stack-list-variables @var{print-values}
29668 Display the names of local variables and function arguments for the selected frame. If
29669 @var{print-values} is 0 or @code{--no-values}, print only the names of
29670 the variables; if it is 1 or @code{--all-values}, print also their
29671 values; and if it is 2 or @code{--simple-values}, print the name,
29672 type and value for simple data types, and the name and type for arrays,
29673 structures and unions.
29675 @subsubheading Example
29679 -stack-list-variables --thread 1 --frame 0 --all-values
29680 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
29685 @subheading The @code{-stack-select-frame} Command
29686 @findex -stack-select-frame
29688 @subsubheading Synopsis
29691 -stack-select-frame @var{framenum}
29694 Change the selected frame. Select a different frame @var{framenum} on
29697 This command in deprecated in favor of passing the @samp{--frame}
29698 option to every command.
29700 @subsubheading @value{GDBN} Command
29702 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
29703 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
29705 @subsubheading Example
29709 -stack-select-frame 2
29714 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29715 @node GDB/MI Variable Objects
29716 @section @sc{gdb/mi} Variable Objects
29720 @subheading Motivation for Variable Objects in @sc{gdb/mi}
29722 For the implementation of a variable debugger window (locals, watched
29723 expressions, etc.), we are proposing the adaptation of the existing code
29724 used by @code{Insight}.
29726 The two main reasons for that are:
29730 It has been proven in practice (it is already on its second generation).
29733 It will shorten development time (needless to say how important it is
29737 The original interface was designed to be used by Tcl code, so it was
29738 slightly changed so it could be used through @sc{gdb/mi}. This section
29739 describes the @sc{gdb/mi} operations that will be available and gives some
29740 hints about their use.
29742 @emph{Note}: In addition to the set of operations described here, we
29743 expect the @sc{gui} implementation of a variable window to require, at
29744 least, the following operations:
29747 @item @code{-gdb-show} @code{output-radix}
29748 @item @code{-stack-list-arguments}
29749 @item @code{-stack-list-locals}
29750 @item @code{-stack-select-frame}
29755 @subheading Introduction to Variable Objects
29757 @cindex variable objects in @sc{gdb/mi}
29759 Variable objects are "object-oriented" MI interface for examining and
29760 changing values of expressions. Unlike some other MI interfaces that
29761 work with expressions, variable objects are specifically designed for
29762 simple and efficient presentation in the frontend. A variable object
29763 is identified by string name. When a variable object is created, the
29764 frontend specifies the expression for that variable object. The
29765 expression can be a simple variable, or it can be an arbitrary complex
29766 expression, and can even involve CPU registers. After creating a
29767 variable object, the frontend can invoke other variable object
29768 operations---for example to obtain or change the value of a variable
29769 object, or to change display format.
29771 Variable objects have hierarchical tree structure. Any variable object
29772 that corresponds to a composite type, such as structure in C, has
29773 a number of child variable objects, for example corresponding to each
29774 element of a structure. A child variable object can itself have
29775 children, recursively. Recursion ends when we reach
29776 leaf variable objects, which always have built-in types. Child variable
29777 objects are created only by explicit request, so if a frontend
29778 is not interested in the children of a particular variable object, no
29779 child will be created.
29781 For a leaf variable object it is possible to obtain its value as a
29782 string, or set the value from a string. String value can be also
29783 obtained for a non-leaf variable object, but it's generally a string
29784 that only indicates the type of the object, and does not list its
29785 contents. Assignment to a non-leaf variable object is not allowed.
29787 A frontend does not need to read the values of all variable objects each time
29788 the program stops. Instead, MI provides an update command that lists all
29789 variable objects whose values has changed since the last update
29790 operation. This considerably reduces the amount of data that must
29791 be transferred to the frontend. As noted above, children variable
29792 objects are created on demand, and only leaf variable objects have a
29793 real value. As result, gdb will read target memory only for leaf
29794 variables that frontend has created.
29796 The automatic update is not always desirable. For example, a frontend
29797 might want to keep a value of some expression for future reference,
29798 and never update it. For another example, fetching memory is
29799 relatively slow for embedded targets, so a frontend might want
29800 to disable automatic update for the variables that are either not
29801 visible on the screen, or ``closed''. This is possible using so
29802 called ``frozen variable objects''. Such variable objects are never
29803 implicitly updated.
29805 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
29806 fixed variable object, the expression is parsed when the variable
29807 object is created, including associating identifiers to specific
29808 variables. The meaning of expression never changes. For a floating
29809 variable object the values of variables whose names appear in the
29810 expressions are re-evaluated every time in the context of the current
29811 frame. Consider this example:
29816 struct work_state state;
29823 If a fixed variable object for the @code{state} variable is created in
29824 this function, and we enter the recursive call, the variable
29825 object will report the value of @code{state} in the top-level
29826 @code{do_work} invocation. On the other hand, a floating variable
29827 object will report the value of @code{state} in the current frame.
29829 If an expression specified when creating a fixed variable object
29830 refers to a local variable, the variable object becomes bound to the
29831 thread and frame in which the variable object is created. When such
29832 variable object is updated, @value{GDBN} makes sure that the
29833 thread/frame combination the variable object is bound to still exists,
29834 and re-evaluates the variable object in context of that thread/frame.
29836 The following is the complete set of @sc{gdb/mi} operations defined to
29837 access this functionality:
29839 @multitable @columnfractions .4 .6
29840 @item @strong{Operation}
29841 @tab @strong{Description}
29843 @item @code{-enable-pretty-printing}
29844 @tab enable Python-based pretty-printing
29845 @item @code{-var-create}
29846 @tab create a variable object
29847 @item @code{-var-delete}
29848 @tab delete the variable object and/or its children
29849 @item @code{-var-set-format}
29850 @tab set the display format of this variable
29851 @item @code{-var-show-format}
29852 @tab show the display format of this variable
29853 @item @code{-var-info-num-children}
29854 @tab tells how many children this object has
29855 @item @code{-var-list-children}
29856 @tab return a list of the object's children
29857 @item @code{-var-info-type}
29858 @tab show the type of this variable object
29859 @item @code{-var-info-expression}
29860 @tab print parent-relative expression that this variable object represents
29861 @item @code{-var-info-path-expression}
29862 @tab print full expression that this variable object represents
29863 @item @code{-var-show-attributes}
29864 @tab is this variable editable? does it exist here?
29865 @item @code{-var-evaluate-expression}
29866 @tab get the value of this variable
29867 @item @code{-var-assign}
29868 @tab set the value of this variable
29869 @item @code{-var-update}
29870 @tab update the variable and its children
29871 @item @code{-var-set-frozen}
29872 @tab set frozeness attribute
29873 @item @code{-var-set-update-range}
29874 @tab set range of children to display on update
29877 In the next subsection we describe each operation in detail and suggest
29878 how it can be used.
29880 @subheading Description And Use of Operations on Variable Objects
29882 @subheading The @code{-enable-pretty-printing} Command
29883 @findex -enable-pretty-printing
29886 -enable-pretty-printing
29889 @value{GDBN} allows Python-based visualizers to affect the output of the
29890 MI variable object commands. However, because there was no way to
29891 implement this in a fully backward-compatible way, a front end must
29892 request that this functionality be enabled.
29894 Once enabled, this feature cannot be disabled.
29896 Note that if Python support has not been compiled into @value{GDBN},
29897 this command will still succeed (and do nothing).
29899 This feature is currently (as of @value{GDBN} 7.0) experimental, and
29900 may work differently in future versions of @value{GDBN}.
29902 @subheading The @code{-var-create} Command
29903 @findex -var-create
29905 @subsubheading Synopsis
29908 -var-create @{@var{name} | "-"@}
29909 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
29912 This operation creates a variable object, which allows the monitoring of
29913 a variable, the result of an expression, a memory cell or a CPU
29916 The @var{name} parameter is the string by which the object can be
29917 referenced. It must be unique. If @samp{-} is specified, the varobj
29918 system will generate a string ``varNNNNNN'' automatically. It will be
29919 unique provided that one does not specify @var{name} of that format.
29920 The command fails if a duplicate name is found.
29922 The frame under which the expression should be evaluated can be
29923 specified by @var{frame-addr}. A @samp{*} indicates that the current
29924 frame should be used. A @samp{@@} indicates that a floating variable
29925 object must be created.
29927 @var{expression} is any expression valid on the current language set (must not
29928 begin with a @samp{*}), or one of the following:
29932 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
29935 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
29938 @samp{$@var{regname}} --- a CPU register name
29941 @cindex dynamic varobj
29942 A varobj's contents may be provided by a Python-based pretty-printer. In this
29943 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
29944 have slightly different semantics in some cases. If the
29945 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
29946 will never create a dynamic varobj. This ensures backward
29947 compatibility for existing clients.
29949 @subsubheading Result
29951 This operation returns attributes of the newly-created varobj. These
29956 The name of the varobj.
29959 The number of children of the varobj. This number is not necessarily
29960 reliable for a dynamic varobj. Instead, you must examine the
29961 @samp{has_more} attribute.
29964 The varobj's scalar value. For a varobj whose type is some sort of
29965 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
29966 will not be interesting.
29969 The varobj's type. This is a string representation of the type, as
29970 would be printed by the @value{GDBN} CLI. If @samp{print object}
29971 (@pxref{Print Settings, set print object}) is set to @code{on}, the
29972 @emph{actual} (derived) type of the object is shown rather than the
29973 @emph{declared} one.
29976 If a variable object is bound to a specific thread, then this is the
29977 thread's identifier.
29980 For a dynamic varobj, this indicates whether there appear to be any
29981 children available. For a non-dynamic varobj, this will be 0.
29984 This attribute will be present and have the value @samp{1} if the
29985 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29986 then this attribute will not be present.
29989 A dynamic varobj can supply a display hint to the front end. The
29990 value comes directly from the Python pretty-printer object's
29991 @code{display_hint} method. @xref{Pretty Printing API}.
29994 Typical output will look like this:
29997 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
29998 has_more="@var{has_more}"
30002 @subheading The @code{-var-delete} Command
30003 @findex -var-delete
30005 @subsubheading Synopsis
30008 -var-delete [ -c ] @var{name}
30011 Deletes a previously created variable object and all of its children.
30012 With the @samp{-c} option, just deletes the children.
30014 Returns an error if the object @var{name} is not found.
30017 @subheading The @code{-var-set-format} Command
30018 @findex -var-set-format
30020 @subsubheading Synopsis
30023 -var-set-format @var{name} @var{format-spec}
30026 Sets the output format for the value of the object @var{name} to be
30029 @anchor{-var-set-format}
30030 The syntax for the @var{format-spec} is as follows:
30033 @var{format-spec} @expansion{}
30034 @{binary | decimal | hexadecimal | octal | natural@}
30037 The natural format is the default format choosen automatically
30038 based on the variable type (like decimal for an @code{int}, hex
30039 for pointers, etc.).
30041 For a variable with children, the format is set only on the
30042 variable itself, and the children are not affected.
30044 @subheading The @code{-var-show-format} Command
30045 @findex -var-show-format
30047 @subsubheading Synopsis
30050 -var-show-format @var{name}
30053 Returns the format used to display the value of the object @var{name}.
30056 @var{format} @expansion{}
30061 @subheading The @code{-var-info-num-children} Command
30062 @findex -var-info-num-children
30064 @subsubheading Synopsis
30067 -var-info-num-children @var{name}
30070 Returns the number of children of a variable object @var{name}:
30076 Note that this number is not completely reliable for a dynamic varobj.
30077 It will return the current number of children, but more children may
30081 @subheading The @code{-var-list-children} Command
30082 @findex -var-list-children
30084 @subsubheading Synopsis
30087 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
30089 @anchor{-var-list-children}
30091 Return a list of the children of the specified variable object and
30092 create variable objects for them, if they do not already exist. With
30093 a single argument or if @var{print-values} has a value of 0 or
30094 @code{--no-values}, print only the names of the variables; if
30095 @var{print-values} is 1 or @code{--all-values}, also print their
30096 values; and if it is 2 or @code{--simple-values} print the name and
30097 value for simple data types and just the name for arrays, structures
30100 @var{from} and @var{to}, if specified, indicate the range of children
30101 to report. If @var{from} or @var{to} is less than zero, the range is
30102 reset and all children will be reported. Otherwise, children starting
30103 at @var{from} (zero-based) and up to and excluding @var{to} will be
30106 If a child range is requested, it will only affect the current call to
30107 @code{-var-list-children}, but not future calls to @code{-var-update}.
30108 For this, you must instead use @code{-var-set-update-range}. The
30109 intent of this approach is to enable a front end to implement any
30110 update approach it likes; for example, scrolling a view may cause the
30111 front end to request more children with @code{-var-list-children}, and
30112 then the front end could call @code{-var-set-update-range} with a
30113 different range to ensure that future updates are restricted to just
30116 For each child the following results are returned:
30121 Name of the variable object created for this child.
30124 The expression to be shown to the user by the front end to designate this child.
30125 For example this may be the name of a structure member.
30127 For a dynamic varobj, this value cannot be used to form an
30128 expression. There is no way to do this at all with a dynamic varobj.
30130 For C/C@t{++} structures there are several pseudo children returned to
30131 designate access qualifiers. For these pseudo children @var{exp} is
30132 @samp{public}, @samp{private}, or @samp{protected}. In this case the
30133 type and value are not present.
30135 A dynamic varobj will not report the access qualifying
30136 pseudo-children, regardless of the language. This information is not
30137 available at all with a dynamic varobj.
30140 Number of children this child has. For a dynamic varobj, this will be
30144 The type of the child. If @samp{print object}
30145 (@pxref{Print Settings, set print object}) is set to @code{on}, the
30146 @emph{actual} (derived) type of the object is shown rather than the
30147 @emph{declared} one.
30150 If values were requested, this is the value.
30153 If this variable object is associated with a thread, this is the thread id.
30154 Otherwise this result is not present.
30157 If the variable object is frozen, this variable will be present with a value of 1.
30160 The result may have its own attributes:
30164 A dynamic varobj can supply a display hint to the front end. The
30165 value comes directly from the Python pretty-printer object's
30166 @code{display_hint} method. @xref{Pretty Printing API}.
30169 This is an integer attribute which is nonzero if there are children
30170 remaining after the end of the selected range.
30173 @subsubheading Example
30177 -var-list-children n
30178 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
30179 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
30181 -var-list-children --all-values n
30182 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
30183 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
30187 @subheading The @code{-var-info-type} Command
30188 @findex -var-info-type
30190 @subsubheading Synopsis
30193 -var-info-type @var{name}
30196 Returns the type of the specified variable @var{name}. The type is
30197 returned as a string in the same format as it is output by the
30201 type=@var{typename}
30205 @subheading The @code{-var-info-expression} Command
30206 @findex -var-info-expression
30208 @subsubheading Synopsis
30211 -var-info-expression @var{name}
30214 Returns a string that is suitable for presenting this
30215 variable object in user interface. The string is generally
30216 not valid expression in the current language, and cannot be evaluated.
30218 For example, if @code{a} is an array, and variable object
30219 @code{A} was created for @code{a}, then we'll get this output:
30222 (gdb) -var-info-expression A.1
30223 ^done,lang="C",exp="1"
30227 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
30229 Note that the output of the @code{-var-list-children} command also
30230 includes those expressions, so the @code{-var-info-expression} command
30233 @subheading The @code{-var-info-path-expression} Command
30234 @findex -var-info-path-expression
30236 @subsubheading Synopsis
30239 -var-info-path-expression @var{name}
30242 Returns an expression that can be evaluated in the current
30243 context and will yield the same value that a variable object has.
30244 Compare this with the @code{-var-info-expression} command, which
30245 result can be used only for UI presentation. Typical use of
30246 the @code{-var-info-path-expression} command is creating a
30247 watchpoint from a variable object.
30249 This command is currently not valid for children of a dynamic varobj,
30250 and will give an error when invoked on one.
30252 For example, suppose @code{C} is a C@t{++} class, derived from class
30253 @code{Base}, and that the @code{Base} class has a member called
30254 @code{m_size}. Assume a variable @code{c} is has the type of
30255 @code{C} and a variable object @code{C} was created for variable
30256 @code{c}. Then, we'll get this output:
30258 (gdb) -var-info-path-expression C.Base.public.m_size
30259 ^done,path_expr=((Base)c).m_size)
30262 @subheading The @code{-var-show-attributes} Command
30263 @findex -var-show-attributes
30265 @subsubheading Synopsis
30268 -var-show-attributes @var{name}
30271 List attributes of the specified variable object @var{name}:
30274 status=@var{attr} [ ( ,@var{attr} )* ]
30278 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
30280 @subheading The @code{-var-evaluate-expression} Command
30281 @findex -var-evaluate-expression
30283 @subsubheading Synopsis
30286 -var-evaluate-expression [-f @var{format-spec}] @var{name}
30289 Evaluates the expression that is represented by the specified variable
30290 object and returns its value as a string. The format of the string
30291 can be specified with the @samp{-f} option. The possible values of
30292 this option are the same as for @code{-var-set-format}
30293 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
30294 the current display format will be used. The current display format
30295 can be changed using the @code{-var-set-format} command.
30301 Note that one must invoke @code{-var-list-children} for a variable
30302 before the value of a child variable can be evaluated.
30304 @subheading The @code{-var-assign} Command
30305 @findex -var-assign
30307 @subsubheading Synopsis
30310 -var-assign @var{name} @var{expression}
30313 Assigns the value of @var{expression} to the variable object specified
30314 by @var{name}. The object must be @samp{editable}. If the variable's
30315 value is altered by the assign, the variable will show up in any
30316 subsequent @code{-var-update} list.
30318 @subsubheading Example
30326 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
30330 @subheading The @code{-var-update} Command
30331 @findex -var-update
30333 @subsubheading Synopsis
30336 -var-update [@var{print-values}] @{@var{name} | "*"@}
30339 Reevaluate the expressions corresponding to the variable object
30340 @var{name} and all its direct and indirect children, and return the
30341 list of variable objects whose values have changed; @var{name} must
30342 be a root variable object. Here, ``changed'' means that the result of
30343 @code{-var-evaluate-expression} before and after the
30344 @code{-var-update} is different. If @samp{*} is used as the variable
30345 object names, all existing variable objects are updated, except
30346 for frozen ones (@pxref{-var-set-frozen}). The option
30347 @var{print-values} determines whether both names and values, or just
30348 names are printed. The possible values of this option are the same
30349 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
30350 recommended to use the @samp{--all-values} option, to reduce the
30351 number of MI commands needed on each program stop.
30353 With the @samp{*} parameter, if a variable object is bound to a
30354 currently running thread, it will not be updated, without any
30357 If @code{-var-set-update-range} was previously used on a varobj, then
30358 only the selected range of children will be reported.
30360 @code{-var-update} reports all the changed varobjs in a tuple named
30363 Each item in the change list is itself a tuple holding:
30367 The name of the varobj.
30370 If values were requested for this update, then this field will be
30371 present and will hold the value of the varobj.
30374 @anchor{-var-update}
30375 This field is a string which may take one of three values:
30379 The variable object's current value is valid.
30382 The variable object does not currently hold a valid value but it may
30383 hold one in the future if its associated expression comes back into
30387 The variable object no longer holds a valid value.
30388 This can occur when the executable file being debugged has changed,
30389 either through recompilation or by using the @value{GDBN} @code{file}
30390 command. The front end should normally choose to delete these variable
30394 In the future new values may be added to this list so the front should
30395 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
30398 This is only present if the varobj is still valid. If the type
30399 changed, then this will be the string @samp{true}; otherwise it will
30402 When a varobj's type changes, its children are also likely to have
30403 become incorrect. Therefore, the varobj's children are automatically
30404 deleted when this attribute is @samp{true}. Also, the varobj's update
30405 range, when set using the @code{-var-set-update-range} command, is
30409 If the varobj's type changed, then this field will be present and will
30412 @item new_num_children
30413 For a dynamic varobj, if the number of children changed, or if the
30414 type changed, this will be the new number of children.
30416 The @samp{numchild} field in other varobj responses is generally not
30417 valid for a dynamic varobj -- it will show the number of children that
30418 @value{GDBN} knows about, but because dynamic varobjs lazily
30419 instantiate their children, this will not reflect the number of
30420 children which may be available.
30422 The @samp{new_num_children} attribute only reports changes to the
30423 number of children known by @value{GDBN}. This is the only way to
30424 detect whether an update has removed children (which necessarily can
30425 only happen at the end of the update range).
30428 The display hint, if any.
30431 This is an integer value, which will be 1 if there are more children
30432 available outside the varobj's update range.
30435 This attribute will be present and have the value @samp{1} if the
30436 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
30437 then this attribute will not be present.
30440 If new children were added to a dynamic varobj within the selected
30441 update range (as set by @code{-var-set-update-range}), then they will
30442 be listed in this attribute.
30445 @subsubheading Example
30452 -var-update --all-values var1
30453 ^done,changelist=[@{name="var1",value="3",in_scope="true",
30454 type_changed="false"@}]
30458 @subheading The @code{-var-set-frozen} Command
30459 @findex -var-set-frozen
30460 @anchor{-var-set-frozen}
30462 @subsubheading Synopsis
30465 -var-set-frozen @var{name} @var{flag}
30468 Set the frozenness flag on the variable object @var{name}. The
30469 @var{flag} parameter should be either @samp{1} to make the variable
30470 frozen or @samp{0} to make it unfrozen. If a variable object is
30471 frozen, then neither itself, nor any of its children, are
30472 implicitly updated by @code{-var-update} of
30473 a parent variable or by @code{-var-update *}. Only
30474 @code{-var-update} of the variable itself will update its value and
30475 values of its children. After a variable object is unfrozen, it is
30476 implicitly updated by all subsequent @code{-var-update} operations.
30477 Unfreezing a variable does not update it, only subsequent
30478 @code{-var-update} does.
30480 @subsubheading Example
30484 -var-set-frozen V 1
30489 @subheading The @code{-var-set-update-range} command
30490 @findex -var-set-update-range
30491 @anchor{-var-set-update-range}
30493 @subsubheading Synopsis
30496 -var-set-update-range @var{name} @var{from} @var{to}
30499 Set the range of children to be returned by future invocations of
30500 @code{-var-update}.
30502 @var{from} and @var{to} indicate the range of children to report. If
30503 @var{from} or @var{to} is less than zero, the range is reset and all
30504 children will be reported. Otherwise, children starting at @var{from}
30505 (zero-based) and up to and excluding @var{to} will be reported.
30507 @subsubheading Example
30511 -var-set-update-range V 1 2
30515 @subheading The @code{-var-set-visualizer} command
30516 @findex -var-set-visualizer
30517 @anchor{-var-set-visualizer}
30519 @subsubheading Synopsis
30522 -var-set-visualizer @var{name} @var{visualizer}
30525 Set a visualizer for the variable object @var{name}.
30527 @var{visualizer} is the visualizer to use. The special value
30528 @samp{None} means to disable any visualizer in use.
30530 If not @samp{None}, @var{visualizer} must be a Python expression.
30531 This expression must evaluate to a callable object which accepts a
30532 single argument. @value{GDBN} will call this object with the value of
30533 the varobj @var{name} as an argument (this is done so that the same
30534 Python pretty-printing code can be used for both the CLI and MI).
30535 When called, this object must return an object which conforms to the
30536 pretty-printing interface (@pxref{Pretty Printing API}).
30538 The pre-defined function @code{gdb.default_visualizer} may be used to
30539 select a visualizer by following the built-in process
30540 (@pxref{Selecting Pretty-Printers}). This is done automatically when
30541 a varobj is created, and so ordinarily is not needed.
30543 This feature is only available if Python support is enabled. The MI
30544 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
30545 can be used to check this.
30547 @subsubheading Example
30549 Resetting the visualizer:
30553 -var-set-visualizer V None
30557 Reselecting the default (type-based) visualizer:
30561 -var-set-visualizer V gdb.default_visualizer
30565 Suppose @code{SomeClass} is a visualizer class. A lambda expression
30566 can be used to instantiate this class for a varobj:
30570 -var-set-visualizer V "lambda val: SomeClass()"
30574 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30575 @node GDB/MI Data Manipulation
30576 @section @sc{gdb/mi} Data Manipulation
30578 @cindex data manipulation, in @sc{gdb/mi}
30579 @cindex @sc{gdb/mi}, data manipulation
30580 This section describes the @sc{gdb/mi} commands that manipulate data:
30581 examine memory and registers, evaluate expressions, etc.
30583 @c REMOVED FROM THE INTERFACE.
30584 @c @subheading -data-assign
30585 @c Change the value of a program variable. Plenty of side effects.
30586 @c @subsubheading GDB Command
30588 @c @subsubheading Example
30591 @subheading The @code{-data-disassemble} Command
30592 @findex -data-disassemble
30594 @subsubheading Synopsis
30598 [ -s @var{start-addr} -e @var{end-addr} ]
30599 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
30607 @item @var{start-addr}
30608 is the beginning address (or @code{$pc})
30609 @item @var{end-addr}
30611 @item @var{filename}
30612 is the name of the file to disassemble
30613 @item @var{linenum}
30614 is the line number to disassemble around
30616 is the number of disassembly lines to be produced. If it is -1,
30617 the whole function will be disassembled, in case no @var{end-addr} is
30618 specified. If @var{end-addr} is specified as a non-zero value, and
30619 @var{lines} is lower than the number of disassembly lines between
30620 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
30621 displayed; if @var{lines} is higher than the number of lines between
30622 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
30625 is either 0 (meaning only disassembly), 1 (meaning mixed source and
30626 disassembly), 2 (meaning disassembly with raw opcodes), or 3 (meaning
30627 mixed source and disassembly with raw opcodes).
30630 @subsubheading Result
30632 The output for each instruction is composed of four fields:
30641 Note that whatever included in the instruction field, is not manipulated
30642 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
30644 @subsubheading @value{GDBN} Command
30646 There's no direct mapping from this command to the CLI.
30648 @subsubheading Example
30650 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
30654 -data-disassemble -s $pc -e "$pc + 20" -- 0
30657 @{address="0x000107c0",func-name="main",offset="4",
30658 inst="mov 2, %o0"@},
30659 @{address="0x000107c4",func-name="main",offset="8",
30660 inst="sethi %hi(0x11800), %o2"@},
30661 @{address="0x000107c8",func-name="main",offset="12",
30662 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
30663 @{address="0x000107cc",func-name="main",offset="16",
30664 inst="sethi %hi(0x11800), %o2"@},
30665 @{address="0x000107d0",func-name="main",offset="20",
30666 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
30670 Disassemble the whole @code{main} function. Line 32 is part of
30674 -data-disassemble -f basics.c -l 32 -- 0
30676 @{address="0x000107bc",func-name="main",offset="0",
30677 inst="save %sp, -112, %sp"@},
30678 @{address="0x000107c0",func-name="main",offset="4",
30679 inst="mov 2, %o0"@},
30680 @{address="0x000107c4",func-name="main",offset="8",
30681 inst="sethi %hi(0x11800), %o2"@},
30683 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
30684 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
30688 Disassemble 3 instructions from the start of @code{main}:
30692 -data-disassemble -f basics.c -l 32 -n 3 -- 0
30694 @{address="0x000107bc",func-name="main",offset="0",
30695 inst="save %sp, -112, %sp"@},
30696 @{address="0x000107c0",func-name="main",offset="4",
30697 inst="mov 2, %o0"@},
30698 @{address="0x000107c4",func-name="main",offset="8",
30699 inst="sethi %hi(0x11800), %o2"@}]
30703 Disassemble 3 instructions from the start of @code{main} in mixed mode:
30707 -data-disassemble -f basics.c -l 32 -n 3 -- 1
30709 src_and_asm_line=@{line="31",
30710 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
30711 testsuite/gdb.mi/basics.c",line_asm_insn=[
30712 @{address="0x000107bc",func-name="main",offset="0",
30713 inst="save %sp, -112, %sp"@}]@},
30714 src_and_asm_line=@{line="32",
30715 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
30716 testsuite/gdb.mi/basics.c",line_asm_insn=[
30717 @{address="0x000107c0",func-name="main",offset="4",
30718 inst="mov 2, %o0"@},
30719 @{address="0x000107c4",func-name="main",offset="8",
30720 inst="sethi %hi(0x11800), %o2"@}]@}]
30725 @subheading The @code{-data-evaluate-expression} Command
30726 @findex -data-evaluate-expression
30728 @subsubheading Synopsis
30731 -data-evaluate-expression @var{expr}
30734 Evaluate @var{expr} as an expression. The expression could contain an
30735 inferior function call. The function call will execute synchronously.
30736 If the expression contains spaces, it must be enclosed in double quotes.
30738 @subsubheading @value{GDBN} Command
30740 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
30741 @samp{call}. In @code{gdbtk} only, there's a corresponding
30742 @samp{gdb_eval} command.
30744 @subsubheading Example
30746 In the following example, the numbers that precede the commands are the
30747 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
30748 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
30752 211-data-evaluate-expression A
30755 311-data-evaluate-expression &A
30756 311^done,value="0xefffeb7c"
30758 411-data-evaluate-expression A+3
30761 511-data-evaluate-expression "A + 3"
30767 @subheading The @code{-data-list-changed-registers} Command
30768 @findex -data-list-changed-registers
30770 @subsubheading Synopsis
30773 -data-list-changed-registers
30776 Display a list of the registers that have changed.
30778 @subsubheading @value{GDBN} Command
30780 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
30781 has the corresponding command @samp{gdb_changed_register_list}.
30783 @subsubheading Example
30785 On a PPC MBX board:
30793 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
30794 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
30797 -data-list-changed-registers
30798 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
30799 "10","11","13","14","15","16","17","18","19","20","21","22","23",
30800 "24","25","26","27","28","30","31","64","65","66","67","69"]
30805 @subheading The @code{-data-list-register-names} Command
30806 @findex -data-list-register-names
30808 @subsubheading Synopsis
30811 -data-list-register-names [ ( @var{regno} )+ ]
30814 Show a list of register names for the current target. If no arguments
30815 are given, it shows a list of the names of all the registers. If
30816 integer numbers are given as arguments, it will print a list of the
30817 names of the registers corresponding to the arguments. To ensure
30818 consistency between a register name and its number, the output list may
30819 include empty register names.
30821 @subsubheading @value{GDBN} Command
30823 @value{GDBN} does not have a command which corresponds to
30824 @samp{-data-list-register-names}. In @code{gdbtk} there is a
30825 corresponding command @samp{gdb_regnames}.
30827 @subsubheading Example
30829 For the PPC MBX board:
30832 -data-list-register-names
30833 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
30834 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
30835 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
30836 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
30837 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
30838 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
30839 "", "pc","ps","cr","lr","ctr","xer"]
30841 -data-list-register-names 1 2 3
30842 ^done,register-names=["r1","r2","r3"]
30846 @subheading The @code{-data-list-register-values} Command
30847 @findex -data-list-register-values
30849 @subsubheading Synopsis
30852 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
30855 Display the registers' contents. @var{fmt} is the format according to
30856 which the registers' contents are to be returned, followed by an optional
30857 list of numbers specifying the registers to display. A missing list of
30858 numbers indicates that the contents of all the registers must be returned.
30860 Allowed formats for @var{fmt} are:
30877 @subsubheading @value{GDBN} Command
30879 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
30880 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
30882 @subsubheading Example
30884 For a PPC MBX board (note: line breaks are for readability only, they
30885 don't appear in the actual output):
30889 -data-list-register-values r 64 65
30890 ^done,register-values=[@{number="64",value="0xfe00a300"@},
30891 @{number="65",value="0x00029002"@}]
30893 -data-list-register-values x
30894 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
30895 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
30896 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
30897 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
30898 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
30899 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
30900 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
30901 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
30902 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
30903 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
30904 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
30905 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
30906 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
30907 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
30908 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
30909 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
30910 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
30911 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
30912 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
30913 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
30914 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
30915 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
30916 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
30917 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
30918 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
30919 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
30920 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
30921 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
30922 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
30923 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
30924 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
30925 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
30926 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
30927 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
30928 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
30929 @{number="69",value="0x20002b03"@}]
30934 @subheading The @code{-data-read-memory} Command
30935 @findex -data-read-memory
30937 This command is deprecated, use @code{-data-read-memory-bytes} instead.
30939 @subsubheading Synopsis
30942 -data-read-memory [ -o @var{byte-offset} ]
30943 @var{address} @var{word-format} @var{word-size}
30944 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
30951 @item @var{address}
30952 An expression specifying the address of the first memory word to be
30953 read. Complex expressions containing embedded white space should be
30954 quoted using the C convention.
30956 @item @var{word-format}
30957 The format to be used to print the memory words. The notation is the
30958 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
30961 @item @var{word-size}
30962 The size of each memory word in bytes.
30964 @item @var{nr-rows}
30965 The number of rows in the output table.
30967 @item @var{nr-cols}
30968 The number of columns in the output table.
30971 If present, indicates that each row should include an @sc{ascii} dump. The
30972 value of @var{aschar} is used as a padding character when a byte is not a
30973 member of the printable @sc{ascii} character set (printable @sc{ascii}
30974 characters are those whose code is between 32 and 126, inclusively).
30976 @item @var{byte-offset}
30977 An offset to add to the @var{address} before fetching memory.
30980 This command displays memory contents as a table of @var{nr-rows} by
30981 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
30982 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
30983 (returned as @samp{total-bytes}). Should less than the requested number
30984 of bytes be returned by the target, the missing words are identified
30985 using @samp{N/A}. The number of bytes read from the target is returned
30986 in @samp{nr-bytes} and the starting address used to read memory in
30989 The address of the next/previous row or page is available in
30990 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
30993 @subsubheading @value{GDBN} Command
30995 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
30996 @samp{gdb_get_mem} memory read command.
30998 @subsubheading Example
31000 Read six bytes of memory starting at @code{bytes+6} but then offset by
31001 @code{-6} bytes. Format as three rows of two columns. One byte per
31002 word. Display each word in hex.
31006 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
31007 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
31008 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
31009 prev-page="0x0000138a",memory=[
31010 @{addr="0x00001390",data=["0x00","0x01"]@},
31011 @{addr="0x00001392",data=["0x02","0x03"]@},
31012 @{addr="0x00001394",data=["0x04","0x05"]@}]
31016 Read two bytes of memory starting at address @code{shorts + 64} and
31017 display as a single word formatted in decimal.
31021 5-data-read-memory shorts+64 d 2 1 1
31022 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
31023 next-row="0x00001512",prev-row="0x0000150e",
31024 next-page="0x00001512",prev-page="0x0000150e",memory=[
31025 @{addr="0x00001510",data=["128"]@}]
31029 Read thirty two bytes of memory starting at @code{bytes+16} and format
31030 as eight rows of four columns. Include a string encoding with @samp{x}
31031 used as the non-printable character.
31035 4-data-read-memory bytes+16 x 1 8 4 x
31036 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
31037 next-row="0x000013c0",prev-row="0x0000139c",
31038 next-page="0x000013c0",prev-page="0x00001380",memory=[
31039 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
31040 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
31041 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
31042 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
31043 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
31044 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
31045 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
31046 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
31050 @subheading The @code{-data-read-memory-bytes} Command
31051 @findex -data-read-memory-bytes
31053 @subsubheading Synopsis
31056 -data-read-memory-bytes [ -o @var{byte-offset} ]
31057 @var{address} @var{count}
31064 @item @var{address}
31065 An expression specifying the address of the first memory word to be
31066 read. Complex expressions containing embedded white space should be
31067 quoted using the C convention.
31070 The number of bytes to read. This should be an integer literal.
31072 @item @var{byte-offset}
31073 The offsets in bytes relative to @var{address} at which to start
31074 reading. This should be an integer literal. This option is provided
31075 so that a frontend is not required to first evaluate address and then
31076 perform address arithmetics itself.
31080 This command attempts to read all accessible memory regions in the
31081 specified range. First, all regions marked as unreadable in the memory
31082 map (if one is defined) will be skipped. @xref{Memory Region
31083 Attributes}. Second, @value{GDBN} will attempt to read the remaining
31084 regions. For each one, if reading full region results in an errors,
31085 @value{GDBN} will try to read a subset of the region.
31087 In general, every single byte in the region may be readable or not,
31088 and the only way to read every readable byte is to try a read at
31089 every address, which is not practical. Therefore, @value{GDBN} will
31090 attempt to read all accessible bytes at either beginning or the end
31091 of the region, using a binary division scheme. This heuristic works
31092 well for reading accross a memory map boundary. Note that if a region
31093 has a readable range that is neither at the beginning or the end,
31094 @value{GDBN} will not read it.
31096 The result record (@pxref{GDB/MI Result Records}) that is output of
31097 the command includes a field named @samp{memory} whose content is a
31098 list of tuples. Each tuple represent a successfully read memory block
31099 and has the following fields:
31103 The start address of the memory block, as hexadecimal literal.
31106 The end address of the memory block, as hexadecimal literal.
31109 The offset of the memory block, as hexadecimal literal, relative to
31110 the start address passed to @code{-data-read-memory-bytes}.
31113 The contents of the memory block, in hex.
31119 @subsubheading @value{GDBN} Command
31121 The corresponding @value{GDBN} command is @samp{x}.
31123 @subsubheading Example
31127 -data-read-memory-bytes &a 10
31128 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
31130 contents="01000000020000000300"@}]
31135 @subheading The @code{-data-write-memory-bytes} Command
31136 @findex -data-write-memory-bytes
31138 @subsubheading Synopsis
31141 -data-write-memory-bytes @var{address} @var{contents}
31148 @item @var{address}
31149 An expression specifying the address of the first memory word to be
31150 read. Complex expressions containing embedded white space should be
31151 quoted using the C convention.
31153 @item @var{contents}
31154 The hex-encoded bytes to write.
31158 @subsubheading @value{GDBN} Command
31160 There's no corresponding @value{GDBN} command.
31162 @subsubheading Example
31166 -data-write-memory-bytes &a "aabbccdd"
31172 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31173 @node GDB/MI Tracepoint Commands
31174 @section @sc{gdb/mi} Tracepoint Commands
31176 The commands defined in this section implement MI support for
31177 tracepoints. For detailed introduction, see @ref{Tracepoints}.
31179 @subheading The @code{-trace-find} Command
31180 @findex -trace-find
31182 @subsubheading Synopsis
31185 -trace-find @var{mode} [@var{parameters}@dots{}]
31188 Find a trace frame using criteria defined by @var{mode} and
31189 @var{parameters}. The following table lists permissible
31190 modes and their parameters. For details of operation, see @ref{tfind}.
31195 No parameters are required. Stops examining trace frames.
31198 An integer is required as parameter. Selects tracepoint frame with
31201 @item tracepoint-number
31202 An integer is required as parameter. Finds next
31203 trace frame that corresponds to tracepoint with the specified number.
31206 An address is required as parameter. Finds
31207 next trace frame that corresponds to any tracepoint at the specified
31210 @item pc-inside-range
31211 Two addresses are required as parameters. Finds next trace
31212 frame that corresponds to a tracepoint at an address inside the
31213 specified range. Both bounds are considered to be inside the range.
31215 @item pc-outside-range
31216 Two addresses are required as parameters. Finds
31217 next trace frame that corresponds to a tracepoint at an address outside
31218 the specified range. Both bounds are considered to be inside the range.
31221 Line specification is required as parameter. @xref{Specify Location}.
31222 Finds next trace frame that corresponds to a tracepoint at
31223 the specified location.
31227 If @samp{none} was passed as @var{mode}, the response does not
31228 have fields. Otherwise, the response may have the following fields:
31232 This field has either @samp{0} or @samp{1} as the value, depending
31233 on whether a matching tracepoint was found.
31236 The index of the found traceframe. This field is present iff
31237 the @samp{found} field has value of @samp{1}.
31240 The index of the found tracepoint. This field is present iff
31241 the @samp{found} field has value of @samp{1}.
31244 The information about the frame corresponding to the found trace
31245 frame. This field is present only if a trace frame was found.
31246 @xref{GDB/MI Frame Information}, for description of this field.
31250 @subsubheading @value{GDBN} Command
31252 The corresponding @value{GDBN} command is @samp{tfind}.
31254 @subheading -trace-define-variable
31255 @findex -trace-define-variable
31257 @subsubheading Synopsis
31260 -trace-define-variable @var{name} [ @var{value} ]
31263 Create trace variable @var{name} if it does not exist. If
31264 @var{value} is specified, sets the initial value of the specified
31265 trace variable to that value. Note that the @var{name} should start
31266 with the @samp{$} character.
31268 @subsubheading @value{GDBN} Command
31270 The corresponding @value{GDBN} command is @samp{tvariable}.
31272 @subheading -trace-list-variables
31273 @findex -trace-list-variables
31275 @subsubheading Synopsis
31278 -trace-list-variables
31281 Return a table of all defined trace variables. Each element of the
31282 table has the following fields:
31286 The name of the trace variable. This field is always present.
31289 The initial value. This is a 64-bit signed integer. This
31290 field is always present.
31293 The value the trace variable has at the moment. This is a 64-bit
31294 signed integer. This field is absent iff current value is
31295 not defined, for example if the trace was never run, or is
31300 @subsubheading @value{GDBN} Command
31302 The corresponding @value{GDBN} command is @samp{tvariables}.
31304 @subsubheading Example
31308 -trace-list-variables
31309 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
31310 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
31311 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
31312 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
31313 body=[variable=@{name="$trace_timestamp",initial="0"@}
31314 variable=@{name="$foo",initial="10",current="15"@}]@}
31318 @subheading -trace-save
31319 @findex -trace-save
31321 @subsubheading Synopsis
31324 -trace-save [-r ] @var{filename}
31327 Saves the collected trace data to @var{filename}. Without the
31328 @samp{-r} option, the data is downloaded from the target and saved
31329 in a local file. With the @samp{-r} option the target is asked
31330 to perform the save.
31332 @subsubheading @value{GDBN} Command
31334 The corresponding @value{GDBN} command is @samp{tsave}.
31337 @subheading -trace-start
31338 @findex -trace-start
31340 @subsubheading Synopsis
31346 Starts a tracing experiments. The result of this command does not
31349 @subsubheading @value{GDBN} Command
31351 The corresponding @value{GDBN} command is @samp{tstart}.
31353 @subheading -trace-status
31354 @findex -trace-status
31356 @subsubheading Synopsis
31362 Obtains the status of a tracing experiment. The result may include
31363 the following fields:
31368 May have a value of either @samp{0}, when no tracing operations are
31369 supported, @samp{1}, when all tracing operations are supported, or
31370 @samp{file} when examining trace file. In the latter case, examining
31371 of trace frame is possible but new tracing experiement cannot be
31372 started. This field is always present.
31375 May have a value of either @samp{0} or @samp{1} depending on whether
31376 tracing experiement is in progress on target. This field is present
31377 if @samp{supported} field is not @samp{0}.
31380 Report the reason why the tracing was stopped last time. This field
31381 may be absent iff tracing was never stopped on target yet. The
31382 value of @samp{request} means the tracing was stopped as result of
31383 the @code{-trace-stop} command. The value of @samp{overflow} means
31384 the tracing buffer is full. The value of @samp{disconnection} means
31385 tracing was automatically stopped when @value{GDBN} has disconnected.
31386 The value of @samp{passcount} means tracing was stopped when a
31387 tracepoint was passed a maximal number of times for that tracepoint.
31388 This field is present if @samp{supported} field is not @samp{0}.
31390 @item stopping-tracepoint
31391 The number of tracepoint whose passcount as exceeded. This field is
31392 present iff the @samp{stop-reason} field has the value of
31396 @itemx frames-created
31397 The @samp{frames} field is a count of the total number of trace frames
31398 in the trace buffer, while @samp{frames-created} is the total created
31399 during the run, including ones that were discarded, such as when a
31400 circular trace buffer filled up. Both fields are optional.
31404 These fields tell the current size of the tracing buffer and the
31405 remaining space. These fields are optional.
31408 The value of the circular trace buffer flag. @code{1} means that the
31409 trace buffer is circular and old trace frames will be discarded if
31410 necessary to make room, @code{0} means that the trace buffer is linear
31414 The value of the disconnected tracing flag. @code{1} means that
31415 tracing will continue after @value{GDBN} disconnects, @code{0} means
31416 that the trace run will stop.
31420 @subsubheading @value{GDBN} Command
31422 The corresponding @value{GDBN} command is @samp{tstatus}.
31424 @subheading -trace-stop
31425 @findex -trace-stop
31427 @subsubheading Synopsis
31433 Stops a tracing experiment. The result of this command has the same
31434 fields as @code{-trace-status}, except that the @samp{supported} and
31435 @samp{running} fields are not output.
31437 @subsubheading @value{GDBN} Command
31439 The corresponding @value{GDBN} command is @samp{tstop}.
31442 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31443 @node GDB/MI Symbol Query
31444 @section @sc{gdb/mi} Symbol Query Commands
31448 @subheading The @code{-symbol-info-address} Command
31449 @findex -symbol-info-address
31451 @subsubheading Synopsis
31454 -symbol-info-address @var{symbol}
31457 Describe where @var{symbol} is stored.
31459 @subsubheading @value{GDBN} Command
31461 The corresponding @value{GDBN} command is @samp{info address}.
31463 @subsubheading Example
31467 @subheading The @code{-symbol-info-file} Command
31468 @findex -symbol-info-file
31470 @subsubheading Synopsis
31476 Show the file for the symbol.
31478 @subsubheading @value{GDBN} Command
31480 There's no equivalent @value{GDBN} command. @code{gdbtk} has
31481 @samp{gdb_find_file}.
31483 @subsubheading Example
31487 @subheading The @code{-symbol-info-function} Command
31488 @findex -symbol-info-function
31490 @subsubheading Synopsis
31493 -symbol-info-function
31496 Show which function the symbol lives in.
31498 @subsubheading @value{GDBN} Command
31500 @samp{gdb_get_function} in @code{gdbtk}.
31502 @subsubheading Example
31506 @subheading The @code{-symbol-info-line} Command
31507 @findex -symbol-info-line
31509 @subsubheading Synopsis
31515 Show the core addresses of the code for a source line.
31517 @subsubheading @value{GDBN} Command
31519 The corresponding @value{GDBN} command is @samp{info line}.
31520 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
31522 @subsubheading Example
31526 @subheading The @code{-symbol-info-symbol} Command
31527 @findex -symbol-info-symbol
31529 @subsubheading Synopsis
31532 -symbol-info-symbol @var{addr}
31535 Describe what symbol is at location @var{addr}.
31537 @subsubheading @value{GDBN} Command
31539 The corresponding @value{GDBN} command is @samp{info symbol}.
31541 @subsubheading Example
31545 @subheading The @code{-symbol-list-functions} Command
31546 @findex -symbol-list-functions
31548 @subsubheading Synopsis
31551 -symbol-list-functions
31554 List the functions in the executable.
31556 @subsubheading @value{GDBN} Command
31558 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
31559 @samp{gdb_search} in @code{gdbtk}.
31561 @subsubheading Example
31566 @subheading The @code{-symbol-list-lines} Command
31567 @findex -symbol-list-lines
31569 @subsubheading Synopsis
31572 -symbol-list-lines @var{filename}
31575 Print the list of lines that contain code and their associated program
31576 addresses for the given source filename. The entries are sorted in
31577 ascending PC order.
31579 @subsubheading @value{GDBN} Command
31581 There is no corresponding @value{GDBN} command.
31583 @subsubheading Example
31586 -symbol-list-lines basics.c
31587 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
31593 @subheading The @code{-symbol-list-types} Command
31594 @findex -symbol-list-types
31596 @subsubheading Synopsis
31602 List all the type names.
31604 @subsubheading @value{GDBN} Command
31606 The corresponding commands are @samp{info types} in @value{GDBN},
31607 @samp{gdb_search} in @code{gdbtk}.
31609 @subsubheading Example
31613 @subheading The @code{-symbol-list-variables} Command
31614 @findex -symbol-list-variables
31616 @subsubheading Synopsis
31619 -symbol-list-variables
31622 List all the global and static variable names.
31624 @subsubheading @value{GDBN} Command
31626 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
31628 @subsubheading Example
31632 @subheading The @code{-symbol-locate} Command
31633 @findex -symbol-locate
31635 @subsubheading Synopsis
31641 @subsubheading @value{GDBN} Command
31643 @samp{gdb_loc} in @code{gdbtk}.
31645 @subsubheading Example
31649 @subheading The @code{-symbol-type} Command
31650 @findex -symbol-type
31652 @subsubheading Synopsis
31655 -symbol-type @var{variable}
31658 Show type of @var{variable}.
31660 @subsubheading @value{GDBN} Command
31662 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
31663 @samp{gdb_obj_variable}.
31665 @subsubheading Example
31670 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31671 @node GDB/MI File Commands
31672 @section @sc{gdb/mi} File Commands
31674 This section describes the GDB/MI commands to specify executable file names
31675 and to read in and obtain symbol table information.
31677 @subheading The @code{-file-exec-and-symbols} Command
31678 @findex -file-exec-and-symbols
31680 @subsubheading Synopsis
31683 -file-exec-and-symbols @var{file}
31686 Specify the executable file to be debugged. This file is the one from
31687 which the symbol table is also read. If no file is specified, the
31688 command clears the executable and symbol information. If breakpoints
31689 are set when using this command with no arguments, @value{GDBN} will produce
31690 error messages. Otherwise, no output is produced, except a completion
31693 @subsubheading @value{GDBN} Command
31695 The corresponding @value{GDBN} command is @samp{file}.
31697 @subsubheading Example
31701 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31707 @subheading The @code{-file-exec-file} Command
31708 @findex -file-exec-file
31710 @subsubheading Synopsis
31713 -file-exec-file @var{file}
31716 Specify the executable file to be debugged. Unlike
31717 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
31718 from this file. If used without argument, @value{GDBN} clears the information
31719 about the executable file. No output is produced, except a completion
31722 @subsubheading @value{GDBN} Command
31724 The corresponding @value{GDBN} command is @samp{exec-file}.
31726 @subsubheading Example
31730 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31737 @subheading The @code{-file-list-exec-sections} Command
31738 @findex -file-list-exec-sections
31740 @subsubheading Synopsis
31743 -file-list-exec-sections
31746 List the sections of the current executable file.
31748 @subsubheading @value{GDBN} Command
31750 The @value{GDBN} command @samp{info file} shows, among the rest, the same
31751 information as this command. @code{gdbtk} has a corresponding command
31752 @samp{gdb_load_info}.
31754 @subsubheading Example
31759 @subheading The @code{-file-list-exec-source-file} Command
31760 @findex -file-list-exec-source-file
31762 @subsubheading Synopsis
31765 -file-list-exec-source-file
31768 List the line number, the current source file, and the absolute path
31769 to the current source file for the current executable. The macro
31770 information field has a value of @samp{1} or @samp{0} depending on
31771 whether or not the file includes preprocessor macro information.
31773 @subsubheading @value{GDBN} Command
31775 The @value{GDBN} equivalent is @samp{info source}
31777 @subsubheading Example
31781 123-file-list-exec-source-file
31782 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
31787 @subheading The @code{-file-list-exec-source-files} Command
31788 @findex -file-list-exec-source-files
31790 @subsubheading Synopsis
31793 -file-list-exec-source-files
31796 List the source files for the current executable.
31798 It will always output the filename, but only when @value{GDBN} can find
31799 the absolute file name of a source file, will it output the fullname.
31801 @subsubheading @value{GDBN} Command
31803 The @value{GDBN} equivalent is @samp{info sources}.
31804 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
31806 @subsubheading Example
31809 -file-list-exec-source-files
31811 @{file=foo.c,fullname=/home/foo.c@},
31812 @{file=/home/bar.c,fullname=/home/bar.c@},
31813 @{file=gdb_could_not_find_fullpath.c@}]
31818 @subheading The @code{-file-list-shared-libraries} Command
31819 @findex -file-list-shared-libraries
31821 @subsubheading Synopsis
31824 -file-list-shared-libraries
31827 List the shared libraries in the program.
31829 @subsubheading @value{GDBN} Command
31831 The corresponding @value{GDBN} command is @samp{info shared}.
31833 @subsubheading Example
31837 @subheading The @code{-file-list-symbol-files} Command
31838 @findex -file-list-symbol-files
31840 @subsubheading Synopsis
31843 -file-list-symbol-files
31848 @subsubheading @value{GDBN} Command
31850 The corresponding @value{GDBN} command is @samp{info file} (part of it).
31852 @subsubheading Example
31857 @subheading The @code{-file-symbol-file} Command
31858 @findex -file-symbol-file
31860 @subsubheading Synopsis
31863 -file-symbol-file @var{file}
31866 Read symbol table info from the specified @var{file} argument. When
31867 used without arguments, clears @value{GDBN}'s symbol table info. No output is
31868 produced, except for a completion notification.
31870 @subsubheading @value{GDBN} Command
31872 The corresponding @value{GDBN} command is @samp{symbol-file}.
31874 @subsubheading Example
31878 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31884 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31885 @node GDB/MI Memory Overlay Commands
31886 @section @sc{gdb/mi} Memory Overlay Commands
31888 The memory overlay commands are not implemented.
31890 @c @subheading -overlay-auto
31892 @c @subheading -overlay-list-mapping-state
31894 @c @subheading -overlay-list-overlays
31896 @c @subheading -overlay-map
31898 @c @subheading -overlay-off
31900 @c @subheading -overlay-on
31902 @c @subheading -overlay-unmap
31904 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31905 @node GDB/MI Signal Handling Commands
31906 @section @sc{gdb/mi} Signal Handling Commands
31908 Signal handling commands are not implemented.
31910 @c @subheading -signal-handle
31912 @c @subheading -signal-list-handle-actions
31914 @c @subheading -signal-list-signal-types
31918 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31919 @node GDB/MI Target Manipulation
31920 @section @sc{gdb/mi} Target Manipulation Commands
31923 @subheading The @code{-target-attach} Command
31924 @findex -target-attach
31926 @subsubheading Synopsis
31929 -target-attach @var{pid} | @var{gid} | @var{file}
31932 Attach to a process @var{pid} or a file @var{file} outside of
31933 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
31934 group, the id previously returned by
31935 @samp{-list-thread-groups --available} must be used.
31937 @subsubheading @value{GDBN} Command
31939 The corresponding @value{GDBN} command is @samp{attach}.
31941 @subsubheading Example
31945 =thread-created,id="1"
31946 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
31952 @subheading The @code{-target-compare-sections} Command
31953 @findex -target-compare-sections
31955 @subsubheading Synopsis
31958 -target-compare-sections [ @var{section} ]
31961 Compare data of section @var{section} on target to the exec file.
31962 Without the argument, all sections are compared.
31964 @subsubheading @value{GDBN} Command
31966 The @value{GDBN} equivalent is @samp{compare-sections}.
31968 @subsubheading Example
31973 @subheading The @code{-target-detach} Command
31974 @findex -target-detach
31976 @subsubheading Synopsis
31979 -target-detach [ @var{pid} | @var{gid} ]
31982 Detach from the remote target which normally resumes its execution.
31983 If either @var{pid} or @var{gid} is specified, detaches from either
31984 the specified process, or specified thread group. There's no output.
31986 @subsubheading @value{GDBN} Command
31988 The corresponding @value{GDBN} command is @samp{detach}.
31990 @subsubheading Example
32000 @subheading The @code{-target-disconnect} Command
32001 @findex -target-disconnect
32003 @subsubheading Synopsis
32009 Disconnect from the remote target. There's no output and the target is
32010 generally not resumed.
32012 @subsubheading @value{GDBN} Command
32014 The corresponding @value{GDBN} command is @samp{disconnect}.
32016 @subsubheading Example
32026 @subheading The @code{-target-download} Command
32027 @findex -target-download
32029 @subsubheading Synopsis
32035 Loads the executable onto the remote target.
32036 It prints out an update message every half second, which includes the fields:
32040 The name of the section.
32042 The size of what has been sent so far for that section.
32044 The size of the section.
32046 The total size of what was sent so far (the current and the previous sections).
32048 The size of the overall executable to download.
32052 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
32053 @sc{gdb/mi} Output Syntax}).
32055 In addition, it prints the name and size of the sections, as they are
32056 downloaded. These messages include the following fields:
32060 The name of the section.
32062 The size of the section.
32064 The size of the overall executable to download.
32068 At the end, a summary is printed.
32070 @subsubheading @value{GDBN} Command
32072 The corresponding @value{GDBN} command is @samp{load}.
32074 @subsubheading Example
32076 Note: each status message appears on a single line. Here the messages
32077 have been broken down so that they can fit onto a page.
32082 +download,@{section=".text",section-size="6668",total-size="9880"@}
32083 +download,@{section=".text",section-sent="512",section-size="6668",
32084 total-sent="512",total-size="9880"@}
32085 +download,@{section=".text",section-sent="1024",section-size="6668",
32086 total-sent="1024",total-size="9880"@}
32087 +download,@{section=".text",section-sent="1536",section-size="6668",
32088 total-sent="1536",total-size="9880"@}
32089 +download,@{section=".text",section-sent="2048",section-size="6668",
32090 total-sent="2048",total-size="9880"@}
32091 +download,@{section=".text",section-sent="2560",section-size="6668",
32092 total-sent="2560",total-size="9880"@}
32093 +download,@{section=".text",section-sent="3072",section-size="6668",
32094 total-sent="3072",total-size="9880"@}
32095 +download,@{section=".text",section-sent="3584",section-size="6668",
32096 total-sent="3584",total-size="9880"@}
32097 +download,@{section=".text",section-sent="4096",section-size="6668",
32098 total-sent="4096",total-size="9880"@}
32099 +download,@{section=".text",section-sent="4608",section-size="6668",
32100 total-sent="4608",total-size="9880"@}
32101 +download,@{section=".text",section-sent="5120",section-size="6668",
32102 total-sent="5120",total-size="9880"@}
32103 +download,@{section=".text",section-sent="5632",section-size="6668",
32104 total-sent="5632",total-size="9880"@}
32105 +download,@{section=".text",section-sent="6144",section-size="6668",
32106 total-sent="6144",total-size="9880"@}
32107 +download,@{section=".text",section-sent="6656",section-size="6668",
32108 total-sent="6656",total-size="9880"@}
32109 +download,@{section=".init",section-size="28",total-size="9880"@}
32110 +download,@{section=".fini",section-size="28",total-size="9880"@}
32111 +download,@{section=".data",section-size="3156",total-size="9880"@}
32112 +download,@{section=".data",section-sent="512",section-size="3156",
32113 total-sent="7236",total-size="9880"@}
32114 +download,@{section=".data",section-sent="1024",section-size="3156",
32115 total-sent="7748",total-size="9880"@}
32116 +download,@{section=".data",section-sent="1536",section-size="3156",
32117 total-sent="8260",total-size="9880"@}
32118 +download,@{section=".data",section-sent="2048",section-size="3156",
32119 total-sent="8772",total-size="9880"@}
32120 +download,@{section=".data",section-sent="2560",section-size="3156",
32121 total-sent="9284",total-size="9880"@}
32122 +download,@{section=".data",section-sent="3072",section-size="3156",
32123 total-sent="9796",total-size="9880"@}
32124 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
32131 @subheading The @code{-target-exec-status} Command
32132 @findex -target-exec-status
32134 @subsubheading Synopsis
32137 -target-exec-status
32140 Provide information on the state of the target (whether it is running or
32141 not, for instance).
32143 @subsubheading @value{GDBN} Command
32145 There's no equivalent @value{GDBN} command.
32147 @subsubheading Example
32151 @subheading The @code{-target-list-available-targets} Command
32152 @findex -target-list-available-targets
32154 @subsubheading Synopsis
32157 -target-list-available-targets
32160 List the possible targets to connect to.
32162 @subsubheading @value{GDBN} Command
32164 The corresponding @value{GDBN} command is @samp{help target}.
32166 @subsubheading Example
32170 @subheading The @code{-target-list-current-targets} Command
32171 @findex -target-list-current-targets
32173 @subsubheading Synopsis
32176 -target-list-current-targets
32179 Describe the current target.
32181 @subsubheading @value{GDBN} Command
32183 The corresponding information is printed by @samp{info file} (among
32186 @subsubheading Example
32190 @subheading The @code{-target-list-parameters} Command
32191 @findex -target-list-parameters
32193 @subsubheading Synopsis
32196 -target-list-parameters
32202 @subsubheading @value{GDBN} Command
32206 @subsubheading Example
32210 @subheading The @code{-target-select} Command
32211 @findex -target-select
32213 @subsubheading Synopsis
32216 -target-select @var{type} @var{parameters @dots{}}
32219 Connect @value{GDBN} to the remote target. This command takes two args:
32223 The type of target, for instance @samp{remote}, etc.
32224 @item @var{parameters}
32225 Device names, host names and the like. @xref{Target Commands, ,
32226 Commands for Managing Targets}, for more details.
32229 The output is a connection notification, followed by the address at
32230 which the target program is, in the following form:
32233 ^connected,addr="@var{address}",func="@var{function name}",
32234 args=[@var{arg list}]
32237 @subsubheading @value{GDBN} Command
32239 The corresponding @value{GDBN} command is @samp{target}.
32241 @subsubheading Example
32245 -target-select remote /dev/ttya
32246 ^connected,addr="0xfe00a300",func="??",args=[]
32250 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32251 @node GDB/MI File Transfer Commands
32252 @section @sc{gdb/mi} File Transfer Commands
32255 @subheading The @code{-target-file-put} Command
32256 @findex -target-file-put
32258 @subsubheading Synopsis
32261 -target-file-put @var{hostfile} @var{targetfile}
32264 Copy file @var{hostfile} from the host system (the machine running
32265 @value{GDBN}) to @var{targetfile} on the target system.
32267 @subsubheading @value{GDBN} Command
32269 The corresponding @value{GDBN} command is @samp{remote put}.
32271 @subsubheading Example
32275 -target-file-put localfile remotefile
32281 @subheading The @code{-target-file-get} Command
32282 @findex -target-file-get
32284 @subsubheading Synopsis
32287 -target-file-get @var{targetfile} @var{hostfile}
32290 Copy file @var{targetfile} from the target system to @var{hostfile}
32291 on the host system.
32293 @subsubheading @value{GDBN} Command
32295 The corresponding @value{GDBN} command is @samp{remote get}.
32297 @subsubheading Example
32301 -target-file-get remotefile localfile
32307 @subheading The @code{-target-file-delete} Command
32308 @findex -target-file-delete
32310 @subsubheading Synopsis
32313 -target-file-delete @var{targetfile}
32316 Delete @var{targetfile} from the target system.
32318 @subsubheading @value{GDBN} Command
32320 The corresponding @value{GDBN} command is @samp{remote delete}.
32322 @subsubheading Example
32326 -target-file-delete remotefile
32332 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32333 @node GDB/MI Miscellaneous Commands
32334 @section Miscellaneous @sc{gdb/mi} Commands
32336 @c @subheading -gdb-complete
32338 @subheading The @code{-gdb-exit} Command
32341 @subsubheading Synopsis
32347 Exit @value{GDBN} immediately.
32349 @subsubheading @value{GDBN} Command
32351 Approximately corresponds to @samp{quit}.
32353 @subsubheading Example
32363 @subheading The @code{-exec-abort} Command
32364 @findex -exec-abort
32366 @subsubheading Synopsis
32372 Kill the inferior running program.
32374 @subsubheading @value{GDBN} Command
32376 The corresponding @value{GDBN} command is @samp{kill}.
32378 @subsubheading Example
32383 @subheading The @code{-gdb-set} Command
32386 @subsubheading Synopsis
32392 Set an internal @value{GDBN} variable.
32393 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
32395 @subsubheading @value{GDBN} Command
32397 The corresponding @value{GDBN} command is @samp{set}.
32399 @subsubheading Example
32409 @subheading The @code{-gdb-show} Command
32412 @subsubheading Synopsis
32418 Show the current value of a @value{GDBN} variable.
32420 @subsubheading @value{GDBN} Command
32422 The corresponding @value{GDBN} command is @samp{show}.
32424 @subsubheading Example
32433 @c @subheading -gdb-source
32436 @subheading The @code{-gdb-version} Command
32437 @findex -gdb-version
32439 @subsubheading Synopsis
32445 Show version information for @value{GDBN}. Used mostly in testing.
32447 @subsubheading @value{GDBN} Command
32449 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
32450 default shows this information when you start an interactive session.
32452 @subsubheading Example
32454 @c This example modifies the actual output from GDB to avoid overfull
32460 ~Copyright 2000 Free Software Foundation, Inc.
32461 ~GDB is free software, covered by the GNU General Public License, and
32462 ~you are welcome to change it and/or distribute copies of it under
32463 ~ certain conditions.
32464 ~Type "show copying" to see the conditions.
32465 ~There is absolutely no warranty for GDB. Type "show warranty" for
32467 ~This GDB was configured as
32468 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
32473 @subheading The @code{-list-features} Command
32474 @findex -list-features
32476 Returns a list of particular features of the MI protocol that
32477 this version of gdb implements. A feature can be a command,
32478 or a new field in an output of some command, or even an
32479 important bugfix. While a frontend can sometimes detect presence
32480 of a feature at runtime, it is easier to perform detection at debugger
32483 The command returns a list of strings, with each string naming an
32484 available feature. Each returned string is just a name, it does not
32485 have any internal structure. The list of possible feature names
32491 (gdb) -list-features
32492 ^done,result=["feature1","feature2"]
32495 The current list of features is:
32498 @item frozen-varobjs
32499 Indicates support for the @code{-var-set-frozen} command, as well
32500 as possible presense of the @code{frozen} field in the output
32501 of @code{-varobj-create}.
32502 @item pending-breakpoints
32503 Indicates support for the @option{-f} option to the @code{-break-insert}
32506 Indicates Python scripting support, Python-based
32507 pretty-printing commands, and possible presence of the
32508 @samp{display_hint} field in the output of @code{-var-list-children}
32510 Indicates support for the @code{-thread-info} command.
32511 @item data-read-memory-bytes
32512 Indicates support for the @code{-data-read-memory-bytes} and the
32513 @code{-data-write-memory-bytes} commands.
32514 @item breakpoint-notifications
32515 Indicates that changes to breakpoints and breakpoints created via the
32516 CLI will be announced via async records.
32517 @item ada-task-info
32518 Indicates support for the @code{-ada-task-info} command.
32521 @subheading The @code{-list-target-features} Command
32522 @findex -list-target-features
32524 Returns a list of particular features that are supported by the
32525 target. Those features affect the permitted MI commands, but
32526 unlike the features reported by the @code{-list-features} command, the
32527 features depend on which target GDB is using at the moment. Whenever
32528 a target can change, due to commands such as @code{-target-select},
32529 @code{-target-attach} or @code{-exec-run}, the list of target features
32530 may change, and the frontend should obtain it again.
32534 (gdb) -list-features
32535 ^done,result=["async"]
32538 The current list of features is:
32542 Indicates that the target is capable of asynchronous command
32543 execution, which means that @value{GDBN} will accept further commands
32544 while the target is running.
32547 Indicates that the target is capable of reverse execution.
32548 @xref{Reverse Execution}, for more information.
32552 @subheading The @code{-list-thread-groups} Command
32553 @findex -list-thread-groups
32555 @subheading Synopsis
32558 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
32561 Lists thread groups (@pxref{Thread groups}). When a single thread
32562 group is passed as the argument, lists the children of that group.
32563 When several thread group are passed, lists information about those
32564 thread groups. Without any parameters, lists information about all
32565 top-level thread groups.
32567 Normally, thread groups that are being debugged are reported.
32568 With the @samp{--available} option, @value{GDBN} reports thread groups
32569 available on the target.
32571 The output of this command may have either a @samp{threads} result or
32572 a @samp{groups} result. The @samp{thread} result has a list of tuples
32573 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
32574 Information}). The @samp{groups} result has a list of tuples as value,
32575 each tuple describing a thread group. If top-level groups are
32576 requested (that is, no parameter is passed), or when several groups
32577 are passed, the output always has a @samp{groups} result. The format
32578 of the @samp{group} result is described below.
32580 To reduce the number of roundtrips it's possible to list thread groups
32581 together with their children, by passing the @samp{--recurse} option
32582 and the recursion depth. Presently, only recursion depth of 1 is
32583 permitted. If this option is present, then every reported thread group
32584 will also include its children, either as @samp{group} or
32585 @samp{threads} field.
32587 In general, any combination of option and parameters is permitted, with
32588 the following caveats:
32592 When a single thread group is passed, the output will typically
32593 be the @samp{threads} result. Because threads may not contain
32594 anything, the @samp{recurse} option will be ignored.
32597 When the @samp{--available} option is passed, limited information may
32598 be available. In particular, the list of threads of a process might
32599 be inaccessible. Further, specifying specific thread groups might
32600 not give any performance advantage over listing all thread groups.
32601 The frontend should assume that @samp{-list-thread-groups --available}
32602 is always an expensive operation and cache the results.
32606 The @samp{groups} result is a list of tuples, where each tuple may
32607 have the following fields:
32611 Identifier of the thread group. This field is always present.
32612 The identifier is an opaque string; frontends should not try to
32613 convert it to an integer, even though it might look like one.
32616 The type of the thread group. At present, only @samp{process} is a
32620 The target-specific process identifier. This field is only present
32621 for thread groups of type @samp{process} and only if the process exists.
32624 The number of children this thread group has. This field may be
32625 absent for an available thread group.
32628 This field has a list of tuples as value, each tuple describing a
32629 thread. It may be present if the @samp{--recurse} option is
32630 specified, and it's actually possible to obtain the threads.
32633 This field is a list of integers, each identifying a core that one
32634 thread of the group is running on. This field may be absent if
32635 such information is not available.
32638 The name of the executable file that corresponds to this thread group.
32639 The field is only present for thread groups of type @samp{process},
32640 and only if there is a corresponding executable file.
32644 @subheading Example
32648 -list-thread-groups
32649 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
32650 -list-thread-groups 17
32651 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
32652 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
32653 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
32654 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
32655 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
32656 -list-thread-groups --available
32657 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
32658 -list-thread-groups --available --recurse 1
32659 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
32660 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
32661 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
32662 -list-thread-groups --available --recurse 1 17 18
32663 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
32664 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
32665 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
32668 @subheading The @code{-info-os} Command
32671 @subsubheading Synopsis
32674 -info-os [ @var{type} ]
32677 If no argument is supplied, the command returns a table of available
32678 operating-system-specific information types. If one of these types is
32679 supplied as an argument @var{type}, then the command returns a table
32680 of data of that type.
32682 The types of information available depend on the target operating
32685 @subsubheading @value{GDBN} Command
32687 The corresponding @value{GDBN} command is @samp{info os}.
32689 @subsubheading Example
32691 When run on a @sc{gnu}/Linux system, the output will look something
32697 ^done,OSDataTable=@{nr_rows="9",nr_cols="3",
32698 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
32699 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
32700 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
32701 body=[item=@{col0="processes",col1="Listing of all processes",
32702 col2="Processes"@},
32703 item=@{col0="procgroups",col1="Listing of all process groups",
32704 col2="Process groups"@},
32705 item=@{col0="threads",col1="Listing of all threads",
32707 item=@{col0="files",col1="Listing of all file descriptors",
32708 col2="File descriptors"@},
32709 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
32711 item=@{col0="shm",col1="Listing of all shared-memory regions",
32712 col2="Shared-memory regions"@},
32713 item=@{col0="semaphores",col1="Listing of all semaphores",
32714 col2="Semaphores"@},
32715 item=@{col0="msg",col1="Listing of all message queues",
32716 col2="Message queues"@},
32717 item=@{col0="modules",col1="Listing of all loaded kernel modules",
32718 col2="Kernel modules"@}]@}
32721 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
32722 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
32723 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
32724 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
32725 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
32726 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
32727 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
32728 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
32730 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
32731 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
32735 (Note that the MI output here includes a @code{"Title"} column that
32736 does not appear in command-line @code{info os}; this column is useful
32737 for MI clients that want to enumerate the types of data, such as in a
32738 popup menu, but is needless clutter on the command line, and
32739 @code{info os} omits it.)
32741 @subheading The @code{-add-inferior} Command
32742 @findex -add-inferior
32744 @subheading Synopsis
32750 Creates a new inferior (@pxref{Inferiors and Programs}). The created
32751 inferior is not associated with any executable. Such association may
32752 be established with the @samp{-file-exec-and-symbols} command
32753 (@pxref{GDB/MI File Commands}). The command response has a single
32754 field, @samp{thread-group}, whose value is the identifier of the
32755 thread group corresponding to the new inferior.
32757 @subheading Example
32762 ^done,thread-group="i3"
32765 @subheading The @code{-interpreter-exec} Command
32766 @findex -interpreter-exec
32768 @subheading Synopsis
32771 -interpreter-exec @var{interpreter} @var{command}
32773 @anchor{-interpreter-exec}
32775 Execute the specified @var{command} in the given @var{interpreter}.
32777 @subheading @value{GDBN} Command
32779 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
32781 @subheading Example
32785 -interpreter-exec console "break main"
32786 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
32787 &"During symbol reading, bad structure-type format.\n"
32788 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
32793 @subheading The @code{-inferior-tty-set} Command
32794 @findex -inferior-tty-set
32796 @subheading Synopsis
32799 -inferior-tty-set /dev/pts/1
32802 Set terminal for future runs of the program being debugged.
32804 @subheading @value{GDBN} Command
32806 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
32808 @subheading Example
32812 -inferior-tty-set /dev/pts/1
32817 @subheading The @code{-inferior-tty-show} Command
32818 @findex -inferior-tty-show
32820 @subheading Synopsis
32826 Show terminal for future runs of program being debugged.
32828 @subheading @value{GDBN} Command
32830 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
32832 @subheading Example
32836 -inferior-tty-set /dev/pts/1
32840 ^done,inferior_tty_terminal="/dev/pts/1"
32844 @subheading The @code{-enable-timings} Command
32845 @findex -enable-timings
32847 @subheading Synopsis
32850 -enable-timings [yes | no]
32853 Toggle the printing of the wallclock, user and system times for an MI
32854 command as a field in its output. This command is to help frontend
32855 developers optimize the performance of their code. No argument is
32856 equivalent to @samp{yes}.
32858 @subheading @value{GDBN} Command
32862 @subheading Example
32870 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
32871 addr="0x080484ed",func="main",file="myprog.c",
32872 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
32873 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
32881 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
32882 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
32883 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
32884 fullname="/home/nickrob/myprog.c",line="73"@}
32889 @chapter @value{GDBN} Annotations
32891 This chapter describes annotations in @value{GDBN}. Annotations were
32892 designed to interface @value{GDBN} to graphical user interfaces or other
32893 similar programs which want to interact with @value{GDBN} at a
32894 relatively high level.
32896 The annotation mechanism has largely been superseded by @sc{gdb/mi}
32900 This is Edition @value{EDITION}, @value{DATE}.
32904 * Annotations Overview:: What annotations are; the general syntax.
32905 * Server Prefix:: Issuing a command without affecting user state.
32906 * Prompting:: Annotations marking @value{GDBN}'s need for input.
32907 * Errors:: Annotations for error messages.
32908 * Invalidation:: Some annotations describe things now invalid.
32909 * Annotations for Running::
32910 Whether the program is running, how it stopped, etc.
32911 * Source Annotations:: Annotations describing source code.
32914 @node Annotations Overview
32915 @section What is an Annotation?
32916 @cindex annotations
32918 Annotations start with a newline character, two @samp{control-z}
32919 characters, and the name of the annotation. If there is no additional
32920 information associated with this annotation, the name of the annotation
32921 is followed immediately by a newline. If there is additional
32922 information, the name of the annotation is followed by a space, the
32923 additional information, and a newline. The additional information
32924 cannot contain newline characters.
32926 Any output not beginning with a newline and two @samp{control-z}
32927 characters denotes literal output from @value{GDBN}. Currently there is
32928 no need for @value{GDBN} to output a newline followed by two
32929 @samp{control-z} characters, but if there was such a need, the
32930 annotations could be extended with an @samp{escape} annotation which
32931 means those three characters as output.
32933 The annotation @var{level}, which is specified using the
32934 @option{--annotate} command line option (@pxref{Mode Options}), controls
32935 how much information @value{GDBN} prints together with its prompt,
32936 values of expressions, source lines, and other types of output. Level 0
32937 is for no annotations, level 1 is for use when @value{GDBN} is run as a
32938 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
32939 for programs that control @value{GDBN}, and level 2 annotations have
32940 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
32941 Interface, annotate, GDB's Obsolete Annotations}).
32944 @kindex set annotate
32945 @item set annotate @var{level}
32946 The @value{GDBN} command @code{set annotate} sets the level of
32947 annotations to the specified @var{level}.
32949 @item show annotate
32950 @kindex show annotate
32951 Show the current annotation level.
32954 This chapter describes level 3 annotations.
32956 A simple example of starting up @value{GDBN} with annotations is:
32959 $ @kbd{gdb --annotate=3}
32961 Copyright 2003 Free Software Foundation, Inc.
32962 GDB is free software, covered by the GNU General Public License,
32963 and you are welcome to change it and/or distribute copies of it
32964 under certain conditions.
32965 Type "show copying" to see the conditions.
32966 There is absolutely no warranty for GDB. Type "show warranty"
32968 This GDB was configured as "i386-pc-linux-gnu"
32979 Here @samp{quit} is input to @value{GDBN}; the rest is output from
32980 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
32981 denotes a @samp{control-z} character) are annotations; the rest is
32982 output from @value{GDBN}.
32984 @node Server Prefix
32985 @section The Server Prefix
32986 @cindex server prefix
32988 If you prefix a command with @samp{server } then it will not affect
32989 the command history, nor will it affect @value{GDBN}'s notion of which
32990 command to repeat if @key{RET} is pressed on a line by itself. This
32991 means that commands can be run behind a user's back by a front-end in
32992 a transparent manner.
32994 The @code{server } prefix does not affect the recording of values into
32995 the value history; to print a value without recording it into the
32996 value history, use the @code{output} command instead of the
32997 @code{print} command.
32999 Using this prefix also disables confirmation requests
33000 (@pxref{confirmation requests}).
33003 @section Annotation for @value{GDBN} Input
33005 @cindex annotations for prompts
33006 When @value{GDBN} prompts for input, it annotates this fact so it is possible
33007 to know when to send output, when the output from a given command is
33010 Different kinds of input each have a different @dfn{input type}. Each
33011 input type has three annotations: a @code{pre-} annotation, which
33012 denotes the beginning of any prompt which is being output, a plain
33013 annotation, which denotes the end of the prompt, and then a @code{post-}
33014 annotation which denotes the end of any echo which may (or may not) be
33015 associated with the input. For example, the @code{prompt} input type
33016 features the following annotations:
33024 The input types are
33027 @findex pre-prompt annotation
33028 @findex prompt annotation
33029 @findex post-prompt annotation
33031 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
33033 @findex pre-commands annotation
33034 @findex commands annotation
33035 @findex post-commands annotation
33037 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
33038 command. The annotations are repeated for each command which is input.
33040 @findex pre-overload-choice annotation
33041 @findex overload-choice annotation
33042 @findex post-overload-choice annotation
33043 @item overload-choice
33044 When @value{GDBN} wants the user to select between various overloaded functions.
33046 @findex pre-query annotation
33047 @findex query annotation
33048 @findex post-query annotation
33050 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
33052 @findex pre-prompt-for-continue annotation
33053 @findex prompt-for-continue annotation
33054 @findex post-prompt-for-continue annotation
33055 @item prompt-for-continue
33056 When @value{GDBN} is asking the user to press return to continue. Note: Don't
33057 expect this to work well; instead use @code{set height 0} to disable
33058 prompting. This is because the counting of lines is buggy in the
33059 presence of annotations.
33064 @cindex annotations for errors, warnings and interrupts
33066 @findex quit annotation
33071 This annotation occurs right before @value{GDBN} responds to an interrupt.
33073 @findex error annotation
33078 This annotation occurs right before @value{GDBN} responds to an error.
33080 Quit and error annotations indicate that any annotations which @value{GDBN} was
33081 in the middle of may end abruptly. For example, if a
33082 @code{value-history-begin} annotation is followed by a @code{error}, one
33083 cannot expect to receive the matching @code{value-history-end}. One
33084 cannot expect not to receive it either, however; an error annotation
33085 does not necessarily mean that @value{GDBN} is immediately returning all the way
33088 @findex error-begin annotation
33089 A quit or error annotation may be preceded by
33095 Any output between that and the quit or error annotation is the error
33098 Warning messages are not yet annotated.
33099 @c If we want to change that, need to fix warning(), type_error(),
33100 @c range_error(), and possibly other places.
33103 @section Invalidation Notices
33105 @cindex annotations for invalidation messages
33106 The following annotations say that certain pieces of state may have
33110 @findex frames-invalid annotation
33111 @item ^Z^Zframes-invalid
33113 The frames (for example, output from the @code{backtrace} command) may
33116 @findex breakpoints-invalid annotation
33117 @item ^Z^Zbreakpoints-invalid
33119 The breakpoints may have changed. For example, the user just added or
33120 deleted a breakpoint.
33123 @node Annotations for Running
33124 @section Running the Program
33125 @cindex annotations for running programs
33127 @findex starting annotation
33128 @findex stopping annotation
33129 When the program starts executing due to a @value{GDBN} command such as
33130 @code{step} or @code{continue},
33136 is output. When the program stops,
33142 is output. Before the @code{stopped} annotation, a variety of
33143 annotations describe how the program stopped.
33146 @findex exited annotation
33147 @item ^Z^Zexited @var{exit-status}
33148 The program exited, and @var{exit-status} is the exit status (zero for
33149 successful exit, otherwise nonzero).
33151 @findex signalled annotation
33152 @findex signal-name annotation
33153 @findex signal-name-end annotation
33154 @findex signal-string annotation
33155 @findex signal-string-end annotation
33156 @item ^Z^Zsignalled
33157 The program exited with a signal. After the @code{^Z^Zsignalled}, the
33158 annotation continues:
33164 ^Z^Zsignal-name-end
33168 ^Z^Zsignal-string-end
33173 where @var{name} is the name of the signal, such as @code{SIGILL} or
33174 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
33175 as @code{Illegal Instruction} or @code{Segmentation fault}.
33176 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
33177 user's benefit and have no particular format.
33179 @findex signal annotation
33181 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
33182 just saying that the program received the signal, not that it was
33183 terminated with it.
33185 @findex breakpoint annotation
33186 @item ^Z^Zbreakpoint @var{number}
33187 The program hit breakpoint number @var{number}.
33189 @findex watchpoint annotation
33190 @item ^Z^Zwatchpoint @var{number}
33191 The program hit watchpoint number @var{number}.
33194 @node Source Annotations
33195 @section Displaying Source
33196 @cindex annotations for source display
33198 @findex source annotation
33199 The following annotation is used instead of displaying source code:
33202 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
33205 where @var{filename} is an absolute file name indicating which source
33206 file, @var{line} is the line number within that file (where 1 is the
33207 first line in the file), @var{character} is the character position
33208 within the file (where 0 is the first character in the file) (for most
33209 debug formats this will necessarily point to the beginning of a line),
33210 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
33211 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
33212 @var{addr} is the address in the target program associated with the
33213 source which is being displayed. @var{addr} is in the form @samp{0x}
33214 followed by one or more lowercase hex digits (note that this does not
33215 depend on the language).
33217 @node JIT Interface
33218 @chapter JIT Compilation Interface
33219 @cindex just-in-time compilation
33220 @cindex JIT compilation interface
33222 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
33223 interface. A JIT compiler is a program or library that generates native
33224 executable code at runtime and executes it, usually in order to achieve good
33225 performance while maintaining platform independence.
33227 Programs that use JIT compilation are normally difficult to debug because
33228 portions of their code are generated at runtime, instead of being loaded from
33229 object files, which is where @value{GDBN} normally finds the program's symbols
33230 and debug information. In order to debug programs that use JIT compilation,
33231 @value{GDBN} has an interface that allows the program to register in-memory
33232 symbol files with @value{GDBN} at runtime.
33234 If you are using @value{GDBN} to debug a program that uses this interface, then
33235 it should work transparently so long as you have not stripped the binary. If
33236 you are developing a JIT compiler, then the interface is documented in the rest
33237 of this chapter. At this time, the only known client of this interface is the
33240 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
33241 JIT compiler communicates with @value{GDBN} by writing data into a global
33242 variable and calling a fuction at a well-known symbol. When @value{GDBN}
33243 attaches, it reads a linked list of symbol files from the global variable to
33244 find existing code, and puts a breakpoint in the function so that it can find
33245 out about additional code.
33248 * Declarations:: Relevant C struct declarations
33249 * Registering Code:: Steps to register code
33250 * Unregistering Code:: Steps to unregister code
33251 * Custom Debug Info:: Emit debug information in a custom format
33255 @section JIT Declarations
33257 These are the relevant struct declarations that a C program should include to
33258 implement the interface:
33268 struct jit_code_entry
33270 struct jit_code_entry *next_entry;
33271 struct jit_code_entry *prev_entry;
33272 const char *symfile_addr;
33273 uint64_t symfile_size;
33276 struct jit_descriptor
33279 /* This type should be jit_actions_t, but we use uint32_t
33280 to be explicit about the bitwidth. */
33281 uint32_t action_flag;
33282 struct jit_code_entry *relevant_entry;
33283 struct jit_code_entry *first_entry;
33286 /* GDB puts a breakpoint in this function. */
33287 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
33289 /* Make sure to specify the version statically, because the
33290 debugger may check the version before we can set it. */
33291 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
33294 If the JIT is multi-threaded, then it is important that the JIT synchronize any
33295 modifications to this global data properly, which can easily be done by putting
33296 a global mutex around modifications to these structures.
33298 @node Registering Code
33299 @section Registering Code
33301 To register code with @value{GDBN}, the JIT should follow this protocol:
33305 Generate an object file in memory with symbols and other desired debug
33306 information. The file must include the virtual addresses of the sections.
33309 Create a code entry for the file, which gives the start and size of the symbol
33313 Add it to the linked list in the JIT descriptor.
33316 Point the relevant_entry field of the descriptor at the entry.
33319 Set @code{action_flag} to @code{JIT_REGISTER} and call
33320 @code{__jit_debug_register_code}.
33323 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
33324 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
33325 new code. However, the linked list must still be maintained in order to allow
33326 @value{GDBN} to attach to a running process and still find the symbol files.
33328 @node Unregistering Code
33329 @section Unregistering Code
33331 If code is freed, then the JIT should use the following protocol:
33335 Remove the code entry corresponding to the code from the linked list.
33338 Point the @code{relevant_entry} field of the descriptor at the code entry.
33341 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
33342 @code{__jit_debug_register_code}.
33345 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
33346 and the JIT will leak the memory used for the associated symbol files.
33348 @node Custom Debug Info
33349 @section Custom Debug Info
33350 @cindex custom JIT debug info
33351 @cindex JIT debug info reader
33353 Generating debug information in platform-native file formats (like ELF
33354 or COFF) may be an overkill for JIT compilers; especially if all the
33355 debug info is used for is displaying a meaningful backtrace. The
33356 issue can be resolved by having the JIT writers decide on a debug info
33357 format and also provide a reader that parses the debug info generated
33358 by the JIT compiler. This section gives a brief overview on writing
33359 such a parser. More specific details can be found in the source file
33360 @file{gdb/jit-reader.in}, which is also installed as a header at
33361 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
33363 The reader is implemented as a shared object (so this functionality is
33364 not available on platforms which don't allow loading shared objects at
33365 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
33366 @code{jit-reader-unload} are provided, to be used to load and unload
33367 the readers from a preconfigured directory. Once loaded, the shared
33368 object is used the parse the debug information emitted by the JIT
33372 * Using JIT Debug Info Readers:: How to use supplied readers correctly
33373 * Writing JIT Debug Info Readers:: Creating a debug-info reader
33376 @node Using JIT Debug Info Readers
33377 @subsection Using JIT Debug Info Readers
33378 @kindex jit-reader-load
33379 @kindex jit-reader-unload
33381 Readers can be loaded and unloaded using the @code{jit-reader-load}
33382 and @code{jit-reader-unload} commands.
33385 @item jit-reader-load @var{reader-name}
33386 Load the JIT reader named @var{reader-name}. On a UNIX system, this
33387 will usually load @file{@var{libdir}/gdb/@var{reader-name}}, where
33388 @var{libdir} is the system library directory, usually
33389 @file{/usr/local/lib}. Only one reader can be active at a time;
33390 trying to load a second reader when one is already loaded will result
33391 in @value{GDBN} reporting an error. A new JIT reader can be loaded by
33392 first unloading the current one using @code{jit-reader-load} and then
33393 invoking @code{jit-reader-load}.
33395 @item jit-reader-unload
33396 Unload the currently loaded JIT reader.
33400 @node Writing JIT Debug Info Readers
33401 @subsection Writing JIT Debug Info Readers
33402 @cindex writing JIT debug info readers
33404 As mentioned, a reader is essentially a shared object conforming to a
33405 certain ABI. This ABI is described in @file{jit-reader.h}.
33407 @file{jit-reader.h} defines the structures, macros and functions
33408 required to write a reader. It is installed (along with
33409 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
33410 the system include directory.
33412 Readers need to be released under a GPL compatible license. A reader
33413 can be declared as released under such a license by placing the macro
33414 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
33416 The entry point for readers is the symbol @code{gdb_init_reader},
33417 which is expected to be a function with the prototype
33419 @findex gdb_init_reader
33421 extern struct gdb_reader_funcs *gdb_init_reader (void);
33424 @cindex @code{struct gdb_reader_funcs}
33426 @code{struct gdb_reader_funcs} contains a set of pointers to callback
33427 functions. These functions are executed to read the debug info
33428 generated by the JIT compiler (@code{read}), to unwind stack frames
33429 (@code{unwind}) and to create canonical frame IDs
33430 (@code{get_Frame_id}). It also has a callback that is called when the
33431 reader is being unloaded (@code{destroy}). The struct looks like this
33434 struct gdb_reader_funcs
33436 /* Must be set to GDB_READER_INTERFACE_VERSION. */
33437 int reader_version;
33439 /* For use by the reader. */
33442 gdb_read_debug_info *read;
33443 gdb_unwind_frame *unwind;
33444 gdb_get_frame_id *get_frame_id;
33445 gdb_destroy_reader *destroy;
33449 @cindex @code{struct gdb_symbol_callbacks}
33450 @cindex @code{struct gdb_unwind_callbacks}
33452 The callbacks are provided with another set of callbacks by
33453 @value{GDBN} to do their job. For @code{read}, these callbacks are
33454 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
33455 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
33456 @code{struct gdb_symbol_callbacks} has callbacks to create new object
33457 files and new symbol tables inside those object files. @code{struct
33458 gdb_unwind_callbacks} has callbacks to read registers off the current
33459 frame and to write out the values of the registers in the previous
33460 frame. Both have a callback (@code{target_read}) to read bytes off the
33461 target's address space.
33463 @node In-Process Agent
33464 @chapter In-Process Agent
33465 @cindex debugging agent
33466 The traditional debugging model is conceptually low-speed, but works fine,
33467 because most bugs can be reproduced in debugging-mode execution. However,
33468 as multi-core or many-core processors are becoming mainstream, and
33469 multi-threaded programs become more and more popular, there should be more
33470 and more bugs that only manifest themselves at normal-mode execution, for
33471 example, thread races, because debugger's interference with the program's
33472 timing may conceal the bugs. On the other hand, in some applications,
33473 it is not feasible for the debugger to interrupt the program's execution
33474 long enough for the developer to learn anything helpful about its behavior.
33475 If the program's correctness depends on its real-time behavior, delays
33476 introduced by a debugger might cause the program to fail, even when the
33477 code itself is correct. It is useful to be able to observe the program's
33478 behavior without interrupting it.
33480 Therefore, traditional debugging model is too intrusive to reproduce
33481 some bugs. In order to reduce the interference with the program, we can
33482 reduce the number of operations performed by debugger. The
33483 @dfn{In-Process Agent}, a shared library, is running within the same
33484 process with inferior, and is able to perform some debugging operations
33485 itself. As a result, debugger is only involved when necessary, and
33486 performance of debugging can be improved accordingly. Note that
33487 interference with program can be reduced but can't be removed completely,
33488 because the in-process agent will still stop or slow down the program.
33490 The in-process agent can interpret and execute Agent Expressions
33491 (@pxref{Agent Expressions}) during performing debugging operations. The
33492 agent expressions can be used for different purposes, such as collecting
33493 data in tracepoints, and condition evaluation in breakpoints.
33495 @anchor{Control Agent}
33496 You can control whether the in-process agent is used as an aid for
33497 debugging with the following commands:
33500 @kindex set agent on
33502 Causes the in-process agent to perform some operations on behalf of the
33503 debugger. Just which operations requested by the user will be done
33504 by the in-process agent depends on the its capabilities. For example,
33505 if you request to evaluate breakpoint conditions in the in-process agent,
33506 and the in-process agent has such capability as well, then breakpoint
33507 conditions will be evaluated in the in-process agent.
33509 @kindex set agent off
33510 @item set agent off
33511 Disables execution of debugging operations by the in-process agent. All
33512 of the operations will be performed by @value{GDBN}.
33516 Display the current setting of execution of debugging operations by
33517 the in-process agent.
33521 * In-Process Agent Protocol::
33524 @node In-Process Agent Protocol
33525 @section In-Process Agent Protocol
33526 @cindex in-process agent protocol
33528 The in-process agent is able to communicate with both @value{GDBN} and
33529 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
33530 used for communications between @value{GDBN} or GDBserver and the IPA.
33531 In general, @value{GDBN} or GDBserver sends commands
33532 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
33533 in-process agent replies back with the return result of the command, or
33534 some other information. The data sent to in-process agent is composed
33535 of primitive data types, such as 4-byte or 8-byte type, and composite
33536 types, which are called objects (@pxref{IPA Protocol Objects}).
33539 * IPA Protocol Objects::
33540 * IPA Protocol Commands::
33543 @node IPA Protocol Objects
33544 @subsection IPA Protocol Objects
33545 @cindex ipa protocol objects
33547 The commands sent to and results received from agent may contain some
33548 complex data types called @dfn{objects}.
33550 The in-process agent is running on the same machine with @value{GDBN}
33551 or GDBserver, so it doesn't have to handle as much differences between
33552 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
33553 However, there are still some differences of two ends in two processes:
33557 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
33558 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
33560 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
33561 GDBserver is compiled with one, and in-process agent is compiled with
33565 Here are the IPA Protocol Objects:
33569 agent expression object. It represents an agent expression
33570 (@pxref{Agent Expressions}).
33571 @anchor{agent expression object}
33573 tracepoint action object. It represents a tracepoint action
33574 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
33575 memory, static trace data and to evaluate expression.
33576 @anchor{tracepoint action object}
33578 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
33579 @anchor{tracepoint object}
33583 The following table describes important attributes of each IPA protocol
33586 @multitable @columnfractions .30 .20 .50
33587 @headitem Name @tab Size @tab Description
33588 @item @emph{agent expression object} @tab @tab
33589 @item length @tab 4 @tab length of bytes code
33590 @item byte code @tab @var{length} @tab contents of byte code
33591 @item @emph{tracepoint action for collecting memory} @tab @tab
33592 @item 'M' @tab 1 @tab type of tracepoint action
33593 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
33594 address of the lowest byte to collect, otherwise @var{addr} is the offset
33595 of @var{basereg} for memory collecting.
33596 @item len @tab 8 @tab length of memory for collecting
33597 @item basereg @tab 4 @tab the register number containing the starting
33598 memory address for collecting.
33599 @item @emph{tracepoint action for collecting registers} @tab @tab
33600 @item 'R' @tab 1 @tab type of tracepoint action
33601 @item @emph{tracepoint action for collecting static trace data} @tab @tab
33602 @item 'L' @tab 1 @tab type of tracepoint action
33603 @item @emph{tracepoint action for expression evaluation} @tab @tab
33604 @item 'X' @tab 1 @tab type of tracepoint action
33605 @item agent expression @tab length of @tab @ref{agent expression object}
33606 @item @emph{tracepoint object} @tab @tab
33607 @item number @tab 4 @tab number of tracepoint
33608 @item address @tab 8 @tab address of tracepoint inserted on
33609 @item type @tab 4 @tab type of tracepoint
33610 @item enabled @tab 1 @tab enable or disable of tracepoint
33611 @item step_count @tab 8 @tab step
33612 @item pass_count @tab 8 @tab pass
33613 @item numactions @tab 4 @tab number of tracepoint actions
33614 @item hit count @tab 8 @tab hit count
33615 @item trace frame usage @tab 8 @tab trace frame usage
33616 @item compiled_cond @tab 8 @tab compiled condition
33617 @item orig_size @tab 8 @tab orig size
33618 @item condition @tab 4 if condition is NULL otherwise length of
33619 @ref{agent expression object}
33620 @tab zero if condition is NULL, otherwise is
33621 @ref{agent expression object}
33622 @item actions @tab variable
33623 @tab numactions number of @ref{tracepoint action object}
33626 @node IPA Protocol Commands
33627 @subsection IPA Protocol Commands
33628 @cindex ipa protocol commands
33630 The spaces in each command are delimiters to ease reading this commands
33631 specification. They don't exist in real commands.
33635 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
33636 Installs a new fast tracepoint described by @var{tracepoint_object}
33637 (@pxref{tracepoint object}). @var{gdb_jump_pad_head}, 8-byte long, is the
33638 head of @dfn{jumppad}, which is used to jump to data collection routine
33643 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
33644 @var{target_address} is address of tracepoint in the inferior.
33645 @var{gdb_jump_pad_head} is updated head of jumppad. Both of
33646 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
33647 @var{fjump} contains a sequence of instructions jump to jumppad entry.
33648 @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
33660 @item probe_marker_at:@var{address}
33661 Asks in-process agent to probe the marker at @var{address}.
33668 @item unprobe_marker_at:@var{address}
33669 Asks in-process agent to unprobe the marker at @var{address}.
33673 @chapter Reporting Bugs in @value{GDBN}
33674 @cindex bugs in @value{GDBN}
33675 @cindex reporting bugs in @value{GDBN}
33677 Your bug reports play an essential role in making @value{GDBN} reliable.
33679 Reporting a bug may help you by bringing a solution to your problem, or it
33680 may not. But in any case the principal function of a bug report is to help
33681 the entire community by making the next version of @value{GDBN} work better. Bug
33682 reports are your contribution to the maintenance of @value{GDBN}.
33684 In order for a bug report to serve its purpose, you must include the
33685 information that enables us to fix the bug.
33688 * Bug Criteria:: Have you found a bug?
33689 * Bug Reporting:: How to report bugs
33693 @section Have You Found a Bug?
33694 @cindex bug criteria
33696 If you are not sure whether you have found a bug, here are some guidelines:
33699 @cindex fatal signal
33700 @cindex debugger crash
33701 @cindex crash of debugger
33703 If the debugger gets a fatal signal, for any input whatever, that is a
33704 @value{GDBN} bug. Reliable debuggers never crash.
33706 @cindex error on valid input
33708 If @value{GDBN} produces an error message for valid input, that is a
33709 bug. (Note that if you're cross debugging, the problem may also be
33710 somewhere in the connection to the target.)
33712 @cindex invalid input
33714 If @value{GDBN} does not produce an error message for invalid input,
33715 that is a bug. However, you should note that your idea of
33716 ``invalid input'' might be our idea of ``an extension'' or ``support
33717 for traditional practice''.
33720 If you are an experienced user of debugging tools, your suggestions
33721 for improvement of @value{GDBN} are welcome in any case.
33724 @node Bug Reporting
33725 @section How to Report Bugs
33726 @cindex bug reports
33727 @cindex @value{GDBN} bugs, reporting
33729 A number of companies and individuals offer support for @sc{gnu} products.
33730 If you obtained @value{GDBN} from a support organization, we recommend you
33731 contact that organization first.
33733 You can find contact information for many support companies and
33734 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
33736 @c should add a web page ref...
33739 @ifset BUGURL_DEFAULT
33740 In any event, we also recommend that you submit bug reports for
33741 @value{GDBN}. The preferred method is to submit them directly using
33742 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
33743 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
33746 @strong{Do not send bug reports to @samp{info-gdb}, or to
33747 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
33748 not want to receive bug reports. Those that do have arranged to receive
33751 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
33752 serves as a repeater. The mailing list and the newsgroup carry exactly
33753 the same messages. Often people think of posting bug reports to the
33754 newsgroup instead of mailing them. This appears to work, but it has one
33755 problem which can be crucial: a newsgroup posting often lacks a mail
33756 path back to the sender. Thus, if we need to ask for more information,
33757 we may be unable to reach you. For this reason, it is better to send
33758 bug reports to the mailing list.
33760 @ifclear BUGURL_DEFAULT
33761 In any event, we also recommend that you submit bug reports for
33762 @value{GDBN} to @value{BUGURL}.
33766 The fundamental principle of reporting bugs usefully is this:
33767 @strong{report all the facts}. If you are not sure whether to state a
33768 fact or leave it out, state it!
33770 Often people omit facts because they think they know what causes the
33771 problem and assume that some details do not matter. Thus, you might
33772 assume that the name of the variable you use in an example does not matter.
33773 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
33774 stray memory reference which happens to fetch from the location where that
33775 name is stored in memory; perhaps, if the name were different, the contents
33776 of that location would fool the debugger into doing the right thing despite
33777 the bug. Play it safe and give a specific, complete example. That is the
33778 easiest thing for you to do, and the most helpful.
33780 Keep in mind that the purpose of a bug report is to enable us to fix the
33781 bug. It may be that the bug has been reported previously, but neither
33782 you nor we can know that unless your bug report is complete and
33785 Sometimes people give a few sketchy facts and ask, ``Does this ring a
33786 bell?'' Those bug reports are useless, and we urge everyone to
33787 @emph{refuse to respond to them} except to chide the sender to report
33790 To enable us to fix the bug, you should include all these things:
33794 The version of @value{GDBN}. @value{GDBN} announces it if you start
33795 with no arguments; you can also print it at any time using @code{show
33798 Without this, we will not know whether there is any point in looking for
33799 the bug in the current version of @value{GDBN}.
33802 The type of machine you are using, and the operating system name and
33806 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
33807 ``@value{GCC}--2.8.1''.
33810 What compiler (and its version) was used to compile the program you are
33811 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
33812 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
33813 to get this information; for other compilers, see the documentation for
33817 The command arguments you gave the compiler to compile your example and
33818 observe the bug. For example, did you use @samp{-O}? To guarantee
33819 you will not omit something important, list them all. A copy of the
33820 Makefile (or the output from make) is sufficient.
33822 If we were to try to guess the arguments, we would probably guess wrong
33823 and then we might not encounter the bug.
33826 A complete input script, and all necessary source files, that will
33830 A description of what behavior you observe that you believe is
33831 incorrect. For example, ``It gets a fatal signal.''
33833 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
33834 will certainly notice it. But if the bug is incorrect output, we might
33835 not notice unless it is glaringly wrong. You might as well not give us
33836 a chance to make a mistake.
33838 Even if the problem you experience is a fatal signal, you should still
33839 say so explicitly. Suppose something strange is going on, such as, your
33840 copy of @value{GDBN} is out of synch, or you have encountered a bug in
33841 the C library on your system. (This has happened!) Your copy might
33842 crash and ours would not. If you told us to expect a crash, then when
33843 ours fails to crash, we would know that the bug was not happening for
33844 us. If you had not told us to expect a crash, then we would not be able
33845 to draw any conclusion from our observations.
33848 @cindex recording a session script
33849 To collect all this information, you can use a session recording program
33850 such as @command{script}, which is available on many Unix systems.
33851 Just run your @value{GDBN} session inside @command{script} and then
33852 include the @file{typescript} file with your bug report.
33854 Another way to record a @value{GDBN} session is to run @value{GDBN}
33855 inside Emacs and then save the entire buffer to a file.
33858 If you wish to suggest changes to the @value{GDBN} source, send us context
33859 diffs. If you even discuss something in the @value{GDBN} source, refer to
33860 it by context, not by line number.
33862 The line numbers in our development sources will not match those in your
33863 sources. Your line numbers would convey no useful information to us.
33867 Here are some things that are not necessary:
33871 A description of the envelope of the bug.
33873 Often people who encounter a bug spend a lot of time investigating
33874 which changes to the input file will make the bug go away and which
33875 changes will not affect it.
33877 This is often time consuming and not very useful, because the way we
33878 will find the bug is by running a single example under the debugger
33879 with breakpoints, not by pure deduction from a series of examples.
33880 We recommend that you save your time for something else.
33882 Of course, if you can find a simpler example to report @emph{instead}
33883 of the original one, that is a convenience for us. Errors in the
33884 output will be easier to spot, running under the debugger will take
33885 less time, and so on.
33887 However, simplification is not vital; if you do not want to do this,
33888 report the bug anyway and send us the entire test case you used.
33891 A patch for the bug.
33893 A patch for the bug does help us if it is a good one. But do not omit
33894 the necessary information, such as the test case, on the assumption that
33895 a patch is all we need. We might see problems with your patch and decide
33896 to fix the problem another way, or we might not understand it at all.
33898 Sometimes with a program as complicated as @value{GDBN} it is very hard to
33899 construct an example that will make the program follow a certain path
33900 through the code. If you do not send us the example, we will not be able
33901 to construct one, so we will not be able to verify that the bug is fixed.
33903 And if we cannot understand what bug you are trying to fix, or why your
33904 patch should be an improvement, we will not install it. A test case will
33905 help us to understand.
33908 A guess about what the bug is or what it depends on.
33910 Such guesses are usually wrong. Even we cannot guess right about such
33911 things without first using the debugger to find the facts.
33914 @c The readline documentation is distributed with the readline code
33915 @c and consists of the two following files:
33918 @c Use -I with makeinfo to point to the appropriate directory,
33919 @c environment var TEXINPUTS with TeX.
33920 @ifclear SYSTEM_READLINE
33921 @include rluser.texi
33922 @include hsuser.texi
33926 @appendix In Memoriam
33928 The @value{GDBN} project mourns the loss of the following long-time
33933 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
33934 to Free Software in general. Outside of @value{GDBN}, he was known in
33935 the Amiga world for his series of Fish Disks, and the GeekGadget project.
33937 @item Michael Snyder
33938 Michael was one of the Global Maintainers of the @value{GDBN} project,
33939 with contributions recorded as early as 1996, until 2011. In addition
33940 to his day to day participation, he was a large driving force behind
33941 adding Reverse Debugging to @value{GDBN}.
33944 Beyond their technical contributions to the project, they were also
33945 enjoyable members of the Free Software Community. We will miss them.
33947 @node Formatting Documentation
33948 @appendix Formatting Documentation
33950 @cindex @value{GDBN} reference card
33951 @cindex reference card
33952 The @value{GDBN} 4 release includes an already-formatted reference card, ready
33953 for printing with PostScript or Ghostscript, in the @file{gdb}
33954 subdirectory of the main source directory@footnote{In
33955 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
33956 release.}. If you can use PostScript or Ghostscript with your printer,
33957 you can print the reference card immediately with @file{refcard.ps}.
33959 The release also includes the source for the reference card. You
33960 can format it, using @TeX{}, by typing:
33966 The @value{GDBN} reference card is designed to print in @dfn{landscape}
33967 mode on US ``letter'' size paper;
33968 that is, on a sheet 11 inches wide by 8.5 inches
33969 high. You will need to specify this form of printing as an option to
33970 your @sc{dvi} output program.
33972 @cindex documentation
33974 All the documentation for @value{GDBN} comes as part of the machine-readable
33975 distribution. The documentation is written in Texinfo format, which is
33976 a documentation system that uses a single source file to produce both
33977 on-line information and a printed manual. You can use one of the Info
33978 formatting commands to create the on-line version of the documentation
33979 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
33981 @value{GDBN} includes an already formatted copy of the on-line Info
33982 version of this manual in the @file{gdb} subdirectory. The main Info
33983 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
33984 subordinate files matching @samp{gdb.info*} in the same directory. If
33985 necessary, you can print out these files, or read them with any editor;
33986 but they are easier to read using the @code{info} subsystem in @sc{gnu}
33987 Emacs or the standalone @code{info} program, available as part of the
33988 @sc{gnu} Texinfo distribution.
33990 If you want to format these Info files yourself, you need one of the
33991 Info formatting programs, such as @code{texinfo-format-buffer} or
33994 If you have @code{makeinfo} installed, and are in the top level
33995 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
33996 version @value{GDBVN}), you can make the Info file by typing:
34003 If you want to typeset and print copies of this manual, you need @TeX{},
34004 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
34005 Texinfo definitions file.
34007 @TeX{} is a typesetting program; it does not print files directly, but
34008 produces output files called @sc{dvi} files. To print a typeset
34009 document, you need a program to print @sc{dvi} files. If your system
34010 has @TeX{} installed, chances are it has such a program. The precise
34011 command to use depends on your system; @kbd{lpr -d} is common; another
34012 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
34013 require a file name without any extension or a @samp{.dvi} extension.
34015 @TeX{} also requires a macro definitions file called
34016 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
34017 written in Texinfo format. On its own, @TeX{} cannot either read or
34018 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
34019 and is located in the @file{gdb-@var{version-number}/texinfo}
34022 If you have @TeX{} and a @sc{dvi} printer program installed, you can
34023 typeset and print this manual. First switch to the @file{gdb}
34024 subdirectory of the main source directory (for example, to
34025 @file{gdb-@value{GDBVN}/gdb}) and type:
34031 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
34033 @node Installing GDB
34034 @appendix Installing @value{GDBN}
34035 @cindex installation
34038 * Requirements:: Requirements for building @value{GDBN}
34039 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
34040 * Separate Objdir:: Compiling @value{GDBN} in another directory
34041 * Config Names:: Specifying names for hosts and targets
34042 * Configure Options:: Summary of options for configure
34043 * System-wide configuration:: Having a system-wide init file
34047 @section Requirements for Building @value{GDBN}
34048 @cindex building @value{GDBN}, requirements for
34050 Building @value{GDBN} requires various tools and packages to be available.
34051 Other packages will be used only if they are found.
34053 @heading Tools/Packages Necessary for Building @value{GDBN}
34055 @item ISO C90 compiler
34056 @value{GDBN} is written in ISO C90. It should be buildable with any
34057 working C90 compiler, e.g.@: GCC.
34061 @heading Tools/Packages Optional for Building @value{GDBN}
34065 @value{GDBN} can use the Expat XML parsing library. This library may be
34066 included with your operating system distribution; if it is not, you
34067 can get the latest version from @url{http://expat.sourceforge.net}.
34068 The @file{configure} script will search for this library in several
34069 standard locations; if it is installed in an unusual path, you can
34070 use the @option{--with-libexpat-prefix} option to specify its location.
34076 Remote protocol memory maps (@pxref{Memory Map Format})
34078 Target descriptions (@pxref{Target Descriptions})
34080 Remote shared library lists (@xref{Library List Format},
34081 or alternatively @pxref{Library List Format for SVR4 Targets})
34083 MS-Windows shared libraries (@pxref{Shared Libraries})
34085 Traceframe info (@pxref{Traceframe Info Format})
34089 @cindex compressed debug sections
34090 @value{GDBN} will use the @samp{zlib} library, if available, to read
34091 compressed debug sections. Some linkers, such as GNU gold, are capable
34092 of producing binaries with compressed debug sections. If @value{GDBN}
34093 is compiled with @samp{zlib}, it will be able to read the debug
34094 information in such binaries.
34096 The @samp{zlib} library is likely included with your operating system
34097 distribution; if it is not, you can get the latest version from
34098 @url{http://zlib.net}.
34101 @value{GDBN}'s features related to character sets (@pxref{Character
34102 Sets}) require a functioning @code{iconv} implementation. If you are
34103 on a GNU system, then this is provided by the GNU C Library. Some
34104 other systems also provide a working @code{iconv}.
34106 If @value{GDBN} is using the @code{iconv} program which is installed
34107 in a non-standard place, you will need to tell @value{GDBN} where to find it.
34108 This is done with @option{--with-iconv-bin} which specifies the
34109 directory that contains the @code{iconv} program.
34111 On systems without @code{iconv}, you can install GNU Libiconv. If you
34112 have previously installed Libiconv, you can use the
34113 @option{--with-libiconv-prefix} option to configure.
34115 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
34116 arrange to build Libiconv if a directory named @file{libiconv} appears
34117 in the top-most source directory. If Libiconv is built this way, and
34118 if the operating system does not provide a suitable @code{iconv}
34119 implementation, then the just-built library will automatically be used
34120 by @value{GDBN}. One easy way to set this up is to download GNU
34121 Libiconv, unpack it, and then rename the directory holding the
34122 Libiconv source code to @samp{libiconv}.
34125 @node Running Configure
34126 @section Invoking the @value{GDBN} @file{configure} Script
34127 @cindex configuring @value{GDBN}
34128 @value{GDBN} comes with a @file{configure} script that automates the process
34129 of preparing @value{GDBN} for installation; you can then use @code{make} to
34130 build the @code{gdb} program.
34132 @c irrelevant in info file; it's as current as the code it lives with.
34133 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
34134 look at the @file{README} file in the sources; we may have improved the
34135 installation procedures since publishing this manual.}
34138 The @value{GDBN} distribution includes all the source code you need for
34139 @value{GDBN} in a single directory, whose name is usually composed by
34140 appending the version number to @samp{gdb}.
34142 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
34143 @file{gdb-@value{GDBVN}} directory. That directory contains:
34146 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
34147 script for configuring @value{GDBN} and all its supporting libraries
34149 @item gdb-@value{GDBVN}/gdb
34150 the source specific to @value{GDBN} itself
34152 @item gdb-@value{GDBVN}/bfd
34153 source for the Binary File Descriptor library
34155 @item gdb-@value{GDBVN}/include
34156 @sc{gnu} include files
34158 @item gdb-@value{GDBVN}/libiberty
34159 source for the @samp{-liberty} free software library
34161 @item gdb-@value{GDBVN}/opcodes
34162 source for the library of opcode tables and disassemblers
34164 @item gdb-@value{GDBVN}/readline
34165 source for the @sc{gnu} command-line interface
34167 @item gdb-@value{GDBVN}/glob
34168 source for the @sc{gnu} filename pattern-matching subroutine
34170 @item gdb-@value{GDBVN}/mmalloc
34171 source for the @sc{gnu} memory-mapped malloc package
34174 The simplest way to configure and build @value{GDBN} is to run @file{configure}
34175 from the @file{gdb-@var{version-number}} source directory, which in
34176 this example is the @file{gdb-@value{GDBVN}} directory.
34178 First switch to the @file{gdb-@var{version-number}} source directory
34179 if you are not already in it; then run @file{configure}. Pass the
34180 identifier for the platform on which @value{GDBN} will run as an
34186 cd gdb-@value{GDBVN}
34187 ./configure @var{host}
34192 where @var{host} is an identifier such as @samp{sun4} or
34193 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
34194 (You can often leave off @var{host}; @file{configure} tries to guess the
34195 correct value by examining your system.)
34197 Running @samp{configure @var{host}} and then running @code{make} builds the
34198 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
34199 libraries, then @code{gdb} itself. The configured source files, and the
34200 binaries, are left in the corresponding source directories.
34203 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
34204 system does not recognize this automatically when you run a different
34205 shell, you may need to run @code{sh} on it explicitly:
34208 sh configure @var{host}
34211 If you run @file{configure} from a directory that contains source
34212 directories for multiple libraries or programs, such as the
34213 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
34215 creates configuration files for every directory level underneath (unless
34216 you tell it not to, with the @samp{--norecursion} option).
34218 You should run the @file{configure} script from the top directory in the
34219 source tree, the @file{gdb-@var{version-number}} directory. If you run
34220 @file{configure} from one of the subdirectories, you will configure only
34221 that subdirectory. That is usually not what you want. In particular,
34222 if you run the first @file{configure} from the @file{gdb} subdirectory
34223 of the @file{gdb-@var{version-number}} directory, you will omit the
34224 configuration of @file{bfd}, @file{readline}, and other sibling
34225 directories of the @file{gdb} subdirectory. This leads to build errors
34226 about missing include files such as @file{bfd/bfd.h}.
34228 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
34229 However, you should make sure that the shell on your path (named by
34230 the @samp{SHELL} environment variable) is publicly readable. Remember
34231 that @value{GDBN} uses the shell to start your program---some systems refuse to
34232 let @value{GDBN} debug child processes whose programs are not readable.
34234 @node Separate Objdir
34235 @section Compiling @value{GDBN} in Another Directory
34237 If you want to run @value{GDBN} versions for several host or target machines,
34238 you need a different @code{gdb} compiled for each combination of
34239 host and target. @file{configure} is designed to make this easy by
34240 allowing you to generate each configuration in a separate subdirectory,
34241 rather than in the source directory. If your @code{make} program
34242 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
34243 @code{make} in each of these directories builds the @code{gdb}
34244 program specified there.
34246 To build @code{gdb} in a separate directory, run @file{configure}
34247 with the @samp{--srcdir} option to specify where to find the source.
34248 (You also need to specify a path to find @file{configure}
34249 itself from your working directory. If the path to @file{configure}
34250 would be the same as the argument to @samp{--srcdir}, you can leave out
34251 the @samp{--srcdir} option; it is assumed.)
34253 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
34254 separate directory for a Sun 4 like this:
34258 cd gdb-@value{GDBVN}
34261 ../gdb-@value{GDBVN}/configure sun4
34266 When @file{configure} builds a configuration using a remote source
34267 directory, it creates a tree for the binaries with the same structure
34268 (and using the same names) as the tree under the source directory. In
34269 the example, you'd find the Sun 4 library @file{libiberty.a} in the
34270 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
34271 @file{gdb-sun4/gdb}.
34273 Make sure that your path to the @file{configure} script has just one
34274 instance of @file{gdb} in it. If your path to @file{configure} looks
34275 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
34276 one subdirectory of @value{GDBN}, not the whole package. This leads to
34277 build errors about missing include files such as @file{bfd/bfd.h}.
34279 One popular reason to build several @value{GDBN} configurations in separate
34280 directories is to configure @value{GDBN} for cross-compiling (where
34281 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
34282 programs that run on another machine---the @dfn{target}).
34283 You specify a cross-debugging target by
34284 giving the @samp{--target=@var{target}} option to @file{configure}.
34286 When you run @code{make} to build a program or library, you must run
34287 it in a configured directory---whatever directory you were in when you
34288 called @file{configure} (or one of its subdirectories).
34290 The @code{Makefile} that @file{configure} generates in each source
34291 directory also runs recursively. If you type @code{make} in a source
34292 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
34293 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
34294 will build all the required libraries, and then build GDB.
34296 When you have multiple hosts or targets configured in separate
34297 directories, you can run @code{make} on them in parallel (for example,
34298 if they are NFS-mounted on each of the hosts); they will not interfere
34302 @section Specifying Names for Hosts and Targets
34304 The specifications used for hosts and targets in the @file{configure}
34305 script are based on a three-part naming scheme, but some short predefined
34306 aliases are also supported. The full naming scheme encodes three pieces
34307 of information in the following pattern:
34310 @var{architecture}-@var{vendor}-@var{os}
34313 For example, you can use the alias @code{sun4} as a @var{host} argument,
34314 or as the value for @var{target} in a @code{--target=@var{target}}
34315 option. The equivalent full name is @samp{sparc-sun-sunos4}.
34317 The @file{configure} script accompanying @value{GDBN} does not provide
34318 any query facility to list all supported host and target names or
34319 aliases. @file{configure} calls the Bourne shell script
34320 @code{config.sub} to map abbreviations to full names; you can read the
34321 script, if you wish, or you can use it to test your guesses on
34322 abbreviations---for example:
34325 % sh config.sub i386-linux
34327 % sh config.sub alpha-linux
34328 alpha-unknown-linux-gnu
34329 % sh config.sub hp9k700
34331 % sh config.sub sun4
34332 sparc-sun-sunos4.1.1
34333 % sh config.sub sun3
34334 m68k-sun-sunos4.1.1
34335 % sh config.sub i986v
34336 Invalid configuration `i986v': machine `i986v' not recognized
34340 @code{config.sub} is also distributed in the @value{GDBN} source
34341 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
34343 @node Configure Options
34344 @section @file{configure} Options
34346 Here is a summary of the @file{configure} options and arguments that
34347 are most often useful for building @value{GDBN}. @file{configure} also has
34348 several other options not listed here. @inforef{What Configure
34349 Does,,configure.info}, for a full explanation of @file{configure}.
34352 configure @r{[}--help@r{]}
34353 @r{[}--prefix=@var{dir}@r{]}
34354 @r{[}--exec-prefix=@var{dir}@r{]}
34355 @r{[}--srcdir=@var{dirname}@r{]}
34356 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
34357 @r{[}--target=@var{target}@r{]}
34362 You may introduce options with a single @samp{-} rather than
34363 @samp{--} if you prefer; but you may abbreviate option names if you use
34368 Display a quick summary of how to invoke @file{configure}.
34370 @item --prefix=@var{dir}
34371 Configure the source to install programs and files under directory
34374 @item --exec-prefix=@var{dir}
34375 Configure the source to install programs under directory
34378 @c avoid splitting the warning from the explanation:
34380 @item --srcdir=@var{dirname}
34381 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
34382 @code{make} that implements the @code{VPATH} feature.}@*
34383 Use this option to make configurations in directories separate from the
34384 @value{GDBN} source directories. Among other things, you can use this to
34385 build (or maintain) several configurations simultaneously, in separate
34386 directories. @file{configure} writes configuration-specific files in
34387 the current directory, but arranges for them to use the source in the
34388 directory @var{dirname}. @file{configure} creates directories under
34389 the working directory in parallel to the source directories below
34392 @item --norecursion
34393 Configure only the directory level where @file{configure} is executed; do not
34394 propagate configuration to subdirectories.
34396 @item --target=@var{target}
34397 Configure @value{GDBN} for cross-debugging programs running on the specified
34398 @var{target}. Without this option, @value{GDBN} is configured to debug
34399 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
34401 There is no convenient way to generate a list of all available targets.
34403 @item @var{host} @dots{}
34404 Configure @value{GDBN} to run on the specified @var{host}.
34406 There is no convenient way to generate a list of all available hosts.
34409 There are many other options available as well, but they are generally
34410 needed for special purposes only.
34412 @node System-wide configuration
34413 @section System-wide configuration and settings
34414 @cindex system-wide init file
34416 @value{GDBN} can be configured to have a system-wide init file;
34417 this file will be read and executed at startup (@pxref{Startup, , What
34418 @value{GDBN} does during startup}).
34420 Here is the corresponding configure option:
34423 @item --with-system-gdbinit=@var{file}
34424 Specify that the default location of the system-wide init file is
34428 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
34429 it may be subject to relocation. Two possible cases:
34433 If the default location of this init file contains @file{$prefix},
34434 it will be subject to relocation. Suppose that the configure options
34435 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
34436 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
34437 init file is looked for as @file{$install/etc/gdbinit} instead of
34438 @file{$prefix/etc/gdbinit}.
34441 By contrast, if the default location does not contain the prefix,
34442 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
34443 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
34444 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
34445 wherever @value{GDBN} is installed.
34448 @node Maintenance Commands
34449 @appendix Maintenance Commands
34450 @cindex maintenance commands
34451 @cindex internal commands
34453 In addition to commands intended for @value{GDBN} users, @value{GDBN}
34454 includes a number of commands intended for @value{GDBN} developers,
34455 that are not documented elsewhere in this manual. These commands are
34456 provided here for reference. (For commands that turn on debugging
34457 messages, see @ref{Debugging Output}.)
34460 @kindex maint agent
34461 @kindex maint agent-eval
34462 @item maint agent @var{expression}
34463 @itemx maint agent-eval @var{expression}
34464 Translate the given @var{expression} into remote agent bytecodes.
34465 This command is useful for debugging the Agent Expression mechanism
34466 (@pxref{Agent Expressions}). The @samp{agent} version produces an
34467 expression useful for data collection, such as by tracepoints, while
34468 @samp{maint agent-eval} produces an expression that evaluates directly
34469 to a result. For instance, a collection expression for @code{globa +
34470 globb} will include bytecodes to record four bytes of memory at each
34471 of the addresses of @code{globa} and @code{globb}, while discarding
34472 the result of the addition, while an evaluation expression will do the
34473 addition and return the sum.
34475 @kindex maint agent-printf
34476 @item maint agent-printf @var{format},@var{expr},...
34477 Translate the given format string and list of argument expressions
34478 into remote agent bytecodes and display them as a disassembled list.
34479 This command is useful for debugging the agent version of dynamic
34480 printf (@pxref{Dynamic Printf}.
34482 @kindex maint info breakpoints
34483 @item @anchor{maint info breakpoints}maint info breakpoints
34484 Using the same format as @samp{info breakpoints}, display both the
34485 breakpoints you've set explicitly, and those @value{GDBN} is using for
34486 internal purposes. Internal breakpoints are shown with negative
34487 breakpoint numbers. The type column identifies what kind of breakpoint
34492 Normal, explicitly set breakpoint.
34495 Normal, explicitly set watchpoint.
34498 Internal breakpoint, used to handle correctly stepping through
34499 @code{longjmp} calls.
34501 @item longjmp resume
34502 Internal breakpoint at the target of a @code{longjmp}.
34505 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
34508 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
34511 Shared library events.
34515 @kindex set displaced-stepping
34516 @kindex show displaced-stepping
34517 @cindex displaced stepping support
34518 @cindex out-of-line single-stepping
34519 @item set displaced-stepping
34520 @itemx show displaced-stepping
34521 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
34522 if the target supports it. Displaced stepping is a way to single-step
34523 over breakpoints without removing them from the inferior, by executing
34524 an out-of-line copy of the instruction that was originally at the
34525 breakpoint location. It is also known as out-of-line single-stepping.
34528 @item set displaced-stepping on
34529 If the target architecture supports it, @value{GDBN} will use
34530 displaced stepping to step over breakpoints.
34532 @item set displaced-stepping off
34533 @value{GDBN} will not use displaced stepping to step over breakpoints,
34534 even if such is supported by the target architecture.
34536 @cindex non-stop mode, and @samp{set displaced-stepping}
34537 @item set displaced-stepping auto
34538 This is the default mode. @value{GDBN} will use displaced stepping
34539 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
34540 architecture supports displaced stepping.
34543 @kindex maint check-symtabs
34544 @item maint check-symtabs
34545 Check the consistency of psymtabs and symtabs.
34547 @kindex maint cplus first_component
34548 @item maint cplus first_component @var{name}
34549 Print the first C@t{++} class/namespace component of @var{name}.
34551 @kindex maint cplus namespace
34552 @item maint cplus namespace
34553 Print the list of possible C@t{++} namespaces.
34555 @kindex maint demangle
34556 @item maint demangle @var{name}
34557 Demangle a C@t{++} or Objective-C mangled @var{name}.
34559 @kindex maint deprecate
34560 @kindex maint undeprecate
34561 @cindex deprecated commands
34562 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
34563 @itemx maint undeprecate @var{command}
34564 Deprecate or undeprecate the named @var{command}. Deprecated commands
34565 cause @value{GDBN} to issue a warning when you use them. The optional
34566 argument @var{replacement} says which newer command should be used in
34567 favor of the deprecated one; if it is given, @value{GDBN} will mention
34568 the replacement as part of the warning.
34570 @kindex maint dump-me
34571 @item maint dump-me
34572 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
34573 Cause a fatal signal in the debugger and force it to dump its core.
34574 This is supported only on systems which support aborting a program
34575 with the @code{SIGQUIT} signal.
34577 @kindex maint internal-error
34578 @kindex maint internal-warning
34579 @item maint internal-error @r{[}@var{message-text}@r{]}
34580 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
34581 Cause @value{GDBN} to call the internal function @code{internal_error}
34582 or @code{internal_warning} and hence behave as though an internal error
34583 or internal warning has been detected. In addition to reporting the
34584 internal problem, these functions give the user the opportunity to
34585 either quit @value{GDBN} or create a core file of the current
34586 @value{GDBN} session.
34588 These commands take an optional parameter @var{message-text} that is
34589 used as the text of the error or warning message.
34591 Here's an example of using @code{internal-error}:
34594 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
34595 @dots{}/maint.c:121: internal-error: testing, 1, 2
34596 A problem internal to GDB has been detected. Further
34597 debugging may prove unreliable.
34598 Quit this debugging session? (y or n) @kbd{n}
34599 Create a core file? (y or n) @kbd{n}
34603 @cindex @value{GDBN} internal error
34604 @cindex internal errors, control of @value{GDBN} behavior
34606 @kindex maint set internal-error
34607 @kindex maint show internal-error
34608 @kindex maint set internal-warning
34609 @kindex maint show internal-warning
34610 @item maint set internal-error @var{action} [ask|yes|no]
34611 @itemx maint show internal-error @var{action}
34612 @itemx maint set internal-warning @var{action} [ask|yes|no]
34613 @itemx maint show internal-warning @var{action}
34614 When @value{GDBN} reports an internal problem (error or warning) it
34615 gives the user the opportunity to both quit @value{GDBN} and create a
34616 core file of the current @value{GDBN} session. These commands let you
34617 override the default behaviour for each particular @var{action},
34618 described in the table below.
34622 You can specify that @value{GDBN} should always (yes) or never (no)
34623 quit. The default is to ask the user what to do.
34626 You can specify that @value{GDBN} should always (yes) or never (no)
34627 create a core file. The default is to ask the user what to do.
34630 @kindex maint packet
34631 @item maint packet @var{text}
34632 If @value{GDBN} is talking to an inferior via the serial protocol,
34633 then this command sends the string @var{text} to the inferior, and
34634 displays the response packet. @value{GDBN} supplies the initial
34635 @samp{$} character, the terminating @samp{#} character, and the
34638 @kindex maint print architecture
34639 @item maint print architecture @r{[}@var{file}@r{]}
34640 Print the entire architecture configuration. The optional argument
34641 @var{file} names the file where the output goes.
34643 @kindex maint print c-tdesc
34644 @item maint print c-tdesc
34645 Print the current target description (@pxref{Target Descriptions}) as
34646 a C source file. The created source file can be used in @value{GDBN}
34647 when an XML parser is not available to parse the description.
34649 @kindex maint print dummy-frames
34650 @item maint print dummy-frames
34651 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
34654 (@value{GDBP}) @kbd{b add}
34656 (@value{GDBP}) @kbd{print add(2,3)}
34657 Breakpoint 2, add (a=2, b=3) at @dots{}
34659 The program being debugged stopped while in a function called from GDB.
34661 (@value{GDBP}) @kbd{maint print dummy-frames}
34662 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
34663 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
34664 call_lo=0x01014000 call_hi=0x01014001
34668 Takes an optional file parameter.
34670 @kindex maint print registers
34671 @kindex maint print raw-registers
34672 @kindex maint print cooked-registers
34673 @kindex maint print register-groups
34674 @kindex maint print remote-registers
34675 @item maint print registers @r{[}@var{file}@r{]}
34676 @itemx maint print raw-registers @r{[}@var{file}@r{]}
34677 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
34678 @itemx maint print register-groups @r{[}@var{file}@r{]}
34679 @itemx maint print remote-registers @r{[}@var{file}@r{]}
34680 Print @value{GDBN}'s internal register data structures.
34682 The command @code{maint print raw-registers} includes the contents of
34683 the raw register cache; the command @code{maint print
34684 cooked-registers} includes the (cooked) value of all registers,
34685 including registers which aren't available on the target nor visible
34686 to user; the command @code{maint print register-groups} includes the
34687 groups that each register is a member of; and the command @code{maint
34688 print remote-registers} includes the remote target's register numbers
34689 and offsets in the `G' packets. @xref{Registers,, Registers, gdbint,
34690 @value{GDBN} Internals}.
34692 These commands take an optional parameter, a file name to which to
34693 write the information.
34695 @kindex maint print reggroups
34696 @item maint print reggroups @r{[}@var{file}@r{]}
34697 Print @value{GDBN}'s internal register group data structures. The
34698 optional argument @var{file} tells to what file to write the
34701 The register groups info looks like this:
34704 (@value{GDBP}) @kbd{maint print reggroups}
34717 This command forces @value{GDBN} to flush its internal register cache.
34719 @kindex maint print objfiles
34720 @cindex info for known object files
34721 @item maint print objfiles
34722 Print a dump of all known object files. For each object file, this
34723 command prints its name, address in memory, and all of its psymtabs
34726 @kindex maint print section-scripts
34727 @cindex info for known .debug_gdb_scripts-loaded scripts
34728 @item maint print section-scripts [@var{regexp}]
34729 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
34730 If @var{regexp} is specified, only print scripts loaded by object files
34731 matching @var{regexp}.
34732 For each script, this command prints its name as specified in the objfile,
34733 and the full path if known.
34734 @xref{dotdebug_gdb_scripts section}.
34736 @kindex maint print statistics
34737 @cindex bcache statistics
34738 @item maint print statistics
34739 This command prints, for each object file in the program, various data
34740 about that object file followed by the byte cache (@dfn{bcache})
34741 statistics for the object file. The objfile data includes the number
34742 of minimal, partial, full, and stabs symbols, the number of types
34743 defined by the objfile, the number of as yet unexpanded psym tables,
34744 the number of line tables and string tables, and the amount of memory
34745 used by the various tables. The bcache statistics include the counts,
34746 sizes, and counts of duplicates of all and unique objects, max,
34747 average, and median entry size, total memory used and its overhead and
34748 savings, and various measures of the hash table size and chain
34751 @kindex maint print target-stack
34752 @cindex target stack description
34753 @item maint print target-stack
34754 A @dfn{target} is an interface between the debugger and a particular
34755 kind of file or process. Targets can be stacked in @dfn{strata},
34756 so that more than one target can potentially respond to a request.
34757 In particular, memory accesses will walk down the stack of targets
34758 until they find a target that is interested in handling that particular
34761 This command prints a short description of each layer that was pushed on
34762 the @dfn{target stack}, starting from the top layer down to the bottom one.
34764 @kindex maint print type
34765 @cindex type chain of a data type
34766 @item maint print type @var{expr}
34767 Print the type chain for a type specified by @var{expr}. The argument
34768 can be either a type name or a symbol. If it is a symbol, the type of
34769 that symbol is described. The type chain produced by this command is
34770 a recursive definition of the data type as stored in @value{GDBN}'s
34771 data structures, including its flags and contained types.
34773 @kindex maint set dwarf2 always-disassemble
34774 @kindex maint show dwarf2 always-disassemble
34775 @item maint set dwarf2 always-disassemble
34776 @item maint show dwarf2 always-disassemble
34777 Control the behavior of @code{info address} when using DWARF debugging
34780 The default is @code{off}, which means that @value{GDBN} should try to
34781 describe a variable's location in an easily readable format. When
34782 @code{on}, @value{GDBN} will instead display the DWARF location
34783 expression in an assembly-like format. Note that some locations are
34784 too complex for @value{GDBN} to describe simply; in this case you will
34785 always see the disassembly form.
34787 Here is an example of the resulting disassembly:
34790 (gdb) info addr argc
34791 Symbol "argc" is a complex DWARF expression:
34795 For more information on these expressions, see
34796 @uref{http://www.dwarfstd.org/, the DWARF standard}.
34798 @kindex maint set dwarf2 max-cache-age
34799 @kindex maint show dwarf2 max-cache-age
34800 @item maint set dwarf2 max-cache-age
34801 @itemx maint show dwarf2 max-cache-age
34802 Control the DWARF 2 compilation unit cache.
34804 @cindex DWARF 2 compilation units cache
34805 In object files with inter-compilation-unit references, such as those
34806 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
34807 reader needs to frequently refer to previously read compilation units.
34808 This setting controls how long a compilation unit will remain in the
34809 cache if it is not referenced. A higher limit means that cached
34810 compilation units will be stored in memory longer, and more total
34811 memory will be used. Setting it to zero disables caching, which will
34812 slow down @value{GDBN} startup, but reduce memory consumption.
34814 @kindex maint set profile
34815 @kindex maint show profile
34816 @cindex profiling GDB
34817 @item maint set profile
34818 @itemx maint show profile
34819 Control profiling of @value{GDBN}.
34821 Profiling will be disabled until you use the @samp{maint set profile}
34822 command to enable it. When you enable profiling, the system will begin
34823 collecting timing and execution count data; when you disable profiling or
34824 exit @value{GDBN}, the results will be written to a log file. Remember that
34825 if you use profiling, @value{GDBN} will overwrite the profiling log file
34826 (often called @file{gmon.out}). If you have a record of important profiling
34827 data in a @file{gmon.out} file, be sure to move it to a safe location.
34829 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
34830 compiled with the @samp{-pg} compiler option.
34832 @kindex maint set show-debug-regs
34833 @kindex maint show show-debug-regs
34834 @cindex hardware debug registers
34835 @item maint set show-debug-regs
34836 @itemx maint show show-debug-regs
34837 Control whether to show variables that mirror the hardware debug
34838 registers. Use @code{ON} to enable, @code{OFF} to disable. If
34839 enabled, the debug registers values are shown when @value{GDBN} inserts or
34840 removes a hardware breakpoint or watchpoint, and when the inferior
34841 triggers a hardware-assisted breakpoint or watchpoint.
34843 @kindex maint set show-all-tib
34844 @kindex maint show show-all-tib
34845 @item maint set show-all-tib
34846 @itemx maint show show-all-tib
34847 Control whether to show all non zero areas within a 1k block starting
34848 at thread local base, when using the @samp{info w32 thread-information-block}
34851 @kindex maint space
34852 @cindex memory used by commands
34854 Control whether to display memory usage for each command. If set to a
34855 nonzero value, @value{GDBN} will display how much memory each command
34856 took, following the command's own output. This can also be requested
34857 by invoking @value{GDBN} with the @option{--statistics} command-line
34858 switch (@pxref{Mode Options}).
34861 @cindex time of command execution
34863 Control whether to display the execution time of @value{GDBN} for each command.
34864 If set to a nonzero value, @value{GDBN} will display how much time it
34865 took to execute each command, following the command's own output.
34866 Both CPU time and wallclock time are printed.
34867 Printing both is useful when trying to determine whether the cost is
34868 CPU or, e.g., disk/network, latency.
34869 Note that the CPU time printed is for @value{GDBN} only, it does not include
34870 the execution time of the inferior because there's no mechanism currently
34871 to compute how much time was spent by @value{GDBN} and how much time was
34872 spent by the program been debugged.
34873 This can also be requested by invoking @value{GDBN} with the
34874 @option{--statistics} command-line switch (@pxref{Mode Options}).
34876 @kindex maint translate-address
34877 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
34878 Find the symbol stored at the location specified by the address
34879 @var{addr} and an optional section name @var{section}. If found,
34880 @value{GDBN} prints the name of the closest symbol and an offset from
34881 the symbol's location to the specified address. This is similar to
34882 the @code{info address} command (@pxref{Symbols}), except that this
34883 command also allows to find symbols in other sections.
34885 If section was not specified, the section in which the symbol was found
34886 is also printed. For dynamically linked executables, the name of
34887 executable or shared library containing the symbol is printed as well.
34891 The following command is useful for non-interactive invocations of
34892 @value{GDBN}, such as in the test suite.
34895 @item set watchdog @var{nsec}
34896 @kindex set watchdog
34897 @cindex watchdog timer
34898 @cindex timeout for commands
34899 Set the maximum number of seconds @value{GDBN} will wait for the
34900 target operation to finish. If this time expires, @value{GDBN}
34901 reports and error and the command is aborted.
34903 @item show watchdog
34904 Show the current setting of the target wait timeout.
34907 @node Remote Protocol
34908 @appendix @value{GDBN} Remote Serial Protocol
34913 * Stop Reply Packets::
34914 * General Query Packets::
34915 * Architecture-Specific Protocol Details::
34916 * Tracepoint Packets::
34917 * Host I/O Packets::
34919 * Notification Packets::
34920 * Remote Non-Stop::
34921 * Packet Acknowledgment::
34923 * File-I/O Remote Protocol Extension::
34924 * Library List Format::
34925 * Library List Format for SVR4 Targets::
34926 * Memory Map Format::
34927 * Thread List Format::
34928 * Traceframe Info Format::
34934 There may be occasions when you need to know something about the
34935 protocol---for example, if there is only one serial port to your target
34936 machine, you might want your program to do something special if it
34937 recognizes a packet meant for @value{GDBN}.
34939 In the examples below, @samp{->} and @samp{<-} are used to indicate
34940 transmitted and received data, respectively.
34942 @cindex protocol, @value{GDBN} remote serial
34943 @cindex serial protocol, @value{GDBN} remote
34944 @cindex remote serial protocol
34945 All @value{GDBN} commands and responses (other than acknowledgments
34946 and notifications, see @ref{Notification Packets}) are sent as a
34947 @var{packet}. A @var{packet} is introduced with the character
34948 @samp{$}, the actual @var{packet-data}, and the terminating character
34949 @samp{#} followed by a two-digit @var{checksum}:
34952 @code{$}@var{packet-data}@code{#}@var{checksum}
34956 @cindex checksum, for @value{GDBN} remote
34958 The two-digit @var{checksum} is computed as the modulo 256 sum of all
34959 characters between the leading @samp{$} and the trailing @samp{#} (an
34960 eight bit unsigned checksum).
34962 Implementors should note that prior to @value{GDBN} 5.0 the protocol
34963 specification also included an optional two-digit @var{sequence-id}:
34966 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
34969 @cindex sequence-id, for @value{GDBN} remote
34971 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
34972 has never output @var{sequence-id}s. Stubs that handle packets added
34973 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
34975 When either the host or the target machine receives a packet, the first
34976 response expected is an acknowledgment: either @samp{+} (to indicate
34977 the package was received correctly) or @samp{-} (to request
34981 -> @code{$}@var{packet-data}@code{#}@var{checksum}
34986 The @samp{+}/@samp{-} acknowledgments can be disabled
34987 once a connection is established.
34988 @xref{Packet Acknowledgment}, for details.
34990 The host (@value{GDBN}) sends @var{command}s, and the target (the
34991 debugging stub incorporated in your program) sends a @var{response}. In
34992 the case of step and continue @var{command}s, the response is only sent
34993 when the operation has completed, and the target has again stopped all
34994 threads in all attached processes. This is the default all-stop mode
34995 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
34996 execution mode; see @ref{Remote Non-Stop}, for details.
34998 @var{packet-data} consists of a sequence of characters with the
34999 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
35002 @cindex remote protocol, field separator
35003 Fields within the packet should be separated using @samp{,} @samp{;} or
35004 @samp{:}. Except where otherwise noted all numbers are represented in
35005 @sc{hex} with leading zeros suppressed.
35007 Implementors should note that prior to @value{GDBN} 5.0, the character
35008 @samp{:} could not appear as the third character in a packet (as it
35009 would potentially conflict with the @var{sequence-id}).
35011 @cindex remote protocol, binary data
35012 @anchor{Binary Data}
35013 Binary data in most packets is encoded either as two hexadecimal
35014 digits per byte of binary data. This allowed the traditional remote
35015 protocol to work over connections which were only seven-bit clean.
35016 Some packets designed more recently assume an eight-bit clean
35017 connection, and use a more efficient encoding to send and receive
35020 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
35021 as an escape character. Any escaped byte is transmitted as the escape
35022 character followed by the original character XORed with @code{0x20}.
35023 For example, the byte @code{0x7d} would be transmitted as the two
35024 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
35025 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
35026 @samp{@}}) must always be escaped. Responses sent by the stub
35027 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
35028 is not interpreted as the start of a run-length encoded sequence
35031 Response @var{data} can be run-length encoded to save space.
35032 Run-length encoding replaces runs of identical characters with one
35033 instance of the repeated character, followed by a @samp{*} and a
35034 repeat count. The repeat count is itself sent encoded, to avoid
35035 binary characters in @var{data}: a value of @var{n} is sent as
35036 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
35037 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
35038 code 32) for a repeat count of 3. (This is because run-length
35039 encoding starts to win for counts 3 or more.) Thus, for example,
35040 @samp{0* } is a run-length encoding of ``0000'': the space character
35041 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
35044 The printable characters @samp{#} and @samp{$} or with a numeric value
35045 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
35046 seven repeats (@samp{$}) can be expanded using a repeat count of only
35047 five (@samp{"}). For example, @samp{00000000} can be encoded as
35050 The error response returned for some packets includes a two character
35051 error number. That number is not well defined.
35053 @cindex empty response, for unsupported packets
35054 For any @var{command} not supported by the stub, an empty response
35055 (@samp{$#00}) should be returned. That way it is possible to extend the
35056 protocol. A newer @value{GDBN} can tell if a packet is supported based
35059 At a minimum, a stub is required to support the @samp{g} and @samp{G}
35060 commands for register access, and the @samp{m} and @samp{M} commands
35061 for memory access. Stubs that only control single-threaded targets
35062 can implement run control with the @samp{c} (continue), and @samp{s}
35063 (step) commands. Stubs that support multi-threading targets should
35064 support the @samp{vCont} command. All other commands are optional.
35069 The following table provides a complete list of all currently defined
35070 @var{command}s and their corresponding response @var{data}.
35071 @xref{File-I/O Remote Protocol Extension}, for details about the File
35072 I/O extension of the remote protocol.
35074 Each packet's description has a template showing the packet's overall
35075 syntax, followed by an explanation of the packet's meaning. We
35076 include spaces in some of the templates for clarity; these are not
35077 part of the packet's syntax. No @value{GDBN} packet uses spaces to
35078 separate its components. For example, a template like @samp{foo
35079 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
35080 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
35081 @var{baz}. @value{GDBN} does not transmit a space character between the
35082 @samp{foo} and the @var{bar}, or between the @var{bar} and the
35085 @cindex @var{thread-id}, in remote protocol
35086 @anchor{thread-id syntax}
35087 Several packets and replies include a @var{thread-id} field to identify
35088 a thread. Normally these are positive numbers with a target-specific
35089 interpretation, formatted as big-endian hex strings. A @var{thread-id}
35090 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
35093 In addition, the remote protocol supports a multiprocess feature in
35094 which the @var{thread-id} syntax is extended to optionally include both
35095 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
35096 The @var{pid} (process) and @var{tid} (thread) components each have the
35097 format described above: a positive number with target-specific
35098 interpretation formatted as a big-endian hex string, literal @samp{-1}
35099 to indicate all processes or threads (respectively), or @samp{0} to
35100 indicate an arbitrary process or thread. Specifying just a process, as
35101 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
35102 error to specify all processes but a specific thread, such as
35103 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
35104 for those packets and replies explicitly documented to include a process
35105 ID, rather than a @var{thread-id}.
35107 The multiprocess @var{thread-id} syntax extensions are only used if both
35108 @value{GDBN} and the stub report support for the @samp{multiprocess}
35109 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
35112 Note that all packet forms beginning with an upper- or lower-case
35113 letter, other than those described here, are reserved for future use.
35115 Here are the packet descriptions.
35120 @cindex @samp{!} packet
35121 @anchor{extended mode}
35122 Enable extended mode. In extended mode, the remote server is made
35123 persistent. The @samp{R} packet is used to restart the program being
35129 The remote target both supports and has enabled extended mode.
35133 @cindex @samp{?} packet
35134 Indicate the reason the target halted. The reply is the same as for
35135 step and continue. This packet has a special interpretation when the
35136 target is in non-stop mode; see @ref{Remote Non-Stop}.
35139 @xref{Stop Reply Packets}, for the reply specifications.
35141 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
35142 @cindex @samp{A} packet
35143 Initialized @code{argv[]} array passed into program. @var{arglen}
35144 specifies the number of bytes in the hex encoded byte stream
35145 @var{arg}. See @code{gdbserver} for more details.
35150 The arguments were set.
35156 @cindex @samp{b} packet
35157 (Don't use this packet; its behavior is not well-defined.)
35158 Change the serial line speed to @var{baud}.
35160 JTC: @emph{When does the transport layer state change? When it's
35161 received, or after the ACK is transmitted. In either case, there are
35162 problems if the command or the acknowledgment packet is dropped.}
35164 Stan: @emph{If people really wanted to add something like this, and get
35165 it working for the first time, they ought to modify ser-unix.c to send
35166 some kind of out-of-band message to a specially-setup stub and have the
35167 switch happen "in between" packets, so that from remote protocol's point
35168 of view, nothing actually happened.}
35170 @item B @var{addr},@var{mode}
35171 @cindex @samp{B} packet
35172 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
35173 breakpoint at @var{addr}.
35175 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
35176 (@pxref{insert breakpoint or watchpoint packet}).
35178 @cindex @samp{bc} packet
35181 Backward continue. Execute the target system in reverse. No parameter.
35182 @xref{Reverse Execution}, for more information.
35185 @xref{Stop Reply Packets}, for the reply specifications.
35187 @cindex @samp{bs} packet
35190 Backward single step. Execute one instruction in reverse. No parameter.
35191 @xref{Reverse Execution}, for more information.
35194 @xref{Stop Reply Packets}, for the reply specifications.
35196 @item c @r{[}@var{addr}@r{]}
35197 @cindex @samp{c} packet
35198 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
35199 resume at current address.
35201 This packet is deprecated for multi-threading support. @xref{vCont
35205 @xref{Stop Reply Packets}, for the reply specifications.
35207 @item C @var{sig}@r{[};@var{addr}@r{]}
35208 @cindex @samp{C} packet
35209 Continue with signal @var{sig} (hex signal number). If
35210 @samp{;@var{addr}} is omitted, resume at same address.
35212 This packet is deprecated for multi-threading support. @xref{vCont
35216 @xref{Stop Reply Packets}, for the reply specifications.
35219 @cindex @samp{d} packet
35222 Don't use this packet; instead, define a general set packet
35223 (@pxref{General Query Packets}).
35227 @cindex @samp{D} packet
35228 The first form of the packet is used to detach @value{GDBN} from the
35229 remote system. It is sent to the remote target
35230 before @value{GDBN} disconnects via the @code{detach} command.
35232 The second form, including a process ID, is used when multiprocess
35233 protocol extensions are enabled (@pxref{multiprocess extensions}), to
35234 detach only a specific process. The @var{pid} is specified as a
35235 big-endian hex string.
35245 @item F @var{RC},@var{EE},@var{CF};@var{XX}
35246 @cindex @samp{F} packet
35247 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
35248 This is part of the File-I/O protocol extension. @xref{File-I/O
35249 Remote Protocol Extension}, for the specification.
35252 @anchor{read registers packet}
35253 @cindex @samp{g} packet
35254 Read general registers.
35258 @item @var{XX@dots{}}
35259 Each byte of register data is described by two hex digits. The bytes
35260 with the register are transmitted in target byte order. The size of
35261 each register and their position within the @samp{g} packet are
35262 determined by the @value{GDBN} internal gdbarch functions
35263 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
35264 specification of several standard @samp{g} packets is specified below.
35266 When reading registers from a trace frame (@pxref{Analyze Collected
35267 Data,,Using the Collected Data}), the stub may also return a string of
35268 literal @samp{x}'s in place of the register data digits, to indicate
35269 that the corresponding register has not been collected, thus its value
35270 is unavailable. For example, for an architecture with 4 registers of
35271 4 bytes each, the following reply indicates to @value{GDBN} that
35272 registers 0 and 2 have not been collected, while registers 1 and 3
35273 have been collected, and both have zero value:
35277 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
35284 @item G @var{XX@dots{}}
35285 @cindex @samp{G} packet
35286 Write general registers. @xref{read registers packet}, for a
35287 description of the @var{XX@dots{}} data.
35297 @item H @var{op} @var{thread-id}
35298 @cindex @samp{H} packet
35299 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
35300 @samp{G}, et.al.). @var{op} depends on the operation to be performed:
35301 it should be @samp{c} for step and continue operations (note that this
35302 is deprecated, supporting the @samp{vCont} command is a better
35303 option), @samp{g} for other operations. The thread designator
35304 @var{thread-id} has the format and interpretation described in
35305 @ref{thread-id syntax}.
35316 @c 'H': How restrictive (or permissive) is the thread model. If a
35317 @c thread is selected and stopped, are other threads allowed
35318 @c to continue to execute? As I mentioned above, I think the
35319 @c semantics of each command when a thread is selected must be
35320 @c described. For example:
35322 @c 'g': If the stub supports threads and a specific thread is
35323 @c selected, returns the register block from that thread;
35324 @c otherwise returns current registers.
35326 @c 'G' If the stub supports threads and a specific thread is
35327 @c selected, sets the registers of the register block of
35328 @c that thread; otherwise sets current registers.
35330 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
35331 @anchor{cycle step packet}
35332 @cindex @samp{i} packet
35333 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
35334 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
35335 step starting at that address.
35338 @cindex @samp{I} packet
35339 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
35343 @cindex @samp{k} packet
35346 FIXME: @emph{There is no description of how to operate when a specific
35347 thread context has been selected (i.e.@: does 'k' kill only that
35350 @item m @var{addr},@var{length}
35351 @cindex @samp{m} packet
35352 Read @var{length} bytes of memory starting at address @var{addr}.
35353 Note that @var{addr} may not be aligned to any particular boundary.
35355 The stub need not use any particular size or alignment when gathering
35356 data from memory for the response; even if @var{addr} is word-aligned
35357 and @var{length} is a multiple of the word size, the stub is free to
35358 use byte accesses, or not. For this reason, this packet may not be
35359 suitable for accessing memory-mapped I/O devices.
35360 @cindex alignment of remote memory accesses
35361 @cindex size of remote memory accesses
35362 @cindex memory, alignment and size of remote accesses
35366 @item @var{XX@dots{}}
35367 Memory contents; each byte is transmitted as a two-digit hexadecimal
35368 number. The reply may contain fewer bytes than requested if the
35369 server was able to read only part of the region of memory.
35374 @item M @var{addr},@var{length}:@var{XX@dots{}}
35375 @cindex @samp{M} packet
35376 Write @var{length} bytes of memory starting at address @var{addr}.
35377 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
35378 hexadecimal number.
35385 for an error (this includes the case where only part of the data was
35390 @cindex @samp{p} packet
35391 Read the value of register @var{n}; @var{n} is in hex.
35392 @xref{read registers packet}, for a description of how the returned
35393 register value is encoded.
35397 @item @var{XX@dots{}}
35398 the register's value
35402 Indicating an unrecognized @var{query}.
35405 @item P @var{n@dots{}}=@var{r@dots{}}
35406 @anchor{write register packet}
35407 @cindex @samp{P} packet
35408 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
35409 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
35410 digits for each byte in the register (target byte order).
35420 @item q @var{name} @var{params}@dots{}
35421 @itemx Q @var{name} @var{params}@dots{}
35422 @cindex @samp{q} packet
35423 @cindex @samp{Q} packet
35424 General query (@samp{q}) and set (@samp{Q}). These packets are
35425 described fully in @ref{General Query Packets}.
35428 @cindex @samp{r} packet
35429 Reset the entire system.
35431 Don't use this packet; use the @samp{R} packet instead.
35434 @cindex @samp{R} packet
35435 Restart the program being debugged. @var{XX}, while needed, is ignored.
35436 This packet is only available in extended mode (@pxref{extended mode}).
35438 The @samp{R} packet has no reply.
35440 @item s @r{[}@var{addr}@r{]}
35441 @cindex @samp{s} packet
35442 Single step. @var{addr} is the address at which to resume. If
35443 @var{addr} is omitted, resume at same address.
35445 This packet is deprecated for multi-threading support. @xref{vCont
35449 @xref{Stop Reply Packets}, for the reply specifications.
35451 @item S @var{sig}@r{[};@var{addr}@r{]}
35452 @anchor{step with signal packet}
35453 @cindex @samp{S} packet
35454 Step with signal. This is analogous to the @samp{C} packet, but
35455 requests a single-step, rather than a normal resumption of execution.
35457 This packet is deprecated for multi-threading support. @xref{vCont
35461 @xref{Stop Reply Packets}, for the reply specifications.
35463 @item t @var{addr}:@var{PP},@var{MM}
35464 @cindex @samp{t} packet
35465 Search backwards starting at address @var{addr} for a match with pattern
35466 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
35467 @var{addr} must be at least 3 digits.
35469 @item T @var{thread-id}
35470 @cindex @samp{T} packet
35471 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
35476 thread is still alive
35482 Packets starting with @samp{v} are identified by a multi-letter name,
35483 up to the first @samp{;} or @samp{?} (or the end of the packet).
35485 @item vAttach;@var{pid}
35486 @cindex @samp{vAttach} packet
35487 Attach to a new process with the specified process ID @var{pid}.
35488 The process ID is a
35489 hexadecimal integer identifying the process. In all-stop mode, all
35490 threads in the attached process are stopped; in non-stop mode, it may be
35491 attached without being stopped if that is supported by the target.
35493 @c In non-stop mode, on a successful vAttach, the stub should set the
35494 @c current thread to a thread of the newly-attached process. After
35495 @c attaching, GDB queries for the attached process's thread ID with qC.
35496 @c Also note that, from a user perspective, whether or not the
35497 @c target is stopped on attach in non-stop mode depends on whether you
35498 @c use the foreground or background version of the attach command, not
35499 @c on what vAttach does; GDB does the right thing with respect to either
35500 @c stopping or restarting threads.
35502 This packet is only available in extended mode (@pxref{extended mode}).
35508 @item @r{Any stop packet}
35509 for success in all-stop mode (@pxref{Stop Reply Packets})
35511 for success in non-stop mode (@pxref{Remote Non-Stop})
35514 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
35515 @cindex @samp{vCont} packet
35516 @anchor{vCont packet}
35517 Resume the inferior, specifying different actions for each thread.
35518 If an action is specified with no @var{thread-id}, then it is applied to any
35519 threads that don't have a specific action specified; if no default action is
35520 specified then other threads should remain stopped in all-stop mode and
35521 in their current state in non-stop mode.
35522 Specifying multiple
35523 default actions is an error; specifying no actions is also an error.
35524 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
35526 Currently supported actions are:
35532 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
35536 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
35541 The optional argument @var{addr} normally associated with the
35542 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
35543 not supported in @samp{vCont}.
35545 The @samp{t} action is only relevant in non-stop mode
35546 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
35547 A stop reply should be generated for any affected thread not already stopped.
35548 When a thread is stopped by means of a @samp{t} action,
35549 the corresponding stop reply should indicate that the thread has stopped with
35550 signal @samp{0}, regardless of whether the target uses some other signal
35551 as an implementation detail.
35553 The stub must support @samp{vCont} if it reports support for
35554 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
35555 this case @samp{vCont} actions can be specified to apply to all threads
35556 in a process by using the @samp{p@var{pid}.-1} form of the
35560 @xref{Stop Reply Packets}, for the reply specifications.
35563 @cindex @samp{vCont?} packet
35564 Request a list of actions supported by the @samp{vCont} packet.
35568 @item vCont@r{[};@var{action}@dots{}@r{]}
35569 The @samp{vCont} packet is supported. Each @var{action} is a supported
35570 command in the @samp{vCont} packet.
35572 The @samp{vCont} packet is not supported.
35575 @item vFile:@var{operation}:@var{parameter}@dots{}
35576 @cindex @samp{vFile} packet
35577 Perform a file operation on the target system. For details,
35578 see @ref{Host I/O Packets}.
35580 @item vFlashErase:@var{addr},@var{length}
35581 @cindex @samp{vFlashErase} packet
35582 Direct the stub to erase @var{length} bytes of flash starting at
35583 @var{addr}. The region may enclose any number of flash blocks, but
35584 its start and end must fall on block boundaries, as indicated by the
35585 flash block size appearing in the memory map (@pxref{Memory Map
35586 Format}). @value{GDBN} groups flash memory programming operations
35587 together, and sends a @samp{vFlashDone} request after each group; the
35588 stub is allowed to delay erase operation until the @samp{vFlashDone}
35589 packet is received.
35599 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
35600 @cindex @samp{vFlashWrite} packet
35601 Direct the stub to write data to flash address @var{addr}. The data
35602 is passed in binary form using the same encoding as for the @samp{X}
35603 packet (@pxref{Binary Data}). The memory ranges specified by
35604 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
35605 not overlap, and must appear in order of increasing addresses
35606 (although @samp{vFlashErase} packets for higher addresses may already
35607 have been received; the ordering is guaranteed only between
35608 @samp{vFlashWrite} packets). If a packet writes to an address that was
35609 neither erased by a preceding @samp{vFlashErase} packet nor by some other
35610 target-specific method, the results are unpredictable.
35618 for vFlashWrite addressing non-flash memory
35624 @cindex @samp{vFlashDone} packet
35625 Indicate to the stub that flash programming operation is finished.
35626 The stub is permitted to delay or batch the effects of a group of
35627 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
35628 @samp{vFlashDone} packet is received. The contents of the affected
35629 regions of flash memory are unpredictable until the @samp{vFlashDone}
35630 request is completed.
35632 @item vKill;@var{pid}
35633 @cindex @samp{vKill} packet
35634 Kill the process with the specified process ID. @var{pid} is a
35635 hexadecimal integer identifying the process. This packet is used in
35636 preference to @samp{k} when multiprocess protocol extensions are
35637 supported; see @ref{multiprocess extensions}.
35647 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
35648 @cindex @samp{vRun} packet
35649 Run the program @var{filename}, passing it each @var{argument} on its
35650 command line. The file and arguments are hex-encoded strings. If
35651 @var{filename} is an empty string, the stub may use a default program
35652 (e.g.@: the last program run). The program is created in the stopped
35655 @c FIXME: What about non-stop mode?
35657 This packet is only available in extended mode (@pxref{extended mode}).
35663 @item @r{Any stop packet}
35664 for success (@pxref{Stop Reply Packets})
35668 @anchor{vStopped packet}
35669 @cindex @samp{vStopped} packet
35671 In non-stop mode (@pxref{Remote Non-Stop}), acknowledge a previous stop
35672 reply and prompt for the stub to report another one.
35676 @item @r{Any stop packet}
35677 if there is another unreported stop event (@pxref{Stop Reply Packets})
35679 if there are no unreported stop events
35682 @item X @var{addr},@var{length}:@var{XX@dots{}}
35684 @cindex @samp{X} packet
35685 Write data to memory, where the data is transmitted in binary.
35686 @var{addr} is address, @var{length} is number of bytes,
35687 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
35697 @item z @var{type},@var{addr},@var{kind}
35698 @itemx Z @var{type},@var{addr},@var{kind}
35699 @anchor{insert breakpoint or watchpoint packet}
35700 @cindex @samp{z} packet
35701 @cindex @samp{Z} packets
35702 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
35703 watchpoint starting at address @var{address} of kind @var{kind}.
35705 Each breakpoint and watchpoint packet @var{type} is documented
35708 @emph{Implementation notes: A remote target shall return an empty string
35709 for an unrecognized breakpoint or watchpoint packet @var{type}. A
35710 remote target shall support either both or neither of a given
35711 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
35712 avoid potential problems with duplicate packets, the operations should
35713 be implemented in an idempotent way.}
35715 @item z0,@var{addr},@var{kind}
35716 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
35717 @cindex @samp{z0} packet
35718 @cindex @samp{Z0} packet
35719 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
35720 @var{addr} of type @var{kind}.
35722 A memory breakpoint is implemented by replacing the instruction at
35723 @var{addr} with a software breakpoint or trap instruction. The
35724 @var{kind} is target-specific and typically indicates the size of
35725 the breakpoint in bytes that should be inserted. E.g., the @sc{arm}
35726 and @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
35727 architectures have additional meanings for @var{kind};
35728 @var{cond_list} is an optional list of conditional expressions in bytecode
35729 form that should be evaluated on the target's side. These are the
35730 conditions that should be taken into consideration when deciding if
35731 the breakpoint trigger should be reported back to @var{GDBN}.
35733 The @var{cond_list} parameter is comprised of a series of expressions,
35734 concatenated without separators. Each expression has the following form:
35738 @item X @var{len},@var{expr}
35739 @var{len} is the length of the bytecode expression and @var{expr} is the
35740 actual conditional expression in bytecode form.
35744 The optional @var{cmd_list} parameter introduces commands that may be
35745 run on the target, rather than being reported back to @value{GDBN}.
35746 The parameter starts with a numeric flag @var{persist}; if the flag is
35747 nonzero, then the breakpoint may remain active and the commands
35748 continue to be run even when @value{GDBN} disconnects from the target.
35749 Following this flag is a series of expressions concatenated with no
35750 separators. Each expression has the following form:
35754 @item X @var{len},@var{expr}
35755 @var{len} is the length of the bytecode expression and @var{expr} is the
35756 actual conditional expression in bytecode form.
35760 see @ref{Architecture-Specific Protocol Details}.
35762 @emph{Implementation note: It is possible for a target to copy or move
35763 code that contains memory breakpoints (e.g., when implementing
35764 overlays). The behavior of this packet, in the presence of such a
35765 target, is not defined.}
35777 @item z1,@var{addr},@var{kind}
35778 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}
35779 @cindex @samp{z1} packet
35780 @cindex @samp{Z1} packet
35781 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
35782 address @var{addr}.
35784 A hardware breakpoint is implemented using a mechanism that is not
35785 dependant on being able to modify the target's memory. @var{kind}
35786 and @var{cond_list} have the same meaning as in @samp{Z0} packets.
35788 @emph{Implementation note: A hardware breakpoint is not affected by code
35801 @item z2,@var{addr},@var{kind}
35802 @itemx Z2,@var{addr},@var{kind}
35803 @cindex @samp{z2} packet
35804 @cindex @samp{Z2} packet
35805 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
35806 @var{kind} is interpreted as the number of bytes to watch.
35818 @item z3,@var{addr},@var{kind}
35819 @itemx Z3,@var{addr},@var{kind}
35820 @cindex @samp{z3} packet
35821 @cindex @samp{Z3} packet
35822 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
35823 @var{kind} is interpreted as the number of bytes to watch.
35835 @item z4,@var{addr},@var{kind}
35836 @itemx Z4,@var{addr},@var{kind}
35837 @cindex @samp{z4} packet
35838 @cindex @samp{Z4} packet
35839 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
35840 @var{kind} is interpreted as the number of bytes to watch.
35854 @node Stop Reply Packets
35855 @section Stop Reply Packets
35856 @cindex stop reply packets
35858 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
35859 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
35860 receive any of the below as a reply. Except for @samp{?}
35861 and @samp{vStopped}, that reply is only returned
35862 when the target halts. In the below the exact meaning of @dfn{signal
35863 number} is defined by the header @file{include/gdb/signals.h} in the
35864 @value{GDBN} source code.
35866 As in the description of request packets, we include spaces in the
35867 reply templates for clarity; these are not part of the reply packet's
35868 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
35874 The program received signal number @var{AA} (a two-digit hexadecimal
35875 number). This is equivalent to a @samp{T} response with no
35876 @var{n}:@var{r} pairs.
35878 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
35879 @cindex @samp{T} packet reply
35880 The program received signal number @var{AA} (a two-digit hexadecimal
35881 number). This is equivalent to an @samp{S} response, except that the
35882 @samp{@var{n}:@var{r}} pairs can carry values of important registers
35883 and other information directly in the stop reply packet, reducing
35884 round-trip latency. Single-step and breakpoint traps are reported
35885 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
35889 If @var{n} is a hexadecimal number, it is a register number, and the
35890 corresponding @var{r} gives that register's value. @var{r} is a
35891 series of bytes in target byte order, with each byte given by a
35892 two-digit hex number.
35895 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
35896 the stopped thread, as specified in @ref{thread-id syntax}.
35899 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
35900 the core on which the stop event was detected.
35903 If @var{n} is a recognized @dfn{stop reason}, it describes a more
35904 specific event that stopped the target. The currently defined stop
35905 reasons are listed below. @var{aa} should be @samp{05}, the trap
35906 signal. At most one stop reason should be present.
35909 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
35910 and go on to the next; this allows us to extend the protocol in the
35914 The currently defined stop reasons are:
35920 The packet indicates a watchpoint hit, and @var{r} is the data address, in
35923 @cindex shared library events, remote reply
35925 The packet indicates that the loaded libraries have changed.
35926 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
35927 list of loaded libraries. @var{r} is ignored.
35929 @cindex replay log events, remote reply
35931 The packet indicates that the target cannot continue replaying
35932 logged execution events, because it has reached the end (or the
35933 beginning when executing backward) of the log. The value of @var{r}
35934 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
35935 for more information.
35939 @itemx W @var{AA} ; process:@var{pid}
35940 The process exited, and @var{AA} is the exit status. This is only
35941 applicable to certain targets.
35943 The second form of the response, including the process ID of the exited
35944 process, can be used only when @value{GDBN} has reported support for
35945 multiprocess protocol extensions; see @ref{multiprocess extensions}.
35946 The @var{pid} is formatted as a big-endian hex string.
35949 @itemx X @var{AA} ; process:@var{pid}
35950 The process terminated with signal @var{AA}.
35952 The second form of the response, including the process ID of the
35953 terminated process, can be used only when @value{GDBN} has reported
35954 support for multiprocess protocol extensions; see @ref{multiprocess
35955 extensions}. The @var{pid} is formatted as a big-endian hex string.
35957 @item O @var{XX}@dots{}
35958 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
35959 written as the program's console output. This can happen at any time
35960 while the program is running and the debugger should continue to wait
35961 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
35963 @item F @var{call-id},@var{parameter}@dots{}
35964 @var{call-id} is the identifier which says which host system call should
35965 be called. This is just the name of the function. Translation into the
35966 correct system call is only applicable as it's defined in @value{GDBN}.
35967 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
35970 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
35971 this very system call.
35973 The target replies with this packet when it expects @value{GDBN} to
35974 call a host system call on behalf of the target. @value{GDBN} replies
35975 with an appropriate @samp{F} packet and keeps up waiting for the next
35976 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
35977 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
35978 Protocol Extension}, for more details.
35982 @node General Query Packets
35983 @section General Query Packets
35984 @cindex remote query requests
35986 Packets starting with @samp{q} are @dfn{general query packets};
35987 packets starting with @samp{Q} are @dfn{general set packets}. General
35988 query and set packets are a semi-unified form for retrieving and
35989 sending information to and from the stub.
35991 The initial letter of a query or set packet is followed by a name
35992 indicating what sort of thing the packet applies to. For example,
35993 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
35994 definitions with the stub. These packet names follow some
35999 The name must not contain commas, colons or semicolons.
36001 Most @value{GDBN} query and set packets have a leading upper case
36004 The names of custom vendor packets should use a company prefix, in
36005 lower case, followed by a period. For example, packets designed at
36006 the Acme Corporation might begin with @samp{qacme.foo} (for querying
36007 foos) or @samp{Qacme.bar} (for setting bars).
36010 The name of a query or set packet should be separated from any
36011 parameters by a @samp{:}; the parameters themselves should be
36012 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
36013 full packet name, and check for a separator or the end of the packet,
36014 in case two packet names share a common prefix. New packets should not begin
36015 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
36016 packets predate these conventions, and have arguments without any terminator
36017 for the packet name; we suspect they are in widespread use in places that
36018 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
36019 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
36022 Like the descriptions of the other packets, each description here
36023 has a template showing the packet's overall syntax, followed by an
36024 explanation of the packet's meaning. We include spaces in some of the
36025 templates for clarity; these are not part of the packet's syntax. No
36026 @value{GDBN} packet uses spaces to separate its components.
36028 Here are the currently defined query and set packets:
36034 Turn on or off the agent as a helper to perform some debugging operations
36035 delegated from @value{GDBN} (@pxref{Control Agent}).
36037 @item QAllow:@var{op}:@var{val}@dots{}
36038 @cindex @samp{QAllow} packet
36039 Specify which operations @value{GDBN} expects to request of the
36040 target, as a semicolon-separated list of operation name and value
36041 pairs. Possible values for @var{op} include @samp{WriteReg},
36042 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
36043 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
36044 indicating that @value{GDBN} will not request the operation, or 1,
36045 indicating that it may. (The target can then use this to set up its
36046 own internals optimally, for instance if the debugger never expects to
36047 insert breakpoints, it may not need to install its own trap handler.)
36050 @cindex current thread, remote request
36051 @cindex @samp{qC} packet
36052 Return the current thread ID.
36056 @item QC @var{thread-id}
36057 Where @var{thread-id} is a thread ID as documented in
36058 @ref{thread-id syntax}.
36059 @item @r{(anything else)}
36060 Any other reply implies the old thread ID.
36063 @item qCRC:@var{addr},@var{length}
36064 @cindex CRC of memory block, remote request
36065 @cindex @samp{qCRC} packet
36066 Compute the CRC checksum of a block of memory using CRC-32 defined in
36067 IEEE 802.3. The CRC is computed byte at a time, taking the most
36068 significant bit of each byte first. The initial pattern code
36069 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
36071 @emph{Note:} This is the same CRC used in validating separate debug
36072 files (@pxref{Separate Debug Files, , Debugging Information in Separate
36073 Files}). However the algorithm is slightly different. When validating
36074 separate debug files, the CRC is computed taking the @emph{least}
36075 significant bit of each byte first, and the final result is inverted to
36076 detect trailing zeros.
36081 An error (such as memory fault)
36082 @item C @var{crc32}
36083 The specified memory region's checksum is @var{crc32}.
36086 @item QDisableRandomization:@var{value}
36087 @cindex disable address space randomization, remote request
36088 @cindex @samp{QDisableRandomization} packet
36089 Some target operating systems will randomize the virtual address space
36090 of the inferior process as a security feature, but provide a feature
36091 to disable such randomization, e.g.@: to allow for a more deterministic
36092 debugging experience. On such systems, this packet with a @var{value}
36093 of 1 directs the target to disable address space randomization for
36094 processes subsequently started via @samp{vRun} packets, while a packet
36095 with a @var{value} of 0 tells the target to enable address space
36098 This packet is only available in extended mode (@pxref{extended mode}).
36103 The request succeeded.
36106 An error occurred. @var{nn} are hex digits.
36109 An empty reply indicates that @samp{QDisableRandomization} is not supported
36113 This packet is not probed by default; the remote stub must request it,
36114 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36115 This should only be done on targets that actually support disabling
36116 address space randomization.
36119 @itemx qsThreadInfo
36120 @cindex list active threads, remote request
36121 @cindex @samp{qfThreadInfo} packet
36122 @cindex @samp{qsThreadInfo} packet
36123 Obtain a list of all active thread IDs from the target (OS). Since there
36124 may be too many active threads to fit into one reply packet, this query
36125 works iteratively: it may require more than one query/reply sequence to
36126 obtain the entire list of threads. The first query of the sequence will
36127 be the @samp{qfThreadInfo} query; subsequent queries in the
36128 sequence will be the @samp{qsThreadInfo} query.
36130 NOTE: This packet replaces the @samp{qL} query (see below).
36134 @item m @var{thread-id}
36136 @item m @var{thread-id},@var{thread-id}@dots{}
36137 a comma-separated list of thread IDs
36139 (lower case letter @samp{L}) denotes end of list.
36142 In response to each query, the target will reply with a list of one or
36143 more thread IDs, separated by commas.
36144 @value{GDBN} will respond to each reply with a request for more thread
36145 ids (using the @samp{qs} form of the query), until the target responds
36146 with @samp{l} (lower-case ell, for @dfn{last}).
36147 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
36150 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
36151 @cindex get thread-local storage address, remote request
36152 @cindex @samp{qGetTLSAddr} packet
36153 Fetch the address associated with thread local storage specified
36154 by @var{thread-id}, @var{offset}, and @var{lm}.
36156 @var{thread-id} is the thread ID associated with the
36157 thread for which to fetch the TLS address. @xref{thread-id syntax}.
36159 @var{offset} is the (big endian, hex encoded) offset associated with the
36160 thread local variable. (This offset is obtained from the debug
36161 information associated with the variable.)
36163 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
36164 load module associated with the thread local storage. For example,
36165 a @sc{gnu}/Linux system will pass the link map address of the shared
36166 object associated with the thread local storage under consideration.
36167 Other operating environments may choose to represent the load module
36168 differently, so the precise meaning of this parameter will vary.
36172 @item @var{XX}@dots{}
36173 Hex encoded (big endian) bytes representing the address of the thread
36174 local storage requested.
36177 An error occurred. @var{nn} are hex digits.
36180 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
36183 @item qGetTIBAddr:@var{thread-id}
36184 @cindex get thread information block address
36185 @cindex @samp{qGetTIBAddr} packet
36186 Fetch address of the Windows OS specific Thread Information Block.
36188 @var{thread-id} is the thread ID associated with the thread.
36192 @item @var{XX}@dots{}
36193 Hex encoded (big endian) bytes representing the linear address of the
36194 thread information block.
36197 An error occured. This means that either the thread was not found, or the
36198 address could not be retrieved.
36201 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
36204 @item qL @var{startflag} @var{threadcount} @var{nextthread}
36205 Obtain thread information from RTOS. Where: @var{startflag} (one hex
36206 digit) is one to indicate the first query and zero to indicate a
36207 subsequent query; @var{threadcount} (two hex digits) is the maximum
36208 number of threads the response packet can contain; and @var{nextthread}
36209 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
36210 returned in the response as @var{argthread}.
36212 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
36216 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
36217 Where: @var{count} (two hex digits) is the number of threads being
36218 returned; @var{done} (one hex digit) is zero to indicate more threads
36219 and one indicates no further threads; @var{argthreadid} (eight hex
36220 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
36221 is a sequence of thread IDs from the target. @var{threadid} (eight hex
36222 digits). See @code{remote.c:parse_threadlist_response()}.
36226 @cindex section offsets, remote request
36227 @cindex @samp{qOffsets} packet
36228 Get section offsets that the target used when relocating the downloaded
36233 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
36234 Relocate the @code{Text} section by @var{xxx} from its original address.
36235 Relocate the @code{Data} section by @var{yyy} from its original address.
36236 If the object file format provides segment information (e.g.@: @sc{elf}
36237 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
36238 segments by the supplied offsets.
36240 @emph{Note: while a @code{Bss} offset may be included in the response,
36241 @value{GDBN} ignores this and instead applies the @code{Data} offset
36242 to the @code{Bss} section.}
36244 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
36245 Relocate the first segment of the object file, which conventionally
36246 contains program code, to a starting address of @var{xxx}. If
36247 @samp{DataSeg} is specified, relocate the second segment, which
36248 conventionally contains modifiable data, to a starting address of
36249 @var{yyy}. @value{GDBN} will report an error if the object file
36250 does not contain segment information, or does not contain at least
36251 as many segments as mentioned in the reply. Extra segments are
36252 kept at fixed offsets relative to the last relocated segment.
36255 @item qP @var{mode} @var{thread-id}
36256 @cindex thread information, remote request
36257 @cindex @samp{qP} packet
36258 Returns information on @var{thread-id}. Where: @var{mode} is a hex
36259 encoded 32 bit mode; @var{thread-id} is a thread ID
36260 (@pxref{thread-id syntax}).
36262 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
36265 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
36269 @cindex non-stop mode, remote request
36270 @cindex @samp{QNonStop} packet
36272 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
36273 @xref{Remote Non-Stop}, for more information.
36278 The request succeeded.
36281 An error occurred. @var{nn} are hex digits.
36284 An empty reply indicates that @samp{QNonStop} is not supported by
36288 This packet is not probed by default; the remote stub must request it,
36289 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36290 Use of this packet is controlled by the @code{set non-stop} command;
36291 @pxref{Non-Stop Mode}.
36293 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
36294 @cindex pass signals to inferior, remote request
36295 @cindex @samp{QPassSignals} packet
36296 @anchor{QPassSignals}
36297 Each listed @var{signal} should be passed directly to the inferior process.
36298 Signals are numbered identically to continue packets and stop replies
36299 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
36300 strictly greater than the previous item. These signals do not need to stop
36301 the inferior, or be reported to @value{GDBN}. All other signals should be
36302 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
36303 combine; any earlier @samp{QPassSignals} list is completely replaced by the
36304 new list. This packet improves performance when using @samp{handle
36305 @var{signal} nostop noprint pass}.
36310 The request succeeded.
36313 An error occurred. @var{nn} are hex digits.
36316 An empty reply indicates that @samp{QPassSignals} is not supported by
36320 Use of this packet is controlled by the @code{set remote pass-signals}
36321 command (@pxref{Remote Configuration, set remote pass-signals}).
36322 This packet is not probed by default; the remote stub must request it,
36323 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36325 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
36326 @cindex signals the inferior may see, remote request
36327 @cindex @samp{QProgramSignals} packet
36328 @anchor{QProgramSignals}
36329 Each listed @var{signal} may be delivered to the inferior process.
36330 Others should be silently discarded.
36332 In some cases, the remote stub may need to decide whether to deliver a
36333 signal to the program or not without @value{GDBN} involvement. One
36334 example of that is while detaching --- the program's threads may have
36335 stopped for signals that haven't yet had a chance of being reported to
36336 @value{GDBN}, and so the remote stub can use the signal list specified
36337 by this packet to know whether to deliver or ignore those pending
36340 This does not influence whether to deliver a signal as requested by a
36341 resumption packet (@pxref{vCont packet}).
36343 Signals are numbered identically to continue packets and stop replies
36344 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
36345 strictly greater than the previous item. Multiple
36346 @samp{QProgramSignals} packets do not combine; any earlier
36347 @samp{QProgramSignals} list is completely replaced by the new list.
36352 The request succeeded.
36355 An error occurred. @var{nn} are hex digits.
36358 An empty reply indicates that @samp{QProgramSignals} is not supported
36362 Use of this packet is controlled by the @code{set remote program-signals}
36363 command (@pxref{Remote Configuration, set remote program-signals}).
36364 This packet is not probed by default; the remote stub must request it,
36365 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36367 @item qRcmd,@var{command}
36368 @cindex execute remote command, remote request
36369 @cindex @samp{qRcmd} packet
36370 @var{command} (hex encoded) is passed to the local interpreter for
36371 execution. Invalid commands should be reported using the output
36372 string. Before the final result packet, the target may also respond
36373 with a number of intermediate @samp{O@var{output}} console output
36374 packets. @emph{Implementors should note that providing access to a
36375 stubs's interpreter may have security implications}.
36380 A command response with no output.
36382 A command response with the hex encoded output string @var{OUTPUT}.
36384 Indicate a badly formed request.
36386 An empty reply indicates that @samp{qRcmd} is not recognized.
36389 (Note that the @code{qRcmd} packet's name is separated from the
36390 command by a @samp{,}, not a @samp{:}, contrary to the naming
36391 conventions above. Please don't use this packet as a model for new
36394 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
36395 @cindex searching memory, in remote debugging
36396 @cindex @samp{qSearch:memory} packet
36397 @anchor{qSearch memory}
36398 Search @var{length} bytes at @var{address} for @var{search-pattern}.
36399 @var{address} and @var{length} are encoded in hex.
36400 @var{search-pattern} is a sequence of bytes, hex encoded.
36405 The pattern was not found.
36407 The pattern was found at @var{address}.
36409 A badly formed request or an error was encountered while searching memory.
36411 An empty reply indicates that @samp{qSearch:memory} is not recognized.
36414 @item QStartNoAckMode
36415 @cindex @samp{QStartNoAckMode} packet
36416 @anchor{QStartNoAckMode}
36417 Request that the remote stub disable the normal @samp{+}/@samp{-}
36418 protocol acknowledgments (@pxref{Packet Acknowledgment}).
36423 The stub has switched to no-acknowledgment mode.
36424 @value{GDBN} acknowledges this reponse,
36425 but neither the stub nor @value{GDBN} shall send or expect further
36426 @samp{+}/@samp{-} acknowledgments in the current connection.
36428 An empty reply indicates that the stub does not support no-acknowledgment mode.
36431 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
36432 @cindex supported packets, remote query
36433 @cindex features of the remote protocol
36434 @cindex @samp{qSupported} packet
36435 @anchor{qSupported}
36436 Tell the remote stub about features supported by @value{GDBN}, and
36437 query the stub for features it supports. This packet allows
36438 @value{GDBN} and the remote stub to take advantage of each others'
36439 features. @samp{qSupported} also consolidates multiple feature probes
36440 at startup, to improve @value{GDBN} performance---a single larger
36441 packet performs better than multiple smaller probe packets on
36442 high-latency links. Some features may enable behavior which must not
36443 be on by default, e.g.@: because it would confuse older clients or
36444 stubs. Other features may describe packets which could be
36445 automatically probed for, but are not. These features must be
36446 reported before @value{GDBN} will use them. This ``default
36447 unsupported'' behavior is not appropriate for all packets, but it
36448 helps to keep the initial connection time under control with new
36449 versions of @value{GDBN} which support increasing numbers of packets.
36453 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
36454 The stub supports or does not support each returned @var{stubfeature},
36455 depending on the form of each @var{stubfeature} (see below for the
36458 An empty reply indicates that @samp{qSupported} is not recognized,
36459 or that no features needed to be reported to @value{GDBN}.
36462 The allowed forms for each feature (either a @var{gdbfeature} in the
36463 @samp{qSupported} packet, or a @var{stubfeature} in the response)
36467 @item @var{name}=@var{value}
36468 The remote protocol feature @var{name} is supported, and associated
36469 with the specified @var{value}. The format of @var{value} depends
36470 on the feature, but it must not include a semicolon.
36472 The remote protocol feature @var{name} is supported, and does not
36473 need an associated value.
36475 The remote protocol feature @var{name} is not supported.
36477 The remote protocol feature @var{name} may be supported, and
36478 @value{GDBN} should auto-detect support in some other way when it is
36479 needed. This form will not be used for @var{gdbfeature} notifications,
36480 but may be used for @var{stubfeature} responses.
36483 Whenever the stub receives a @samp{qSupported} request, the
36484 supplied set of @value{GDBN} features should override any previous
36485 request. This allows @value{GDBN} to put the stub in a known
36486 state, even if the stub had previously been communicating with
36487 a different version of @value{GDBN}.
36489 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
36494 This feature indicates whether @value{GDBN} supports multiprocess
36495 extensions to the remote protocol. @value{GDBN} does not use such
36496 extensions unless the stub also reports that it supports them by
36497 including @samp{multiprocess+} in its @samp{qSupported} reply.
36498 @xref{multiprocess extensions}, for details.
36501 This feature indicates that @value{GDBN} supports the XML target
36502 description. If the stub sees @samp{xmlRegisters=} with target
36503 specific strings separated by a comma, it will report register
36507 This feature indicates whether @value{GDBN} supports the
36508 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
36509 instruction reply packet}).
36512 Stubs should ignore any unknown values for
36513 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
36514 packet supports receiving packets of unlimited length (earlier
36515 versions of @value{GDBN} may reject overly long responses). Additional values
36516 for @var{gdbfeature} may be defined in the future to let the stub take
36517 advantage of new features in @value{GDBN}, e.g.@: incompatible
36518 improvements in the remote protocol---the @samp{multiprocess} feature is
36519 an example of such a feature. The stub's reply should be independent
36520 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
36521 describes all the features it supports, and then the stub replies with
36522 all the features it supports.
36524 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
36525 responses, as long as each response uses one of the standard forms.
36527 Some features are flags. A stub which supports a flag feature
36528 should respond with a @samp{+} form response. Other features
36529 require values, and the stub should respond with an @samp{=}
36532 Each feature has a default value, which @value{GDBN} will use if
36533 @samp{qSupported} is not available or if the feature is not mentioned
36534 in the @samp{qSupported} response. The default values are fixed; a
36535 stub is free to omit any feature responses that match the defaults.
36537 Not all features can be probed, but for those which can, the probing
36538 mechanism is useful: in some cases, a stub's internal
36539 architecture may not allow the protocol layer to know some information
36540 about the underlying target in advance. This is especially common in
36541 stubs which may be configured for multiple targets.
36543 These are the currently defined stub features and their properties:
36545 @multitable @columnfractions 0.35 0.2 0.12 0.2
36546 @c NOTE: The first row should be @headitem, but we do not yet require
36547 @c a new enough version of Texinfo (4.7) to use @headitem.
36549 @tab Value Required
36553 @item @samp{PacketSize}
36558 @item @samp{qXfer:auxv:read}
36563 @item @samp{qXfer:features:read}
36568 @item @samp{qXfer:libraries:read}
36573 @item @samp{qXfer:memory-map:read}
36578 @item @samp{qXfer:sdata:read}
36583 @item @samp{qXfer:spu:read}
36588 @item @samp{qXfer:spu:write}
36593 @item @samp{qXfer:siginfo:read}
36598 @item @samp{qXfer:siginfo:write}
36603 @item @samp{qXfer:threads:read}
36608 @item @samp{qXfer:traceframe-info:read}
36613 @item @samp{qXfer:uib:read}
36618 @item @samp{qXfer:fdpic:read}
36623 @item @samp{QNonStop}
36628 @item @samp{QPassSignals}
36633 @item @samp{QStartNoAckMode}
36638 @item @samp{multiprocess}
36643 @item @samp{ConditionalBreakpoints}
36648 @item @samp{ConditionalTracepoints}
36653 @item @samp{ReverseContinue}
36658 @item @samp{ReverseStep}
36663 @item @samp{TracepointSource}
36668 @item @samp{QAgent}
36673 @item @samp{QAllow}
36678 @item @samp{QDisableRandomization}
36683 @item @samp{EnableDisableTracepoints}
36688 @item @samp{tracenz}
36693 @item @samp{BreakpointCommands}
36700 These are the currently defined stub features, in more detail:
36703 @cindex packet size, remote protocol
36704 @item PacketSize=@var{bytes}
36705 The remote stub can accept packets up to at least @var{bytes} in
36706 length. @value{GDBN} will send packets up to this size for bulk
36707 transfers, and will never send larger packets. This is a limit on the
36708 data characters in the packet, including the frame and checksum.
36709 There is no trailing NUL byte in a remote protocol packet; if the stub
36710 stores packets in a NUL-terminated format, it should allow an extra
36711 byte in its buffer for the NUL. If this stub feature is not supported,
36712 @value{GDBN} guesses based on the size of the @samp{g} packet response.
36714 @item qXfer:auxv:read
36715 The remote stub understands the @samp{qXfer:auxv:read} packet
36716 (@pxref{qXfer auxiliary vector read}).
36718 @item qXfer:features:read
36719 The remote stub understands the @samp{qXfer:features:read} packet
36720 (@pxref{qXfer target description read}).
36722 @item qXfer:libraries:read
36723 The remote stub understands the @samp{qXfer:libraries:read} packet
36724 (@pxref{qXfer library list read}).
36726 @item qXfer:libraries-svr4:read
36727 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
36728 (@pxref{qXfer svr4 library list read}).
36730 @item qXfer:memory-map:read
36731 The remote stub understands the @samp{qXfer:memory-map:read} packet
36732 (@pxref{qXfer memory map read}).
36734 @item qXfer:sdata:read
36735 The remote stub understands the @samp{qXfer:sdata:read} packet
36736 (@pxref{qXfer sdata read}).
36738 @item qXfer:spu:read
36739 The remote stub understands the @samp{qXfer:spu:read} packet
36740 (@pxref{qXfer spu read}).
36742 @item qXfer:spu:write
36743 The remote stub understands the @samp{qXfer:spu:write} packet
36744 (@pxref{qXfer spu write}).
36746 @item qXfer:siginfo:read
36747 The remote stub understands the @samp{qXfer:siginfo:read} packet
36748 (@pxref{qXfer siginfo read}).
36750 @item qXfer:siginfo:write
36751 The remote stub understands the @samp{qXfer:siginfo:write} packet
36752 (@pxref{qXfer siginfo write}).
36754 @item qXfer:threads:read
36755 The remote stub understands the @samp{qXfer:threads:read} packet
36756 (@pxref{qXfer threads read}).
36758 @item qXfer:traceframe-info:read
36759 The remote stub understands the @samp{qXfer:traceframe-info:read}
36760 packet (@pxref{qXfer traceframe info read}).
36762 @item qXfer:uib:read
36763 The remote stub understands the @samp{qXfer:uib:read}
36764 packet (@pxref{qXfer unwind info block}).
36766 @item qXfer:fdpic:read
36767 The remote stub understands the @samp{qXfer:fdpic:read}
36768 packet (@pxref{qXfer fdpic loadmap read}).
36771 The remote stub understands the @samp{QNonStop} packet
36772 (@pxref{QNonStop}).
36775 The remote stub understands the @samp{QPassSignals} packet
36776 (@pxref{QPassSignals}).
36778 @item QStartNoAckMode
36779 The remote stub understands the @samp{QStartNoAckMode} packet and
36780 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
36783 @anchor{multiprocess extensions}
36784 @cindex multiprocess extensions, in remote protocol
36785 The remote stub understands the multiprocess extensions to the remote
36786 protocol syntax. The multiprocess extensions affect the syntax of
36787 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
36788 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
36789 replies. Note that reporting this feature indicates support for the
36790 syntactic extensions only, not that the stub necessarily supports
36791 debugging of more than one process at a time. The stub must not use
36792 multiprocess extensions in packet replies unless @value{GDBN} has also
36793 indicated it supports them in its @samp{qSupported} request.
36795 @item qXfer:osdata:read
36796 The remote stub understands the @samp{qXfer:osdata:read} packet
36797 ((@pxref{qXfer osdata read}).
36799 @item ConditionalBreakpoints
36800 The target accepts and implements evaluation of conditional expressions
36801 defined for breakpoints. The target will only report breakpoint triggers
36802 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
36804 @item ConditionalTracepoints
36805 The remote stub accepts and implements conditional expressions defined
36806 for tracepoints (@pxref{Tracepoint Conditions}).
36808 @item ReverseContinue
36809 The remote stub accepts and implements the reverse continue packet
36813 The remote stub accepts and implements the reverse step packet
36816 @item TracepointSource
36817 The remote stub understands the @samp{QTDPsrc} packet that supplies
36818 the source form of tracepoint definitions.
36821 The remote stub understands the @samp{QAgent} packet.
36824 The remote stub understands the @samp{QAllow} packet.
36826 @item QDisableRandomization
36827 The remote stub understands the @samp{QDisableRandomization} packet.
36829 @item StaticTracepoint
36830 @cindex static tracepoints, in remote protocol
36831 The remote stub supports static tracepoints.
36833 @item InstallInTrace
36834 @anchor{install tracepoint in tracing}
36835 The remote stub supports installing tracepoint in tracing.
36837 @item EnableDisableTracepoints
36838 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
36839 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
36840 to be enabled and disabled while a trace experiment is running.
36843 @cindex string tracing, in remote protocol
36844 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
36845 See @ref{Bytecode Descriptions} for details about the bytecode.
36847 @item BreakpointCommands
36848 @cindex breakpoint commands, in remote protocol
36849 The remote stub supports running a breakpoint's command list itself,
36850 rather than reporting the hit to @value{GDBN}.
36855 @cindex symbol lookup, remote request
36856 @cindex @samp{qSymbol} packet
36857 Notify the target that @value{GDBN} is prepared to serve symbol lookup
36858 requests. Accept requests from the target for the values of symbols.
36863 The target does not need to look up any (more) symbols.
36864 @item qSymbol:@var{sym_name}
36865 The target requests the value of symbol @var{sym_name} (hex encoded).
36866 @value{GDBN} may provide the value by using the
36867 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
36871 @item qSymbol:@var{sym_value}:@var{sym_name}
36872 Set the value of @var{sym_name} to @var{sym_value}.
36874 @var{sym_name} (hex encoded) is the name of a symbol whose value the
36875 target has previously requested.
36877 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
36878 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
36884 The target does not need to look up any (more) symbols.
36885 @item qSymbol:@var{sym_name}
36886 The target requests the value of a new symbol @var{sym_name} (hex
36887 encoded). @value{GDBN} will continue to supply the values of symbols
36888 (if available), until the target ceases to request them.
36893 @item QTDisconnected
36900 @itemx qTMinFTPILen
36902 @xref{Tracepoint Packets}.
36904 @item qThreadExtraInfo,@var{thread-id}
36905 @cindex thread attributes info, remote request
36906 @cindex @samp{qThreadExtraInfo} packet
36907 Obtain a printable string description of a thread's attributes from
36908 the target OS. @var{thread-id} is a thread ID;
36909 see @ref{thread-id syntax}. This
36910 string may contain anything that the target OS thinks is interesting
36911 for @value{GDBN} to tell the user about the thread. The string is
36912 displayed in @value{GDBN}'s @code{info threads} display. Some
36913 examples of possible thread extra info strings are @samp{Runnable}, or
36914 @samp{Blocked on Mutex}.
36918 @item @var{XX}@dots{}
36919 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
36920 comprising the printable string containing the extra information about
36921 the thread's attributes.
36924 (Note that the @code{qThreadExtraInfo} packet's name is separated from
36925 the command by a @samp{,}, not a @samp{:}, contrary to the naming
36926 conventions above. Please don't use this packet as a model for new
36945 @xref{Tracepoint Packets}.
36947 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
36948 @cindex read special object, remote request
36949 @cindex @samp{qXfer} packet
36950 @anchor{qXfer read}
36951 Read uninterpreted bytes from the target's special data area
36952 identified by the keyword @var{object}. Request @var{length} bytes
36953 starting at @var{offset} bytes into the data. The content and
36954 encoding of @var{annex} is specific to @var{object}; it can supply
36955 additional details about what data to access.
36957 Here are the specific requests of this form defined so far. All
36958 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
36959 formats, listed below.
36962 @item qXfer:auxv:read::@var{offset},@var{length}
36963 @anchor{qXfer auxiliary vector read}
36964 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
36965 auxiliary vector}. Note @var{annex} must be empty.
36967 This packet is not probed by default; the remote stub must request it,
36968 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36970 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
36971 @anchor{qXfer target description read}
36972 Access the @dfn{target description}. @xref{Target Descriptions}. The
36973 annex specifies which XML document to access. The main description is
36974 always loaded from the @samp{target.xml} annex.
36976 This packet is not probed by default; the remote stub must request it,
36977 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36979 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
36980 @anchor{qXfer library list read}
36981 Access the target's list of loaded libraries. @xref{Library List Format}.
36982 The annex part of the generic @samp{qXfer} packet must be empty
36983 (@pxref{qXfer read}).
36985 Targets which maintain a list of libraries in the program's memory do
36986 not need to implement this packet; it is designed for platforms where
36987 the operating system manages the list of loaded libraries.
36989 This packet is not probed by default; the remote stub must request it,
36990 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36992 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
36993 @anchor{qXfer svr4 library list read}
36994 Access the target's list of loaded libraries when the target is an SVR4
36995 platform. @xref{Library List Format for SVR4 Targets}. The annex part
36996 of the generic @samp{qXfer} packet must be empty (@pxref{qXfer read}).
36998 This packet is optional for better performance on SVR4 targets.
36999 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
37001 This packet is not probed by default; the remote stub must request it,
37002 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37004 @item qXfer:memory-map:read::@var{offset},@var{length}
37005 @anchor{qXfer memory map read}
37006 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
37007 annex part of the generic @samp{qXfer} packet must be empty
37008 (@pxref{qXfer read}).
37010 This packet is not probed by default; the remote stub must request it,
37011 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37013 @item qXfer:sdata:read::@var{offset},@var{length}
37014 @anchor{qXfer sdata read}
37016 Read contents of the extra collected static tracepoint marker
37017 information. The annex part of the generic @samp{qXfer} packet must
37018 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
37021 This packet is not probed by default; the remote stub must request it,
37022 by supplying an appropriate @samp{qSupported} response
37023 (@pxref{qSupported}).
37025 @item qXfer:siginfo:read::@var{offset},@var{length}
37026 @anchor{qXfer siginfo read}
37027 Read contents of the extra signal information on the target
37028 system. The annex part of the generic @samp{qXfer} packet must be
37029 empty (@pxref{qXfer read}).
37031 This packet is not probed by default; the remote stub must request it,
37032 by supplying an appropriate @samp{qSupported} response
37033 (@pxref{qSupported}).
37035 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
37036 @anchor{qXfer spu read}
37037 Read contents of an @code{spufs} file on the target system. The
37038 annex specifies which file to read; it must be of the form
37039 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
37040 in the target process, and @var{name} identifes the @code{spufs} file
37041 in that context to be accessed.
37043 This packet is not probed by default; the remote stub must request it,
37044 by supplying an appropriate @samp{qSupported} response
37045 (@pxref{qSupported}).
37047 @item qXfer:threads:read::@var{offset},@var{length}
37048 @anchor{qXfer threads read}
37049 Access the list of threads on target. @xref{Thread List Format}. The
37050 annex part of the generic @samp{qXfer} packet must be empty
37051 (@pxref{qXfer read}).
37053 This packet is not probed by default; the remote stub must request it,
37054 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37056 @item qXfer:traceframe-info:read::@var{offset},@var{length}
37057 @anchor{qXfer traceframe info read}
37059 Return a description of the current traceframe's contents.
37060 @xref{Traceframe Info Format}. The annex part of the generic
37061 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
37063 This packet is not probed by default; the remote stub must request it,
37064 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37066 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
37067 @anchor{qXfer unwind info block}
37069 Return the unwind information block for @var{pc}. This packet is used
37070 on OpenVMS/ia64 to ask the kernel unwind information.
37072 This packet is not probed by default.
37074 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
37075 @anchor{qXfer fdpic loadmap read}
37076 Read contents of @code{loadmap}s on the target system. The
37077 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
37078 executable @code{loadmap} or interpreter @code{loadmap} to read.
37080 This packet is not probed by default; the remote stub must request it,
37081 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37083 @item qXfer:osdata:read::@var{offset},@var{length}
37084 @anchor{qXfer osdata read}
37085 Access the target's @dfn{operating system information}.
37086 @xref{Operating System Information}.
37093 Data @var{data} (@pxref{Binary Data}) has been read from the
37094 target. There may be more data at a higher address (although
37095 it is permitted to return @samp{m} even for the last valid
37096 block of data, as long as at least one byte of data was read).
37097 @var{data} may have fewer bytes than the @var{length} in the
37101 Data @var{data} (@pxref{Binary Data}) has been read from the target.
37102 There is no more data to be read. @var{data} may have fewer bytes
37103 than the @var{length} in the request.
37106 The @var{offset} in the request is at the end of the data.
37107 There is no more data to be read.
37110 The request was malformed, or @var{annex} was invalid.
37113 The offset was invalid, or there was an error encountered reading the data.
37114 @var{nn} is a hex-encoded @code{errno} value.
37117 An empty reply indicates the @var{object} string was not recognized by
37118 the stub, or that the object does not support reading.
37121 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
37122 @cindex write data into object, remote request
37123 @anchor{qXfer write}
37124 Write uninterpreted bytes into the target's special data area
37125 identified by the keyword @var{object}, starting at @var{offset} bytes
37126 into the data. @var{data}@dots{} is the binary-encoded data
37127 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
37128 is specific to @var{object}; it can supply additional details about what data
37131 Here are the specific requests of this form defined so far. All
37132 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
37133 formats, listed below.
37136 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
37137 @anchor{qXfer siginfo write}
37138 Write @var{data} to the extra signal information on the target system.
37139 The annex part of the generic @samp{qXfer} packet must be
37140 empty (@pxref{qXfer write}).
37142 This packet is not probed by default; the remote stub must request it,
37143 by supplying an appropriate @samp{qSupported} response
37144 (@pxref{qSupported}).
37146 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
37147 @anchor{qXfer spu write}
37148 Write @var{data} to an @code{spufs} file on the target system. The
37149 annex specifies which file to write; it must be of the form
37150 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
37151 in the target process, and @var{name} identifes the @code{spufs} file
37152 in that context to be accessed.
37154 This packet is not probed by default; the remote stub must request it,
37155 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37161 @var{nn} (hex encoded) is the number of bytes written.
37162 This may be fewer bytes than supplied in the request.
37165 The request was malformed, or @var{annex} was invalid.
37168 The offset was invalid, or there was an error encountered writing the data.
37169 @var{nn} is a hex-encoded @code{errno} value.
37172 An empty reply indicates the @var{object} string was not
37173 recognized by the stub, or that the object does not support writing.
37176 @item qXfer:@var{object}:@var{operation}:@dots{}
37177 Requests of this form may be added in the future. When a stub does
37178 not recognize the @var{object} keyword, or its support for
37179 @var{object} does not recognize the @var{operation} keyword, the stub
37180 must respond with an empty packet.
37182 @item qAttached:@var{pid}
37183 @cindex query attached, remote request
37184 @cindex @samp{qAttached} packet
37185 Return an indication of whether the remote server attached to an
37186 existing process or created a new process. When the multiprocess
37187 protocol extensions are supported (@pxref{multiprocess extensions}),
37188 @var{pid} is an integer in hexadecimal format identifying the target
37189 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
37190 the query packet will be simplified as @samp{qAttached}.
37192 This query is used, for example, to know whether the remote process
37193 should be detached or killed when a @value{GDBN} session is ended with
37194 the @code{quit} command.
37199 The remote server attached to an existing process.
37201 The remote server created a new process.
37203 A badly formed request or an error was encountered.
37208 @node Architecture-Specific Protocol Details
37209 @section Architecture-Specific Protocol Details
37211 This section describes how the remote protocol is applied to specific
37212 target architectures. Also see @ref{Standard Target Features}, for
37213 details of XML target descriptions for each architecture.
37216 * ARM-Specific Protocol Details::
37217 * MIPS-Specific Protocol Details::
37220 @node ARM-Specific Protocol Details
37221 @subsection @acronym{ARM}-specific Protocol Details
37224 * ARM Breakpoint Kinds::
37227 @node ARM Breakpoint Kinds
37228 @subsubsection @acronym{ARM} Breakpoint Kinds
37229 @cindex breakpoint kinds, @acronym{ARM}
37231 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
37236 16-bit Thumb mode breakpoint.
37239 32-bit Thumb mode (Thumb-2) breakpoint.
37242 32-bit @acronym{ARM} mode breakpoint.
37246 @node MIPS-Specific Protocol Details
37247 @subsection @acronym{MIPS}-specific Protocol Details
37250 * MIPS Register packet Format::
37251 * MIPS Breakpoint Kinds::
37254 @node MIPS Register packet Format
37255 @subsubsection @acronym{MIPS} Register Packet Format
37256 @cindex register packet format, @acronym{MIPS}
37258 The following @code{g}/@code{G} packets have previously been defined.
37259 In the below, some thirty-two bit registers are transferred as
37260 sixty-four bits. Those registers should be zero/sign extended (which?)
37261 to fill the space allocated. Register bytes are transferred in target
37262 byte order. The two nibbles within a register byte are transferred
37263 most-significant -- least-significant.
37268 All registers are transferred as thirty-two bit quantities in the order:
37269 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
37270 registers; fsr; fir; fp.
37273 All registers are transferred as sixty-four bit quantities (including
37274 thirty-two bit registers such as @code{sr}). The ordering is the same
37279 @node MIPS Breakpoint Kinds
37280 @subsubsection @acronym{MIPS} Breakpoint Kinds
37281 @cindex breakpoint kinds, @acronym{MIPS}
37283 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
37288 16-bit @acronym{MIPS16} mode breakpoint.
37291 16-bit @acronym{microMIPS} mode breakpoint.
37294 32-bit standard @acronym{MIPS} mode breakpoint.
37297 32-bit @acronym{microMIPS} mode breakpoint.
37301 @node Tracepoint Packets
37302 @section Tracepoint Packets
37303 @cindex tracepoint packets
37304 @cindex packets, tracepoint
37306 Here we describe the packets @value{GDBN} uses to implement
37307 tracepoints (@pxref{Tracepoints}).
37311 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
37312 @cindex @samp{QTDP} packet
37313 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
37314 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
37315 the tracepoint is disabled. @var{step} is the tracepoint's step
37316 count, and @var{pass} is its pass count. If an @samp{F} is present,
37317 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
37318 the number of bytes that the target should copy elsewhere to make room
37319 for the tracepoint. If an @samp{X} is present, it introduces a
37320 tracepoint condition, which consists of a hexadecimal length, followed
37321 by a comma and hex-encoded bytes, in a manner similar to action
37322 encodings as described below. If the trailing @samp{-} is present,
37323 further @samp{QTDP} packets will follow to specify this tracepoint's
37329 The packet was understood and carried out.
37331 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
37333 The packet was not recognized.
37336 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
37337 Define actions to be taken when a tracepoint is hit. @var{n} and
37338 @var{addr} must be the same as in the initial @samp{QTDP} packet for
37339 this tracepoint. This packet may only be sent immediately after
37340 another @samp{QTDP} packet that ended with a @samp{-}. If the
37341 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
37342 specifying more actions for this tracepoint.
37344 In the series of action packets for a given tracepoint, at most one
37345 can have an @samp{S} before its first @var{action}. If such a packet
37346 is sent, it and the following packets define ``while-stepping''
37347 actions. Any prior packets define ordinary actions --- that is, those
37348 taken when the tracepoint is first hit. If no action packet has an
37349 @samp{S}, then all the packets in the series specify ordinary
37350 tracepoint actions.
37352 The @samp{@var{action}@dots{}} portion of the packet is a series of
37353 actions, concatenated without separators. Each action has one of the
37359 Collect the registers whose bits are set in @var{mask}. @var{mask} is
37360 a hexadecimal number whose @var{i}'th bit is set if register number
37361 @var{i} should be collected. (The least significant bit is numbered
37362 zero.) Note that @var{mask} may be any number of digits long; it may
37363 not fit in a 32-bit word.
37365 @item M @var{basereg},@var{offset},@var{len}
37366 Collect @var{len} bytes of memory starting at the address in register
37367 number @var{basereg}, plus @var{offset}. If @var{basereg} is
37368 @samp{-1}, then the range has a fixed address: @var{offset} is the
37369 address of the lowest byte to collect. The @var{basereg},
37370 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
37371 values (the @samp{-1} value for @var{basereg} is a special case).
37373 @item X @var{len},@var{expr}
37374 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
37375 it directs. @var{expr} is an agent expression, as described in
37376 @ref{Agent Expressions}. Each byte of the expression is encoded as a
37377 two-digit hex number in the packet; @var{len} is the number of bytes
37378 in the expression (and thus one-half the number of hex digits in the
37383 Any number of actions may be packed together in a single @samp{QTDP}
37384 packet, as long as the packet does not exceed the maximum packet
37385 length (400 bytes, for many stubs). There may be only one @samp{R}
37386 action per tracepoint, and it must precede any @samp{M} or @samp{X}
37387 actions. Any registers referred to by @samp{M} and @samp{X} actions
37388 must be collected by a preceding @samp{R} action. (The
37389 ``while-stepping'' actions are treated as if they were attached to a
37390 separate tracepoint, as far as these restrictions are concerned.)
37395 The packet was understood and carried out.
37397 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
37399 The packet was not recognized.
37402 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
37403 @cindex @samp{QTDPsrc} packet
37404 Specify a source string of tracepoint @var{n} at address @var{addr}.
37405 This is useful to get accurate reproduction of the tracepoints
37406 originally downloaded at the beginning of the trace run. @var{type}
37407 is the name of the tracepoint part, such as @samp{cond} for the
37408 tracepoint's conditional expression (see below for a list of types), while
37409 @var{bytes} is the string, encoded in hexadecimal.
37411 @var{start} is the offset of the @var{bytes} within the overall source
37412 string, while @var{slen} is the total length of the source string.
37413 This is intended for handling source strings that are longer than will
37414 fit in a single packet.
37415 @c Add detailed example when this info is moved into a dedicated
37416 @c tracepoint descriptions section.
37418 The available string types are @samp{at} for the location,
37419 @samp{cond} for the conditional, and @samp{cmd} for an action command.
37420 @value{GDBN} sends a separate packet for each command in the action
37421 list, in the same order in which the commands are stored in the list.
37423 The target does not need to do anything with source strings except
37424 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
37427 Although this packet is optional, and @value{GDBN} will only send it
37428 if the target replies with @samp{TracepointSource} @xref{General
37429 Query Packets}, it makes both disconnected tracing and trace files
37430 much easier to use. Otherwise the user must be careful that the
37431 tracepoints in effect while looking at trace frames are identical to
37432 the ones in effect during the trace run; even a small discrepancy
37433 could cause @samp{tdump} not to work, or a particular trace frame not
37436 @item QTDV:@var{n}:@var{value}
37437 @cindex define trace state variable, remote request
37438 @cindex @samp{QTDV} packet
37439 Create a new trace state variable, number @var{n}, with an initial
37440 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
37441 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
37442 the option of not using this packet for initial values of zero; the
37443 target should simply create the trace state variables as they are
37444 mentioned in expressions.
37446 @item QTFrame:@var{n}
37447 @cindex @samp{QTFrame} packet
37448 Select the @var{n}'th tracepoint frame from the buffer, and use the
37449 register and memory contents recorded there to answer subsequent
37450 request packets from @value{GDBN}.
37452 A successful reply from the stub indicates that the stub has found the
37453 requested frame. The response is a series of parts, concatenated
37454 without separators, describing the frame we selected. Each part has
37455 one of the following forms:
37459 The selected frame is number @var{n} in the trace frame buffer;
37460 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
37461 was no frame matching the criteria in the request packet.
37464 The selected trace frame records a hit of tracepoint number @var{t};
37465 @var{t} is a hexadecimal number.
37469 @item QTFrame:pc:@var{addr}
37470 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
37471 currently selected frame whose PC is @var{addr};
37472 @var{addr} is a hexadecimal number.
37474 @item QTFrame:tdp:@var{t}
37475 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
37476 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
37477 is a hexadecimal number.
37479 @item QTFrame:range:@var{start}:@var{end}
37480 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
37481 currently selected frame whose PC is between @var{start} (inclusive)
37482 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
37485 @item QTFrame:outside:@var{start}:@var{end}
37486 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
37487 frame @emph{outside} the given range of addresses (exclusive).
37490 @cindex @samp{qTMinFTPILen} packet
37491 This packet requests the minimum length of instruction at which a fast
37492 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
37493 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
37494 it depends on the target system being able to create trampolines in
37495 the first 64K of memory, which might or might not be possible for that
37496 system. So the reply to this packet will be 4 if it is able to
37503 The minimum instruction length is currently unknown.
37505 The minimum instruction length is @var{length}, where @var{length} is greater
37506 or equal to 1. @var{length} is a hexadecimal number. A reply of 1 means
37507 that a fast tracepoint may be placed on any instruction regardless of size.
37509 An error has occurred.
37511 An empty reply indicates that the request is not supported by the stub.
37515 @cindex @samp{QTStart} packet
37516 Begin the tracepoint experiment. Begin collecting data from
37517 tracepoint hits in the trace frame buffer. This packet supports the
37518 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
37519 instruction reply packet}).
37522 @cindex @samp{QTStop} packet
37523 End the tracepoint experiment. Stop collecting trace frames.
37525 @item QTEnable:@var{n}:@var{addr}
37527 @cindex @samp{QTEnable} packet
37528 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
37529 experiment. If the tracepoint was previously disabled, then collection
37530 of data from it will resume.
37532 @item QTDisable:@var{n}:@var{addr}
37534 @cindex @samp{QTDisable} packet
37535 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
37536 experiment. No more data will be collected from the tracepoint unless
37537 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
37540 @cindex @samp{QTinit} packet
37541 Clear the table of tracepoints, and empty the trace frame buffer.
37543 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
37544 @cindex @samp{QTro} packet
37545 Establish the given ranges of memory as ``transparent''. The stub
37546 will answer requests for these ranges from memory's current contents,
37547 if they were not collected as part of the tracepoint hit.
37549 @value{GDBN} uses this to mark read-only regions of memory, like those
37550 containing program code. Since these areas never change, they should
37551 still have the same contents they did when the tracepoint was hit, so
37552 there's no reason for the stub to refuse to provide their contents.
37554 @item QTDisconnected:@var{value}
37555 @cindex @samp{QTDisconnected} packet
37556 Set the choice to what to do with the tracing run when @value{GDBN}
37557 disconnects from the target. A @var{value} of 1 directs the target to
37558 continue the tracing run, while 0 tells the target to stop tracing if
37559 @value{GDBN} is no longer in the picture.
37562 @cindex @samp{qTStatus} packet
37563 Ask the stub if there is a trace experiment running right now.
37565 The reply has the form:
37569 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
37570 @var{running} is a single digit @code{1} if the trace is presently
37571 running, or @code{0} if not. It is followed by semicolon-separated
37572 optional fields that an agent may use to report additional status.
37576 If the trace is not running, the agent may report any of several
37577 explanations as one of the optional fields:
37582 No trace has been run yet.
37584 @item tstop[:@var{text}]:0
37585 The trace was stopped by a user-originated stop command. The optional
37586 @var{text} field is a user-supplied string supplied as part of the
37587 stop command (for instance, an explanation of why the trace was
37588 stopped manually). It is hex-encoded.
37591 The trace stopped because the trace buffer filled up.
37593 @item tdisconnected:0
37594 The trace stopped because @value{GDBN} disconnected from the target.
37596 @item tpasscount:@var{tpnum}
37597 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
37599 @item terror:@var{text}:@var{tpnum}
37600 The trace stopped because tracepoint @var{tpnum} had an error. The
37601 string @var{text} is available to describe the nature of the error
37602 (for instance, a divide by zero in the condition expression).
37603 @var{text} is hex encoded.
37606 The trace stopped for some other reason.
37610 Additional optional fields supply statistical and other information.
37611 Although not required, they are extremely useful for users monitoring
37612 the progress of a trace run. If a trace has stopped, and these
37613 numbers are reported, they must reflect the state of the just-stopped
37618 @item tframes:@var{n}
37619 The number of trace frames in the buffer.
37621 @item tcreated:@var{n}
37622 The total number of trace frames created during the run. This may
37623 be larger than the trace frame count, if the buffer is circular.
37625 @item tsize:@var{n}
37626 The total size of the trace buffer, in bytes.
37628 @item tfree:@var{n}
37629 The number of bytes still unused in the buffer.
37631 @item circular:@var{n}
37632 The value of the circular trace buffer flag. @code{1} means that the
37633 trace buffer is circular and old trace frames will be discarded if
37634 necessary to make room, @code{0} means that the trace buffer is linear
37637 @item disconn:@var{n}
37638 The value of the disconnected tracing flag. @code{1} means that
37639 tracing will continue after @value{GDBN} disconnects, @code{0} means
37640 that the trace run will stop.
37644 @item qTP:@var{tp}:@var{addr}
37645 @cindex tracepoint status, remote request
37646 @cindex @samp{qTP} packet
37647 Ask the stub for the current state of tracepoint number @var{tp} at
37648 address @var{addr}.
37652 @item V@var{hits}:@var{usage}
37653 The tracepoint has been hit @var{hits} times so far during the trace
37654 run, and accounts for @var{usage} in the trace buffer. Note that
37655 @code{while-stepping} steps are not counted as separate hits, but the
37656 steps' space consumption is added into the usage number.
37660 @item qTV:@var{var}
37661 @cindex trace state variable value, remote request
37662 @cindex @samp{qTV} packet
37663 Ask the stub for the value of the trace state variable number @var{var}.
37668 The value of the variable is @var{value}. This will be the current
37669 value of the variable if the user is examining a running target, or a
37670 saved value if the variable was collected in the trace frame that the
37671 user is looking at. Note that multiple requests may result in
37672 different reply values, such as when requesting values while the
37673 program is running.
37676 The value of the variable is unknown. This would occur, for example,
37677 if the user is examining a trace frame in which the requested variable
37682 @cindex @samp{qTfP} packet
37684 @cindex @samp{qTsP} packet
37685 These packets request data about tracepoints that are being used by
37686 the target. @value{GDBN} sends @code{qTfP} to get the first piece
37687 of data, and multiple @code{qTsP} to get additional pieces. Replies
37688 to these packets generally take the form of the @code{QTDP} packets
37689 that define tracepoints. (FIXME add detailed syntax)
37692 @cindex @samp{qTfV} packet
37694 @cindex @samp{qTsV} packet
37695 These packets request data about trace state variables that are on the
37696 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
37697 and multiple @code{qTsV} to get additional variables. Replies to
37698 these packets follow the syntax of the @code{QTDV} packets that define
37699 trace state variables.
37705 @cindex @samp{qTfSTM} packet
37706 @cindex @samp{qTsSTM} packet
37707 These packets request data about static tracepoint markers that exist
37708 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
37709 first piece of data, and multiple @code{qTsSTM} to get additional
37710 pieces. Replies to these packets take the following form:
37714 @item m @var{address}:@var{id}:@var{extra}
37716 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
37717 a comma-separated list of markers
37719 (lower case letter @samp{L}) denotes end of list.
37721 An error occurred. @var{nn} are hex digits.
37723 An empty reply indicates that the request is not supported by the
37727 @var{address} is encoded in hex.
37728 @var{id} and @var{extra} are strings encoded in hex.
37730 In response to each query, the target will reply with a list of one or
37731 more markers, separated by commas. @value{GDBN} will respond to each
37732 reply with a request for more markers (using the @samp{qs} form of the
37733 query), until the target responds with @samp{l} (lower-case ell, for
37736 @item qTSTMat:@var{address}
37738 @cindex @samp{qTSTMat} packet
37739 This packets requests data about static tracepoint markers in the
37740 target program at @var{address}. Replies to this packet follow the
37741 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
37742 tracepoint markers.
37744 @item QTSave:@var{filename}
37745 @cindex @samp{QTSave} packet
37746 This packet directs the target to save trace data to the file name
37747 @var{filename} in the target's filesystem. @var{filename} is encoded
37748 as a hex string; the interpretation of the file name (relative vs
37749 absolute, wild cards, etc) is up to the target.
37751 @item qTBuffer:@var{offset},@var{len}
37752 @cindex @samp{qTBuffer} packet
37753 Return up to @var{len} bytes of the current contents of trace buffer,
37754 starting at @var{offset}. The trace buffer is treated as if it were
37755 a contiguous collection of traceframes, as per the trace file format.
37756 The reply consists as many hex-encoded bytes as the target can deliver
37757 in a packet; it is not an error to return fewer than were asked for.
37758 A reply consisting of just @code{l} indicates that no bytes are
37761 @item QTBuffer:circular:@var{value}
37762 This packet directs the target to use a circular trace buffer if
37763 @var{value} is 1, or a linear buffer if the value is 0.
37765 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
37766 @cindex @samp{QTNotes} packet
37767 This packet adds optional textual notes to the trace run. Allowable
37768 types include @code{user}, @code{notes}, and @code{tstop}, the
37769 @var{text} fields are arbitrary strings, hex-encoded.
37773 @subsection Relocate instruction reply packet
37774 When installing fast tracepoints in memory, the target may need to
37775 relocate the instruction currently at the tracepoint address to a
37776 different address in memory. For most instructions, a simple copy is
37777 enough, but, for example, call instructions that implicitly push the
37778 return address on the stack, and relative branches or other
37779 PC-relative instructions require offset adjustment, so that the effect
37780 of executing the instruction at a different address is the same as if
37781 it had executed in the original location.
37783 In response to several of the tracepoint packets, the target may also
37784 respond with a number of intermediate @samp{qRelocInsn} request
37785 packets before the final result packet, to have @value{GDBN} handle
37786 this relocation operation. If a packet supports this mechanism, its
37787 documentation will explicitly say so. See for example the above
37788 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
37789 format of the request is:
37792 @item qRelocInsn:@var{from};@var{to}
37794 This requests @value{GDBN} to copy instruction at address @var{from}
37795 to address @var{to}, possibly adjusted so that executing the
37796 instruction at @var{to} has the same effect as executing it at
37797 @var{from}. @value{GDBN} writes the adjusted instruction to target
37798 memory starting at @var{to}.
37803 @item qRelocInsn:@var{adjusted_size}
37804 Informs the stub the relocation is complete. @var{adjusted_size} is
37805 the length in bytes of resulting relocated instruction sequence.
37807 A badly formed request was detected, or an error was encountered while
37808 relocating the instruction.
37811 @node Host I/O Packets
37812 @section Host I/O Packets
37813 @cindex Host I/O, remote protocol
37814 @cindex file transfer, remote protocol
37816 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
37817 operations on the far side of a remote link. For example, Host I/O is
37818 used to upload and download files to a remote target with its own
37819 filesystem. Host I/O uses the same constant values and data structure
37820 layout as the target-initiated File-I/O protocol. However, the
37821 Host I/O packets are structured differently. The target-initiated
37822 protocol relies on target memory to store parameters and buffers.
37823 Host I/O requests are initiated by @value{GDBN}, and the
37824 target's memory is not involved. @xref{File-I/O Remote Protocol
37825 Extension}, for more details on the target-initiated protocol.
37827 The Host I/O request packets all encode a single operation along with
37828 its arguments. They have this format:
37832 @item vFile:@var{operation}: @var{parameter}@dots{}
37833 @var{operation} is the name of the particular request; the target
37834 should compare the entire packet name up to the second colon when checking
37835 for a supported operation. The format of @var{parameter} depends on
37836 the operation. Numbers are always passed in hexadecimal. Negative
37837 numbers have an explicit minus sign (i.e.@: two's complement is not
37838 used). Strings (e.g.@: filenames) are encoded as a series of
37839 hexadecimal bytes. The last argument to a system call may be a
37840 buffer of escaped binary data (@pxref{Binary Data}).
37844 The valid responses to Host I/O packets are:
37848 @item F @var{result} [, @var{errno}] [; @var{attachment}]
37849 @var{result} is the integer value returned by this operation, usually
37850 non-negative for success and -1 for errors. If an error has occured,
37851 @var{errno} will be included in the result. @var{errno} will have a
37852 value defined by the File-I/O protocol (@pxref{Errno Values}). For
37853 operations which return data, @var{attachment} supplies the data as a
37854 binary buffer. Binary buffers in response packets are escaped in the
37855 normal way (@pxref{Binary Data}). See the individual packet
37856 documentation for the interpretation of @var{result} and
37860 An empty response indicates that this operation is not recognized.
37864 These are the supported Host I/O operations:
37867 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
37868 Open a file at @var{pathname} and return a file descriptor for it, or
37869 return -1 if an error occurs. @var{pathname} is a string,
37870 @var{flags} is an integer indicating a mask of open flags
37871 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
37872 of mode bits to use if the file is created (@pxref{mode_t Values}).
37873 @xref{open}, for details of the open flags and mode values.
37875 @item vFile:close: @var{fd}
37876 Close the open file corresponding to @var{fd} and return 0, or
37877 -1 if an error occurs.
37879 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
37880 Read data from the open file corresponding to @var{fd}. Up to
37881 @var{count} bytes will be read from the file, starting at @var{offset}
37882 relative to the start of the file. The target may read fewer bytes;
37883 common reasons include packet size limits and an end-of-file
37884 condition. The number of bytes read is returned. Zero should only be
37885 returned for a successful read at the end of the file, or if
37886 @var{count} was zero.
37888 The data read should be returned as a binary attachment on success.
37889 If zero bytes were read, the response should include an empty binary
37890 attachment (i.e.@: a trailing semicolon). The return value is the
37891 number of target bytes read; the binary attachment may be longer if
37892 some characters were escaped.
37894 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
37895 Write @var{data} (a binary buffer) to the open file corresponding
37896 to @var{fd}. Start the write at @var{offset} from the start of the
37897 file. Unlike many @code{write} system calls, there is no
37898 separate @var{count} argument; the length of @var{data} in the
37899 packet is used. @samp{vFile:write} returns the number of bytes written,
37900 which may be shorter than the length of @var{data}, or -1 if an
37903 @item vFile:unlink: @var{pathname}
37904 Delete the file at @var{pathname} on the target. Return 0,
37905 or -1 if an error occurs. @var{pathname} is a string.
37907 @item vFile:readlink: @var{filename}
37908 Read value of symbolic link @var{filename} on the target. Return
37909 the number of bytes read, or -1 if an error occurs.
37911 The data read should be returned as a binary attachment on success.
37912 If zero bytes were read, the response should include an empty binary
37913 attachment (i.e.@: a trailing semicolon). The return value is the
37914 number of target bytes read; the binary attachment may be longer if
37915 some characters were escaped.
37920 @section Interrupts
37921 @cindex interrupts (remote protocol)
37923 When a program on the remote target is running, @value{GDBN} may
37924 attempt to interrupt it by sending a @samp{Ctrl-C}, @code{BREAK} or
37925 a @code{BREAK} followed by @code{g},
37926 control of which is specified via @value{GDBN}'s @samp{interrupt-sequence}.
37928 The precise meaning of @code{BREAK} is defined by the transport
37929 mechanism and may, in fact, be undefined. @value{GDBN} does not
37930 currently define a @code{BREAK} mechanism for any of the network
37931 interfaces except for TCP, in which case @value{GDBN} sends the
37932 @code{telnet} BREAK sequence.
37934 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
37935 transport mechanisms. It is represented by sending the single byte
37936 @code{0x03} without any of the usual packet overhead described in
37937 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
37938 transmitted as part of a packet, it is considered to be packet data
37939 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
37940 (@pxref{X packet}), used for binary downloads, may include an unescaped
37941 @code{0x03} as part of its packet.
37943 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
37944 When Linux kernel receives this sequence from serial port,
37945 it stops execution and connects to gdb.
37947 Stubs are not required to recognize these interrupt mechanisms and the
37948 precise meaning associated with receipt of the interrupt is
37949 implementation defined. If the target supports debugging of multiple
37950 threads and/or processes, it should attempt to interrupt all
37951 currently-executing threads and processes.
37952 If the stub is successful at interrupting the
37953 running program, it should send one of the stop
37954 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
37955 of successfully stopping the program in all-stop mode, and a stop reply
37956 for each stopped thread in non-stop mode.
37957 Interrupts received while the
37958 program is stopped are discarded.
37960 @node Notification Packets
37961 @section Notification Packets
37962 @cindex notification packets
37963 @cindex packets, notification
37965 The @value{GDBN} remote serial protocol includes @dfn{notifications},
37966 packets that require no acknowledgment. Both the GDB and the stub
37967 may send notifications (although the only notifications defined at
37968 present are sent by the stub). Notifications carry information
37969 without incurring the round-trip latency of an acknowledgment, and so
37970 are useful for low-impact communications where occasional packet loss
37973 A notification packet has the form @samp{% @var{data} #
37974 @var{checksum}}, where @var{data} is the content of the notification,
37975 and @var{checksum} is a checksum of @var{data}, computed and formatted
37976 as for ordinary @value{GDBN} packets. A notification's @var{data}
37977 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
37978 receiving a notification, the recipient sends no @samp{+} or @samp{-}
37979 to acknowledge the notification's receipt or to report its corruption.
37981 Every notification's @var{data} begins with a name, which contains no
37982 colon characters, followed by a colon character.
37984 Recipients should silently ignore corrupted notifications and
37985 notifications they do not understand. Recipients should restart
37986 timeout periods on receipt of a well-formed notification, whether or
37987 not they understand it.
37989 Senders should only send the notifications described here when this
37990 protocol description specifies that they are permitted. In the
37991 future, we may extend the protocol to permit existing notifications in
37992 new contexts; this rule helps older senders avoid confusing newer
37995 (Older versions of @value{GDBN} ignore bytes received until they see
37996 the @samp{$} byte that begins an ordinary packet, so new stubs may
37997 transmit notifications without fear of confusing older clients. There
37998 are no notifications defined for @value{GDBN} to send at the moment, but we
37999 assume that most older stubs would ignore them, as well.)
38001 The following notification packets from the stub to @value{GDBN} are
38005 @item Stop: @var{reply}
38006 Report an asynchronous stop event in non-stop mode.
38007 The @var{reply} has the form of a stop reply, as
38008 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
38009 for information on how these notifications are acknowledged by
38013 @node Remote Non-Stop
38014 @section Remote Protocol Support for Non-Stop Mode
38016 @value{GDBN}'s remote protocol supports non-stop debugging of
38017 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
38018 supports non-stop mode, it should report that to @value{GDBN} by including
38019 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
38021 @value{GDBN} typically sends a @samp{QNonStop} packet only when
38022 establishing a new connection with the stub. Entering non-stop mode
38023 does not alter the state of any currently-running threads, but targets
38024 must stop all threads in any already-attached processes when entering
38025 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
38026 probe the target state after a mode change.
38028 In non-stop mode, when an attached process encounters an event that
38029 would otherwise be reported with a stop reply, it uses the
38030 asynchronous notification mechanism (@pxref{Notification Packets}) to
38031 inform @value{GDBN}. In contrast to all-stop mode, where all threads
38032 in all processes are stopped when a stop reply is sent, in non-stop
38033 mode only the thread reporting the stop event is stopped. That is,
38034 when reporting a @samp{S} or @samp{T} response to indicate completion
38035 of a step operation, hitting a breakpoint, or a fault, only the
38036 affected thread is stopped; any other still-running threads continue
38037 to run. When reporting a @samp{W} or @samp{X} response, all running
38038 threads belonging to other attached processes continue to run.
38040 Only one stop reply notification at a time may be pending; if
38041 additional stop events occur before @value{GDBN} has acknowledged the
38042 previous notification, they must be queued by the stub for later
38043 synchronous transmission in response to @samp{vStopped} packets from
38044 @value{GDBN}. Because the notification mechanism is unreliable,
38045 the stub is permitted to resend a stop reply notification
38046 if it believes @value{GDBN} may not have received it. @value{GDBN}
38047 ignores additional stop reply notifications received before it has
38048 finished processing a previous notification and the stub has completed
38049 sending any queued stop events.
38051 Otherwise, @value{GDBN} must be prepared to receive a stop reply
38052 notification at any time. Specifically, they may appear when
38053 @value{GDBN} is not otherwise reading input from the stub, or when
38054 @value{GDBN} is expecting to read a normal synchronous response or a
38055 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
38056 Notification packets are distinct from any other communication from
38057 the stub so there is no ambiguity.
38059 After receiving a stop reply notification, @value{GDBN} shall
38060 acknowledge it by sending a @samp{vStopped} packet (@pxref{vStopped packet})
38061 as a regular, synchronous request to the stub. Such acknowledgment
38062 is not required to happen immediately, as @value{GDBN} is permitted to
38063 send other, unrelated packets to the stub first, which the stub should
38066 Upon receiving a @samp{vStopped} packet, if the stub has other queued
38067 stop events to report to @value{GDBN}, it shall respond by sending a
38068 normal stop reply response. @value{GDBN} shall then send another
38069 @samp{vStopped} packet to solicit further responses; again, it is
38070 permitted to send other, unrelated packets as well which the stub
38071 should process normally.
38073 If the stub receives a @samp{vStopped} packet and there are no
38074 additional stop events to report, the stub shall return an @samp{OK}
38075 response. At this point, if further stop events occur, the stub shall
38076 send a new stop reply notification, @value{GDBN} shall accept the
38077 notification, and the process shall be repeated.
38079 In non-stop mode, the target shall respond to the @samp{?} packet as
38080 follows. First, any incomplete stop reply notification/@samp{vStopped}
38081 sequence in progress is abandoned. The target must begin a new
38082 sequence reporting stop events for all stopped threads, whether or not
38083 it has previously reported those events to @value{GDBN}. The first
38084 stop reply is sent as a synchronous reply to the @samp{?} packet, and
38085 subsequent stop replies are sent as responses to @samp{vStopped} packets
38086 using the mechanism described above. The target must not send
38087 asynchronous stop reply notifications until the sequence is complete.
38088 If all threads are running when the target receives the @samp{?} packet,
38089 or if the target is not attached to any process, it shall respond
38092 @node Packet Acknowledgment
38093 @section Packet Acknowledgment
38095 @cindex acknowledgment, for @value{GDBN} remote
38096 @cindex packet acknowledgment, for @value{GDBN} remote
38097 By default, when either the host or the target machine receives a packet,
38098 the first response expected is an acknowledgment: either @samp{+} (to indicate
38099 the package was received correctly) or @samp{-} (to request retransmission).
38100 This mechanism allows the @value{GDBN} remote protocol to operate over
38101 unreliable transport mechanisms, such as a serial line.
38103 In cases where the transport mechanism is itself reliable (such as a pipe or
38104 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
38105 It may be desirable to disable them in that case to reduce communication
38106 overhead, or for other reasons. This can be accomplished by means of the
38107 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
38109 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
38110 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
38111 and response format still includes the normal checksum, as described in
38112 @ref{Overview}, but the checksum may be ignored by the receiver.
38114 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
38115 no-acknowledgment mode, it should report that to @value{GDBN}
38116 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
38117 @pxref{qSupported}.
38118 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
38119 disabled via the @code{set remote noack-packet off} command
38120 (@pxref{Remote Configuration}),
38121 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
38122 Only then may the stub actually turn off packet acknowledgments.
38123 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
38124 response, which can be safely ignored by the stub.
38126 Note that @code{set remote noack-packet} command only affects negotiation
38127 between @value{GDBN} and the stub when subsequent connections are made;
38128 it does not affect the protocol acknowledgment state for any current
38130 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
38131 new connection is established,
38132 there is also no protocol request to re-enable the acknowledgments
38133 for the current connection, once disabled.
38138 Example sequence of a target being re-started. Notice how the restart
38139 does not get any direct output:
38144 @emph{target restarts}
38147 <- @code{T001:1234123412341234}
38151 Example sequence of a target being stepped by a single instruction:
38154 -> @code{G1445@dots{}}
38159 <- @code{T001:1234123412341234}
38163 <- @code{1455@dots{}}
38167 @node File-I/O Remote Protocol Extension
38168 @section File-I/O Remote Protocol Extension
38169 @cindex File-I/O remote protocol extension
38172 * File-I/O Overview::
38173 * Protocol Basics::
38174 * The F Request Packet::
38175 * The F Reply Packet::
38176 * The Ctrl-C Message::
38178 * List of Supported Calls::
38179 * Protocol-specific Representation of Datatypes::
38181 * File-I/O Examples::
38184 @node File-I/O Overview
38185 @subsection File-I/O Overview
38186 @cindex file-i/o overview
38188 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
38189 target to use the host's file system and console I/O to perform various
38190 system calls. System calls on the target system are translated into a
38191 remote protocol packet to the host system, which then performs the needed
38192 actions and returns a response packet to the target system.
38193 This simulates file system operations even on targets that lack file systems.
38195 The protocol is defined to be independent of both the host and target systems.
38196 It uses its own internal representation of datatypes and values. Both
38197 @value{GDBN} and the target's @value{GDBN} stub are responsible for
38198 translating the system-dependent value representations into the internal
38199 protocol representations when data is transmitted.
38201 The communication is synchronous. A system call is possible only when
38202 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
38203 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
38204 the target is stopped to allow deterministic access to the target's
38205 memory. Therefore File-I/O is not interruptible by target signals. On
38206 the other hand, it is possible to interrupt File-I/O by a user interrupt
38207 (@samp{Ctrl-C}) within @value{GDBN}.
38209 The target's request to perform a host system call does not finish
38210 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
38211 after finishing the system call, the target returns to continuing the
38212 previous activity (continue, step). No additional continue or step
38213 request from @value{GDBN} is required.
38216 (@value{GDBP}) continue
38217 <- target requests 'system call X'
38218 target is stopped, @value{GDBN} executes system call
38219 -> @value{GDBN} returns result
38220 ... target continues, @value{GDBN} returns to wait for the target
38221 <- target hits breakpoint and sends a Txx packet
38224 The protocol only supports I/O on the console and to regular files on
38225 the host file system. Character or block special devices, pipes,
38226 named pipes, sockets or any other communication method on the host
38227 system are not supported by this protocol.
38229 File I/O is not supported in non-stop mode.
38231 @node Protocol Basics
38232 @subsection Protocol Basics
38233 @cindex protocol basics, file-i/o
38235 The File-I/O protocol uses the @code{F} packet as the request as well
38236 as reply packet. Since a File-I/O system call can only occur when
38237 @value{GDBN} is waiting for a response from the continuing or stepping target,
38238 the File-I/O request is a reply that @value{GDBN} has to expect as a result
38239 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
38240 This @code{F} packet contains all information needed to allow @value{GDBN}
38241 to call the appropriate host system call:
38245 A unique identifier for the requested system call.
38248 All parameters to the system call. Pointers are given as addresses
38249 in the target memory address space. Pointers to strings are given as
38250 pointer/length pair. Numerical values are given as they are.
38251 Numerical control flags are given in a protocol-specific representation.
38255 At this point, @value{GDBN} has to perform the following actions.
38259 If the parameters include pointer values to data needed as input to a
38260 system call, @value{GDBN} requests this data from the target with a
38261 standard @code{m} packet request. This additional communication has to be
38262 expected by the target implementation and is handled as any other @code{m}
38266 @value{GDBN} translates all value from protocol representation to host
38267 representation as needed. Datatypes are coerced into the host types.
38270 @value{GDBN} calls the system call.
38273 It then coerces datatypes back to protocol representation.
38276 If the system call is expected to return data in buffer space specified
38277 by pointer parameters to the call, the data is transmitted to the
38278 target using a @code{M} or @code{X} packet. This packet has to be expected
38279 by the target implementation and is handled as any other @code{M} or @code{X}
38284 Eventually @value{GDBN} replies with another @code{F} packet which contains all
38285 necessary information for the target to continue. This at least contains
38292 @code{errno}, if has been changed by the system call.
38299 After having done the needed type and value coercion, the target continues
38300 the latest continue or step action.
38302 @node The F Request Packet
38303 @subsection The @code{F} Request Packet
38304 @cindex file-i/o request packet
38305 @cindex @code{F} request packet
38307 The @code{F} request packet has the following format:
38310 @item F@var{call-id},@var{parameter@dots{}}
38312 @var{call-id} is the identifier to indicate the host system call to be called.
38313 This is just the name of the function.
38315 @var{parameter@dots{}} are the parameters to the system call.
38316 Parameters are hexadecimal integer values, either the actual values in case
38317 of scalar datatypes, pointers to target buffer space in case of compound
38318 datatypes and unspecified memory areas, or pointer/length pairs in case
38319 of string parameters. These are appended to the @var{call-id} as a
38320 comma-delimited list. All values are transmitted in ASCII
38321 string representation, pointer/length pairs separated by a slash.
38327 @node The F Reply Packet
38328 @subsection The @code{F} Reply Packet
38329 @cindex file-i/o reply packet
38330 @cindex @code{F} reply packet
38332 The @code{F} reply packet has the following format:
38336 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
38338 @var{retcode} is the return code of the system call as hexadecimal value.
38340 @var{errno} is the @code{errno} set by the call, in protocol-specific
38342 This parameter can be omitted if the call was successful.
38344 @var{Ctrl-C flag} is only sent if the user requested a break. In this
38345 case, @var{errno} must be sent as well, even if the call was successful.
38346 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
38353 or, if the call was interrupted before the host call has been performed:
38360 assuming 4 is the protocol-specific representation of @code{EINTR}.
38365 @node The Ctrl-C Message
38366 @subsection The @samp{Ctrl-C} Message
38367 @cindex ctrl-c message, in file-i/o protocol
38369 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
38370 reply packet (@pxref{The F Reply Packet}),
38371 the target should behave as if it had
38372 gotten a break message. The meaning for the target is ``system call
38373 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
38374 (as with a break message) and return to @value{GDBN} with a @code{T02}
38377 It's important for the target to know in which
38378 state the system call was interrupted. There are two possible cases:
38382 The system call hasn't been performed on the host yet.
38385 The system call on the host has been finished.
38389 These two states can be distinguished by the target by the value of the
38390 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
38391 call hasn't been performed. This is equivalent to the @code{EINTR} handling
38392 on POSIX systems. In any other case, the target may presume that the
38393 system call has been finished --- successfully or not --- and should behave
38394 as if the break message arrived right after the system call.
38396 @value{GDBN} must behave reliably. If the system call has not been called
38397 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
38398 @code{errno} in the packet. If the system call on the host has been finished
38399 before the user requests a break, the full action must be finished by
38400 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
38401 The @code{F} packet may only be sent when either nothing has happened
38402 or the full action has been completed.
38405 @subsection Console I/O
38406 @cindex console i/o as part of file-i/o
38408 By default and if not explicitly closed by the target system, the file
38409 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
38410 on the @value{GDBN} console is handled as any other file output operation
38411 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
38412 by @value{GDBN} so that after the target read request from file descriptor
38413 0 all following typing is buffered until either one of the following
38418 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
38420 system call is treated as finished.
38423 The user presses @key{RET}. This is treated as end of input with a trailing
38427 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
38428 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
38432 If the user has typed more characters than fit in the buffer given to
38433 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
38434 either another @code{read(0, @dots{})} is requested by the target, or debugging
38435 is stopped at the user's request.
38438 @node List of Supported Calls
38439 @subsection List of Supported Calls
38440 @cindex list of supported file-i/o calls
38457 @unnumberedsubsubsec open
38458 @cindex open, file-i/o system call
38463 int open(const char *pathname, int flags);
38464 int open(const char *pathname, int flags, mode_t mode);
38468 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
38471 @var{flags} is the bitwise @code{OR} of the following values:
38475 If the file does not exist it will be created. The host
38476 rules apply as far as file ownership and time stamps
38480 When used with @code{O_CREAT}, if the file already exists it is
38481 an error and open() fails.
38484 If the file already exists and the open mode allows
38485 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
38486 truncated to zero length.
38489 The file is opened in append mode.
38492 The file is opened for reading only.
38495 The file is opened for writing only.
38498 The file is opened for reading and writing.
38502 Other bits are silently ignored.
38506 @var{mode} is the bitwise @code{OR} of the following values:
38510 User has read permission.
38513 User has write permission.
38516 Group has read permission.
38519 Group has write permission.
38522 Others have read permission.
38525 Others have write permission.
38529 Other bits are silently ignored.
38532 @item Return value:
38533 @code{open} returns the new file descriptor or -1 if an error
38540 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
38543 @var{pathname} refers to a directory.
38546 The requested access is not allowed.
38549 @var{pathname} was too long.
38552 A directory component in @var{pathname} does not exist.
38555 @var{pathname} refers to a device, pipe, named pipe or socket.
38558 @var{pathname} refers to a file on a read-only filesystem and
38559 write access was requested.
38562 @var{pathname} is an invalid pointer value.
38565 No space on device to create the file.
38568 The process already has the maximum number of files open.
38571 The limit on the total number of files open on the system
38575 The call was interrupted by the user.
38581 @unnumberedsubsubsec close
38582 @cindex close, file-i/o system call
38591 @samp{Fclose,@var{fd}}
38593 @item Return value:
38594 @code{close} returns zero on success, or -1 if an error occurred.
38600 @var{fd} isn't a valid open file descriptor.
38603 The call was interrupted by the user.
38609 @unnumberedsubsubsec read
38610 @cindex read, file-i/o system call
38615 int read(int fd, void *buf, unsigned int count);
38619 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
38621 @item Return value:
38622 On success, the number of bytes read is returned.
38623 Zero indicates end of file. If count is zero, read
38624 returns zero as well. On error, -1 is returned.
38630 @var{fd} is not a valid file descriptor or is not open for
38634 @var{bufptr} is an invalid pointer value.
38637 The call was interrupted by the user.
38643 @unnumberedsubsubsec write
38644 @cindex write, file-i/o system call
38649 int write(int fd, const void *buf, unsigned int count);
38653 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
38655 @item Return value:
38656 On success, the number of bytes written are returned.
38657 Zero indicates nothing was written. On error, -1
38664 @var{fd} is not a valid file descriptor or is not open for
38668 @var{bufptr} is an invalid pointer value.
38671 An attempt was made to write a file that exceeds the
38672 host-specific maximum file size allowed.
38675 No space on device to write the data.
38678 The call was interrupted by the user.
38684 @unnumberedsubsubsec lseek
38685 @cindex lseek, file-i/o system call
38690 long lseek (int fd, long offset, int flag);
38694 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
38696 @var{flag} is one of:
38700 The offset is set to @var{offset} bytes.
38703 The offset is set to its current location plus @var{offset}
38707 The offset is set to the size of the file plus @var{offset}
38711 @item Return value:
38712 On success, the resulting unsigned offset in bytes from
38713 the beginning of the file is returned. Otherwise, a
38714 value of -1 is returned.
38720 @var{fd} is not a valid open file descriptor.
38723 @var{fd} is associated with the @value{GDBN} console.
38726 @var{flag} is not a proper value.
38729 The call was interrupted by the user.
38735 @unnumberedsubsubsec rename
38736 @cindex rename, file-i/o system call
38741 int rename(const char *oldpath, const char *newpath);
38745 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
38747 @item Return value:
38748 On success, zero is returned. On error, -1 is returned.
38754 @var{newpath} is an existing directory, but @var{oldpath} is not a
38758 @var{newpath} is a non-empty directory.
38761 @var{oldpath} or @var{newpath} is a directory that is in use by some
38765 An attempt was made to make a directory a subdirectory
38769 A component used as a directory in @var{oldpath} or new
38770 path is not a directory. Or @var{oldpath} is a directory
38771 and @var{newpath} exists but is not a directory.
38774 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
38777 No access to the file or the path of the file.
38781 @var{oldpath} or @var{newpath} was too long.
38784 A directory component in @var{oldpath} or @var{newpath} does not exist.
38787 The file is on a read-only filesystem.
38790 The device containing the file has no room for the new
38794 The call was interrupted by the user.
38800 @unnumberedsubsubsec unlink
38801 @cindex unlink, file-i/o system call
38806 int unlink(const char *pathname);
38810 @samp{Funlink,@var{pathnameptr}/@var{len}}
38812 @item Return value:
38813 On success, zero is returned. On error, -1 is returned.
38819 No access to the file or the path of the file.
38822 The system does not allow unlinking of directories.
38825 The file @var{pathname} cannot be unlinked because it's
38826 being used by another process.
38829 @var{pathnameptr} is an invalid pointer value.
38832 @var{pathname} was too long.
38835 A directory component in @var{pathname} does not exist.
38838 A component of the path is not a directory.
38841 The file is on a read-only filesystem.
38844 The call was interrupted by the user.
38850 @unnumberedsubsubsec stat/fstat
38851 @cindex fstat, file-i/o system call
38852 @cindex stat, file-i/o system call
38857 int stat(const char *pathname, struct stat *buf);
38858 int fstat(int fd, struct stat *buf);
38862 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
38863 @samp{Ffstat,@var{fd},@var{bufptr}}
38865 @item Return value:
38866 On success, zero is returned. On error, -1 is returned.
38872 @var{fd} is not a valid open file.
38875 A directory component in @var{pathname} does not exist or the
38876 path is an empty string.
38879 A component of the path is not a directory.
38882 @var{pathnameptr} is an invalid pointer value.
38885 No access to the file or the path of the file.
38888 @var{pathname} was too long.
38891 The call was interrupted by the user.
38897 @unnumberedsubsubsec gettimeofday
38898 @cindex gettimeofday, file-i/o system call
38903 int gettimeofday(struct timeval *tv, void *tz);
38907 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
38909 @item Return value:
38910 On success, 0 is returned, -1 otherwise.
38916 @var{tz} is a non-NULL pointer.
38919 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
38925 @unnumberedsubsubsec isatty
38926 @cindex isatty, file-i/o system call
38931 int isatty(int fd);
38935 @samp{Fisatty,@var{fd}}
38937 @item Return value:
38938 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
38944 The call was interrupted by the user.
38949 Note that the @code{isatty} call is treated as a special case: it returns
38950 1 to the target if the file descriptor is attached
38951 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
38952 would require implementing @code{ioctl} and would be more complex than
38957 @unnumberedsubsubsec system
38958 @cindex system, file-i/o system call
38963 int system(const char *command);
38967 @samp{Fsystem,@var{commandptr}/@var{len}}
38969 @item Return value:
38970 If @var{len} is zero, the return value indicates whether a shell is
38971 available. A zero return value indicates a shell is not available.
38972 For non-zero @var{len}, the value returned is -1 on error and the
38973 return status of the command otherwise. Only the exit status of the
38974 command is returned, which is extracted from the host's @code{system}
38975 return value by calling @code{WEXITSTATUS(retval)}. In case
38976 @file{/bin/sh} could not be executed, 127 is returned.
38982 The call was interrupted by the user.
38987 @value{GDBN} takes over the full task of calling the necessary host calls
38988 to perform the @code{system} call. The return value of @code{system} on
38989 the host is simplified before it's returned
38990 to the target. Any termination signal information from the child process
38991 is discarded, and the return value consists
38992 entirely of the exit status of the called command.
38994 Due to security concerns, the @code{system} call is by default refused
38995 by @value{GDBN}. The user has to allow this call explicitly with the
38996 @code{set remote system-call-allowed 1} command.
38999 @item set remote system-call-allowed
39000 @kindex set remote system-call-allowed
39001 Control whether to allow the @code{system} calls in the File I/O
39002 protocol for the remote target. The default is zero (disabled).
39004 @item show remote system-call-allowed
39005 @kindex show remote system-call-allowed
39006 Show whether the @code{system} calls are allowed in the File I/O
39010 @node Protocol-specific Representation of Datatypes
39011 @subsection Protocol-specific Representation of Datatypes
39012 @cindex protocol-specific representation of datatypes, in file-i/o protocol
39015 * Integral Datatypes::
39017 * Memory Transfer::
39022 @node Integral Datatypes
39023 @unnumberedsubsubsec Integral Datatypes
39024 @cindex integral datatypes, in file-i/o protocol
39026 The integral datatypes used in the system calls are @code{int},
39027 @code{unsigned int}, @code{long}, @code{unsigned long},
39028 @code{mode_t}, and @code{time_t}.
39030 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
39031 implemented as 32 bit values in this protocol.
39033 @code{long} and @code{unsigned long} are implemented as 64 bit types.
39035 @xref{Limits}, for corresponding MIN and MAX values (similar to those
39036 in @file{limits.h}) to allow range checking on host and target.
39038 @code{time_t} datatypes are defined as seconds since the Epoch.
39040 All integral datatypes transferred as part of a memory read or write of a
39041 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
39044 @node Pointer Values
39045 @unnumberedsubsubsec Pointer Values
39046 @cindex pointer values, in file-i/o protocol
39048 Pointers to target data are transmitted as they are. An exception
39049 is made for pointers to buffers for which the length isn't
39050 transmitted as part of the function call, namely strings. Strings
39051 are transmitted as a pointer/length pair, both as hex values, e.g.@:
39058 which is a pointer to data of length 18 bytes at position 0x1aaf.
39059 The length is defined as the full string length in bytes, including
39060 the trailing null byte. For example, the string @code{"hello world"}
39061 at address 0x123456 is transmitted as
39067 @node Memory Transfer
39068 @unnumberedsubsubsec Memory Transfer
39069 @cindex memory transfer, in file-i/o protocol
39071 Structured data which is transferred using a memory read or write (for
39072 example, a @code{struct stat}) is expected to be in a protocol-specific format
39073 with all scalar multibyte datatypes being big endian. Translation to
39074 this representation needs to be done both by the target before the @code{F}
39075 packet is sent, and by @value{GDBN} before
39076 it transfers memory to the target. Transferred pointers to structured
39077 data should point to the already-coerced data at any time.
39081 @unnumberedsubsubsec struct stat
39082 @cindex struct stat, in file-i/o protocol
39084 The buffer of type @code{struct stat} used by the target and @value{GDBN}
39085 is defined as follows:
39089 unsigned int st_dev; /* device */
39090 unsigned int st_ino; /* inode */
39091 mode_t st_mode; /* protection */
39092 unsigned int st_nlink; /* number of hard links */
39093 unsigned int st_uid; /* user ID of owner */
39094 unsigned int st_gid; /* group ID of owner */
39095 unsigned int st_rdev; /* device type (if inode device) */
39096 unsigned long st_size; /* total size, in bytes */
39097 unsigned long st_blksize; /* blocksize for filesystem I/O */
39098 unsigned long st_blocks; /* number of blocks allocated */
39099 time_t st_atime; /* time of last access */
39100 time_t st_mtime; /* time of last modification */
39101 time_t st_ctime; /* time of last change */
39105 The integral datatypes conform to the definitions given in the
39106 appropriate section (see @ref{Integral Datatypes}, for details) so this
39107 structure is of size 64 bytes.
39109 The values of several fields have a restricted meaning and/or
39115 A value of 0 represents a file, 1 the console.
39118 No valid meaning for the target. Transmitted unchanged.
39121 Valid mode bits are described in @ref{Constants}. Any other
39122 bits have currently no meaning for the target.
39127 No valid meaning for the target. Transmitted unchanged.
39132 These values have a host and file system dependent
39133 accuracy. Especially on Windows hosts, the file system may not
39134 support exact timing values.
39137 The target gets a @code{struct stat} of the above representation and is
39138 responsible for coercing it to the target representation before
39141 Note that due to size differences between the host, target, and protocol
39142 representations of @code{struct stat} members, these members could eventually
39143 get truncated on the target.
39145 @node struct timeval
39146 @unnumberedsubsubsec struct timeval
39147 @cindex struct timeval, in file-i/o protocol
39149 The buffer of type @code{struct timeval} used by the File-I/O protocol
39150 is defined as follows:
39154 time_t tv_sec; /* second */
39155 long tv_usec; /* microsecond */
39159 The integral datatypes conform to the definitions given in the
39160 appropriate section (see @ref{Integral Datatypes}, for details) so this
39161 structure is of size 8 bytes.
39164 @subsection Constants
39165 @cindex constants, in file-i/o protocol
39167 The following values are used for the constants inside of the
39168 protocol. @value{GDBN} and target are responsible for translating these
39169 values before and after the call as needed.
39180 @unnumberedsubsubsec Open Flags
39181 @cindex open flags, in file-i/o protocol
39183 All values are given in hexadecimal representation.
39195 @node mode_t Values
39196 @unnumberedsubsubsec mode_t Values
39197 @cindex mode_t values, in file-i/o protocol
39199 All values are given in octal representation.
39216 @unnumberedsubsubsec Errno Values
39217 @cindex errno values, in file-i/o protocol
39219 All values are given in decimal representation.
39244 @code{EUNKNOWN} is used as a fallback error value if a host system returns
39245 any error value not in the list of supported error numbers.
39248 @unnumberedsubsubsec Lseek Flags
39249 @cindex lseek flags, in file-i/o protocol
39258 @unnumberedsubsubsec Limits
39259 @cindex limits, in file-i/o protocol
39261 All values are given in decimal representation.
39264 INT_MIN -2147483648
39266 UINT_MAX 4294967295
39267 LONG_MIN -9223372036854775808
39268 LONG_MAX 9223372036854775807
39269 ULONG_MAX 18446744073709551615
39272 @node File-I/O Examples
39273 @subsection File-I/O Examples
39274 @cindex file-i/o examples
39276 Example sequence of a write call, file descriptor 3, buffer is at target
39277 address 0x1234, 6 bytes should be written:
39280 <- @code{Fwrite,3,1234,6}
39281 @emph{request memory read from target}
39284 @emph{return "6 bytes written"}
39288 Example sequence of a read call, file descriptor 3, buffer is at target
39289 address 0x1234, 6 bytes should be read:
39292 <- @code{Fread,3,1234,6}
39293 @emph{request memory write to target}
39294 -> @code{X1234,6:XXXXXX}
39295 @emph{return "6 bytes read"}
39299 Example sequence of a read call, call fails on the host due to invalid
39300 file descriptor (@code{EBADF}):
39303 <- @code{Fread,3,1234,6}
39307 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
39311 <- @code{Fread,3,1234,6}
39316 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
39320 <- @code{Fread,3,1234,6}
39321 -> @code{X1234,6:XXXXXX}
39325 @node Library List Format
39326 @section Library List Format
39327 @cindex library list format, remote protocol
39329 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
39330 same process as your application to manage libraries. In this case,
39331 @value{GDBN} can use the loader's symbol table and normal memory
39332 operations to maintain a list of shared libraries. On other
39333 platforms, the operating system manages loaded libraries.
39334 @value{GDBN} can not retrieve the list of currently loaded libraries
39335 through memory operations, so it uses the @samp{qXfer:libraries:read}
39336 packet (@pxref{qXfer library list read}) instead. The remote stub
39337 queries the target's operating system and reports which libraries
39340 The @samp{qXfer:libraries:read} packet returns an XML document which
39341 lists loaded libraries and their offsets. Each library has an
39342 associated name and one or more segment or section base addresses,
39343 which report where the library was loaded in memory.
39345 For the common case of libraries that are fully linked binaries, the
39346 library should have a list of segments. If the target supports
39347 dynamic linking of a relocatable object file, its library XML element
39348 should instead include a list of allocated sections. The segment or
39349 section bases are start addresses, not relocation offsets; they do not
39350 depend on the library's link-time base addresses.
39352 @value{GDBN} must be linked with the Expat library to support XML
39353 library lists. @xref{Expat}.
39355 A simple memory map, with one loaded library relocated by a single
39356 offset, looks like this:
39360 <library name="/lib/libc.so.6">
39361 <segment address="0x10000000"/>
39366 Another simple memory map, with one loaded library with three
39367 allocated sections (.text, .data, .bss), looks like this:
39371 <library name="sharedlib.o">
39372 <section address="0x10000000"/>
39373 <section address="0x20000000"/>
39374 <section address="0x30000000"/>
39379 The format of a library list is described by this DTD:
39382 <!-- library-list: Root element with versioning -->
39383 <!ELEMENT library-list (library)*>
39384 <!ATTLIST library-list version CDATA #FIXED "1.0">
39385 <!ELEMENT library (segment*, section*)>
39386 <!ATTLIST library name CDATA #REQUIRED>
39387 <!ELEMENT segment EMPTY>
39388 <!ATTLIST segment address CDATA #REQUIRED>
39389 <!ELEMENT section EMPTY>
39390 <!ATTLIST section address CDATA #REQUIRED>
39393 In addition, segments and section descriptors cannot be mixed within a
39394 single library element, and you must supply at least one segment or
39395 section for each library.
39397 @node Library List Format for SVR4 Targets
39398 @section Library List Format for SVR4 Targets
39399 @cindex library list format, remote protocol
39401 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
39402 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
39403 shared libraries. Still a special library list provided by this packet is
39404 more efficient for the @value{GDBN} remote protocol.
39406 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
39407 loaded libraries and their SVR4 linker parameters. For each library on SVR4
39408 target, the following parameters are reported:
39412 @code{name}, the absolute file name from the @code{l_name} field of
39413 @code{struct link_map}.
39415 @code{lm} with address of @code{struct link_map} used for TLS
39416 (Thread Local Storage) access.
39418 @code{l_addr}, the displacement as read from the field @code{l_addr} of
39419 @code{struct link_map}. For prelinked libraries this is not an absolute
39420 memory address. It is a displacement of absolute memory address against
39421 address the file was prelinked to during the library load.
39423 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
39426 Additionally the single @code{main-lm} attribute specifies address of
39427 @code{struct link_map} used for the main executable. This parameter is used
39428 for TLS access and its presence is optional.
39430 @value{GDBN} must be linked with the Expat library to support XML
39431 SVR4 library lists. @xref{Expat}.
39433 A simple memory map, with two loaded libraries (which do not use prelink),
39437 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
39438 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
39440 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
39442 </library-list-svr>
39445 The format of an SVR4 library list is described by this DTD:
39448 <!-- library-list-svr4: Root element with versioning -->
39449 <!ELEMENT library-list-svr4 (library)*>
39450 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
39451 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
39452 <!ELEMENT library EMPTY>
39453 <!ATTLIST library name CDATA #REQUIRED>
39454 <!ATTLIST library lm CDATA #REQUIRED>
39455 <!ATTLIST library l_addr CDATA #REQUIRED>
39456 <!ATTLIST library l_ld CDATA #REQUIRED>
39459 @node Memory Map Format
39460 @section Memory Map Format
39461 @cindex memory map format
39463 To be able to write into flash memory, @value{GDBN} needs to obtain a
39464 memory map from the target. This section describes the format of the
39467 The memory map is obtained using the @samp{qXfer:memory-map:read}
39468 (@pxref{qXfer memory map read}) packet and is an XML document that
39469 lists memory regions.
39471 @value{GDBN} must be linked with the Expat library to support XML
39472 memory maps. @xref{Expat}.
39474 The top-level structure of the document is shown below:
39477 <?xml version="1.0"?>
39478 <!DOCTYPE memory-map
39479 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
39480 "http://sourceware.org/gdb/gdb-memory-map.dtd">
39486 Each region can be either:
39491 A region of RAM starting at @var{addr} and extending for @var{length}
39495 <memory type="ram" start="@var{addr}" length="@var{length}"/>
39500 A region of read-only memory:
39503 <memory type="rom" start="@var{addr}" length="@var{length}"/>
39508 A region of flash memory, with erasure blocks @var{blocksize}
39512 <memory type="flash" start="@var{addr}" length="@var{length}">
39513 <property name="blocksize">@var{blocksize}</property>
39519 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
39520 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
39521 packets to write to addresses in such ranges.
39523 The formal DTD for memory map format is given below:
39526 <!-- ................................................... -->
39527 <!-- Memory Map XML DTD ................................ -->
39528 <!-- File: memory-map.dtd .............................. -->
39529 <!-- .................................... .............. -->
39530 <!-- memory-map.dtd -->
39531 <!-- memory-map: Root element with versioning -->
39532 <!ELEMENT memory-map (memory | property)>
39533 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
39534 <!ELEMENT memory (property)>
39535 <!-- memory: Specifies a memory region,
39536 and its type, or device. -->
39537 <!ATTLIST memory type CDATA #REQUIRED
39538 start CDATA #REQUIRED
39539 length CDATA #REQUIRED
39540 device CDATA #IMPLIED>
39541 <!-- property: Generic attribute tag -->
39542 <!ELEMENT property (#PCDATA | property)*>
39543 <!ATTLIST property name CDATA #REQUIRED>
39546 @node Thread List Format
39547 @section Thread List Format
39548 @cindex thread list format
39550 To efficiently update the list of threads and their attributes,
39551 @value{GDBN} issues the @samp{qXfer:threads:read} packet
39552 (@pxref{qXfer threads read}) and obtains the XML document with
39553 the following structure:
39556 <?xml version="1.0"?>
39558 <thread id="id" core="0">
39559 ... description ...
39564 Each @samp{thread} element must have the @samp{id} attribute that
39565 identifies the thread (@pxref{thread-id syntax}). The
39566 @samp{core} attribute, if present, specifies which processor core
39567 the thread was last executing on. The content of the of @samp{thread}
39568 element is interpreted as human-readable auxilliary information.
39570 @node Traceframe Info Format
39571 @section Traceframe Info Format
39572 @cindex traceframe info format
39574 To be able to know which objects in the inferior can be examined when
39575 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
39576 memory ranges, registers and trace state variables that have been
39577 collected in a traceframe.
39579 This list is obtained using the @samp{qXfer:traceframe-info:read}
39580 (@pxref{qXfer traceframe info read}) packet and is an XML document.
39582 @value{GDBN} must be linked with the Expat library to support XML
39583 traceframe info discovery. @xref{Expat}.
39585 The top-level structure of the document is shown below:
39588 <?xml version="1.0"?>
39589 <!DOCTYPE traceframe-info
39590 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
39591 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
39597 Each traceframe block can be either:
39602 A region of collected memory starting at @var{addr} and extending for
39603 @var{length} bytes from there:
39606 <memory start="@var{addr}" length="@var{length}"/>
39611 The formal DTD for the traceframe info format is given below:
39614 <!ELEMENT traceframe-info (memory)* >
39615 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
39617 <!ELEMENT memory EMPTY>
39618 <!ATTLIST memory start CDATA #REQUIRED
39619 length CDATA #REQUIRED>
39622 @include agentexpr.texi
39624 @node Target Descriptions
39625 @appendix Target Descriptions
39626 @cindex target descriptions
39628 One of the challenges of using @value{GDBN} to debug embedded systems
39629 is that there are so many minor variants of each processor
39630 architecture in use. It is common practice for vendors to start with
39631 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
39632 and then make changes to adapt it to a particular market niche. Some
39633 architectures have hundreds of variants, available from dozens of
39634 vendors. This leads to a number of problems:
39638 With so many different customized processors, it is difficult for
39639 the @value{GDBN} maintainers to keep up with the changes.
39641 Since individual variants may have short lifetimes or limited
39642 audiences, it may not be worthwhile to carry information about every
39643 variant in the @value{GDBN} source tree.
39645 When @value{GDBN} does support the architecture of the embedded system
39646 at hand, the task of finding the correct architecture name to give the
39647 @command{set architecture} command can be error-prone.
39650 To address these problems, the @value{GDBN} remote protocol allows a
39651 target system to not only identify itself to @value{GDBN}, but to
39652 actually describe its own features. This lets @value{GDBN} support
39653 processor variants it has never seen before --- to the extent that the
39654 descriptions are accurate, and that @value{GDBN} understands them.
39656 @value{GDBN} must be linked with the Expat library to support XML
39657 target descriptions. @xref{Expat}.
39660 * Retrieving Descriptions:: How descriptions are fetched from a target.
39661 * Target Description Format:: The contents of a target description.
39662 * Predefined Target Types:: Standard types available for target
39664 * Standard Target Features:: Features @value{GDBN} knows about.
39667 @node Retrieving Descriptions
39668 @section Retrieving Descriptions
39670 Target descriptions can be read from the target automatically, or
39671 specified by the user manually. The default behavior is to read the
39672 description from the target. @value{GDBN} retrieves it via the remote
39673 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
39674 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
39675 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
39676 XML document, of the form described in @ref{Target Description
39679 Alternatively, you can specify a file to read for the target description.
39680 If a file is set, the target will not be queried. The commands to
39681 specify a file are:
39684 @cindex set tdesc filename
39685 @item set tdesc filename @var{path}
39686 Read the target description from @var{path}.
39688 @cindex unset tdesc filename
39689 @item unset tdesc filename
39690 Do not read the XML target description from a file. @value{GDBN}
39691 will use the description supplied by the current target.
39693 @cindex show tdesc filename
39694 @item show tdesc filename
39695 Show the filename to read for a target description, if any.
39699 @node Target Description Format
39700 @section Target Description Format
39701 @cindex target descriptions, XML format
39703 A target description annex is an @uref{http://www.w3.org/XML/, XML}
39704 document which complies with the Document Type Definition provided in
39705 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
39706 means you can use generally available tools like @command{xmllint} to
39707 check that your feature descriptions are well-formed and valid.
39708 However, to help people unfamiliar with XML write descriptions for
39709 their targets, we also describe the grammar here.
39711 Target descriptions can identify the architecture of the remote target
39712 and (for some architectures) provide information about custom register
39713 sets. They can also identify the OS ABI of the remote target.
39714 @value{GDBN} can use this information to autoconfigure for your
39715 target, or to warn you if you connect to an unsupported target.
39717 Here is a simple target description:
39720 <target version="1.0">
39721 <architecture>i386:x86-64</architecture>
39726 This minimal description only says that the target uses
39727 the x86-64 architecture.
39729 A target description has the following overall form, with [ ] marking
39730 optional elements and @dots{} marking repeatable elements. The elements
39731 are explained further below.
39734 <?xml version="1.0"?>
39735 <!DOCTYPE target SYSTEM "gdb-target.dtd">
39736 <target version="1.0">
39737 @r{[}@var{architecture}@r{]}
39738 @r{[}@var{osabi}@r{]}
39739 @r{[}@var{compatible}@r{]}
39740 @r{[}@var{feature}@dots{}@r{]}
39745 The description is generally insensitive to whitespace and line
39746 breaks, under the usual common-sense rules. The XML version
39747 declaration and document type declaration can generally be omitted
39748 (@value{GDBN} does not require them), but specifying them may be
39749 useful for XML validation tools. The @samp{version} attribute for
39750 @samp{<target>} may also be omitted, but we recommend
39751 including it; if future versions of @value{GDBN} use an incompatible
39752 revision of @file{gdb-target.dtd}, they will detect and report
39753 the version mismatch.
39755 @subsection Inclusion
39756 @cindex target descriptions, inclusion
39759 @cindex <xi:include>
39762 It can sometimes be valuable to split a target description up into
39763 several different annexes, either for organizational purposes, or to
39764 share files between different possible target descriptions. You can
39765 divide a description into multiple files by replacing any element of
39766 the target description with an inclusion directive of the form:
39769 <xi:include href="@var{document}"/>
39773 When @value{GDBN} encounters an element of this form, it will retrieve
39774 the named XML @var{document}, and replace the inclusion directive with
39775 the contents of that document. If the current description was read
39776 using @samp{qXfer}, then so will be the included document;
39777 @var{document} will be interpreted as the name of an annex. If the
39778 current description was read from a file, @value{GDBN} will look for
39779 @var{document} as a file in the same directory where it found the
39780 original description.
39782 @subsection Architecture
39783 @cindex <architecture>
39785 An @samp{<architecture>} element has this form:
39788 <architecture>@var{arch}</architecture>
39791 @var{arch} is one of the architectures from the set accepted by
39792 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
39795 @cindex @code{<osabi>}
39797 This optional field was introduced in @value{GDBN} version 7.0.
39798 Previous versions of @value{GDBN} ignore it.
39800 An @samp{<osabi>} element has this form:
39803 <osabi>@var{abi-name}</osabi>
39806 @var{abi-name} is an OS ABI name from the same selection accepted by
39807 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
39809 @subsection Compatible Architecture
39810 @cindex @code{<compatible>}
39812 This optional field was introduced in @value{GDBN} version 7.0.
39813 Previous versions of @value{GDBN} ignore it.
39815 A @samp{<compatible>} element has this form:
39818 <compatible>@var{arch}</compatible>
39821 @var{arch} is one of the architectures from the set accepted by
39822 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
39824 A @samp{<compatible>} element is used to specify that the target
39825 is able to run binaries in some other than the main target architecture
39826 given by the @samp{<architecture>} element. For example, on the
39827 Cell Broadband Engine, the main architecture is @code{powerpc:common}
39828 or @code{powerpc:common64}, but the system is able to run binaries
39829 in the @code{spu} architecture as well. The way to describe this
39830 capability with @samp{<compatible>} is as follows:
39833 <architecture>powerpc:common</architecture>
39834 <compatible>spu</compatible>
39837 @subsection Features
39840 Each @samp{<feature>} describes some logical portion of the target
39841 system. Features are currently used to describe available CPU
39842 registers and the types of their contents. A @samp{<feature>} element
39846 <feature name="@var{name}">
39847 @r{[}@var{type}@dots{}@r{]}
39853 Each feature's name should be unique within the description. The name
39854 of a feature does not matter unless @value{GDBN} has some special
39855 knowledge of the contents of that feature; if it does, the feature
39856 should have its standard name. @xref{Standard Target Features}.
39860 Any register's value is a collection of bits which @value{GDBN} must
39861 interpret. The default interpretation is a two's complement integer,
39862 but other types can be requested by name in the register description.
39863 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
39864 Target Types}), and the description can define additional composite types.
39866 Each type element must have an @samp{id} attribute, which gives
39867 a unique (within the containing @samp{<feature>}) name to the type.
39868 Types must be defined before they are used.
39871 Some targets offer vector registers, which can be treated as arrays
39872 of scalar elements. These types are written as @samp{<vector>} elements,
39873 specifying the array element type, @var{type}, and the number of elements,
39877 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
39881 If a register's value is usefully viewed in multiple ways, define it
39882 with a union type containing the useful representations. The
39883 @samp{<union>} element contains one or more @samp{<field>} elements,
39884 each of which has a @var{name} and a @var{type}:
39887 <union id="@var{id}">
39888 <field name="@var{name}" type="@var{type}"/>
39894 If a register's value is composed from several separate values, define
39895 it with a structure type. There are two forms of the @samp{<struct>}
39896 element; a @samp{<struct>} element must either contain only bitfields
39897 or contain no bitfields. If the structure contains only bitfields,
39898 its total size in bytes must be specified, each bitfield must have an
39899 explicit start and end, and bitfields are automatically assigned an
39900 integer type. The field's @var{start} should be less than or
39901 equal to its @var{end}, and zero represents the least significant bit.
39904 <struct id="@var{id}" size="@var{size}">
39905 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
39910 If the structure contains no bitfields, then each field has an
39911 explicit type, and no implicit padding is added.
39914 <struct id="@var{id}">
39915 <field name="@var{name}" type="@var{type}"/>
39921 If a register's value is a series of single-bit flags, define it with
39922 a flags type. The @samp{<flags>} element has an explicit @var{size}
39923 and contains one or more @samp{<field>} elements. Each field has a
39924 @var{name}, a @var{start}, and an @var{end}. Only single-bit flags
39928 <flags id="@var{id}" size="@var{size}">
39929 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
39934 @subsection Registers
39937 Each register is represented as an element with this form:
39940 <reg name="@var{name}"
39941 bitsize="@var{size}"
39942 @r{[}regnum="@var{num}"@r{]}
39943 @r{[}save-restore="@var{save-restore}"@r{]}
39944 @r{[}type="@var{type}"@r{]}
39945 @r{[}group="@var{group}"@r{]}/>
39949 The components are as follows:
39954 The register's name; it must be unique within the target description.
39957 The register's size, in bits.
39960 The register's number. If omitted, a register's number is one greater
39961 than that of the previous register (either in the current feature or in
39962 a preceding feature); the first register in the target description
39963 defaults to zero. This register number is used to read or write
39964 the register; e.g.@: it is used in the remote @code{p} and @code{P}
39965 packets, and registers appear in the @code{g} and @code{G} packets
39966 in order of increasing register number.
39969 Whether the register should be preserved across inferior function
39970 calls; this must be either @code{yes} or @code{no}. The default is
39971 @code{yes}, which is appropriate for most registers except for
39972 some system control registers; this is not related to the target's
39976 The type of the register. @var{type} may be a predefined type, a type
39977 defined in the current feature, or one of the special types @code{int}
39978 and @code{float}. @code{int} is an integer type of the correct size
39979 for @var{bitsize}, and @code{float} is a floating point type (in the
39980 architecture's normal floating point format) of the correct size for
39981 @var{bitsize}. The default is @code{int}.
39984 The register group to which this register belongs. @var{group} must
39985 be either @code{general}, @code{float}, or @code{vector}. If no
39986 @var{group} is specified, @value{GDBN} will not display the register
39987 in @code{info registers}.
39991 @node Predefined Target Types
39992 @section Predefined Target Types
39993 @cindex target descriptions, predefined types
39995 Type definitions in the self-description can build up composite types
39996 from basic building blocks, but can not define fundamental types. Instead,
39997 standard identifiers are provided by @value{GDBN} for the fundamental
39998 types. The currently supported types are:
40007 Signed integer types holding the specified number of bits.
40014 Unsigned integer types holding the specified number of bits.
40018 Pointers to unspecified code and data. The program counter and
40019 any dedicated return address register may be marked as code
40020 pointers; printing a code pointer converts it into a symbolic
40021 address. The stack pointer and any dedicated address registers
40022 may be marked as data pointers.
40025 Single precision IEEE floating point.
40028 Double precision IEEE floating point.
40031 The 12-byte extended precision format used by ARM FPA registers.
40034 The 10-byte extended precision format used by x87 registers.
40037 32bit @sc{eflags} register used by x86.
40040 32bit @sc{mxcsr} register used by x86.
40044 @node Standard Target Features
40045 @section Standard Target Features
40046 @cindex target descriptions, standard features
40048 A target description must contain either no registers or all the
40049 target's registers. If the description contains no registers, then
40050 @value{GDBN} will assume a default register layout, selected based on
40051 the architecture. If the description contains any registers, the
40052 default layout will not be used; the standard registers must be
40053 described in the target description, in such a way that @value{GDBN}
40054 can recognize them.
40056 This is accomplished by giving specific names to feature elements
40057 which contain standard registers. @value{GDBN} will look for features
40058 with those names and verify that they contain the expected registers;
40059 if any known feature is missing required registers, or if any required
40060 feature is missing, @value{GDBN} will reject the target
40061 description. You can add additional registers to any of the
40062 standard features --- @value{GDBN} will display them just as if
40063 they were added to an unrecognized feature.
40065 This section lists the known features and their expected contents.
40066 Sample XML documents for these features are included in the
40067 @value{GDBN} source tree, in the directory @file{gdb/features}.
40069 Names recognized by @value{GDBN} should include the name of the
40070 company or organization which selected the name, and the overall
40071 architecture to which the feature applies; so e.g.@: the feature
40072 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
40074 The names of registers are not case sensitive for the purpose
40075 of recognizing standard features, but @value{GDBN} will only display
40076 registers using the capitalization used in the description.
40083 * PowerPC Features::
40089 @subsection ARM Features
40090 @cindex target descriptions, ARM features
40092 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
40094 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
40095 @samp{lr}, @samp{pc}, and @samp{cpsr}.
40097 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
40098 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
40099 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
40102 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
40103 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
40105 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
40106 it should contain at least registers @samp{wR0} through @samp{wR15} and
40107 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
40108 @samp{wCSSF}, and @samp{wCASF} registers are optional.
40110 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
40111 should contain at least registers @samp{d0} through @samp{d15}. If
40112 they are present, @samp{d16} through @samp{d31} should also be included.
40113 @value{GDBN} will synthesize the single-precision registers from
40114 halves of the double-precision registers.
40116 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
40117 need to contain registers; it instructs @value{GDBN} to display the
40118 VFP double-precision registers as vectors and to synthesize the
40119 quad-precision registers from pairs of double-precision registers.
40120 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
40121 be present and include 32 double-precision registers.
40123 @node i386 Features
40124 @subsection i386 Features
40125 @cindex target descriptions, i386 features
40127 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
40128 targets. It should describe the following registers:
40132 @samp{eax} through @samp{edi} plus @samp{eip} for i386
40134 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
40136 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
40137 @samp{fs}, @samp{gs}
40139 @samp{st0} through @samp{st7}
40141 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
40142 @samp{foseg}, @samp{fooff} and @samp{fop}
40145 The register sets may be different, depending on the target.
40147 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
40148 describe registers:
40152 @samp{xmm0} through @samp{xmm7} for i386
40154 @samp{xmm0} through @samp{xmm15} for amd64
40159 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
40160 @samp{org.gnu.gdb.i386.sse} feature. It should
40161 describe the upper 128 bits of @sc{ymm} registers:
40165 @samp{ymm0h} through @samp{ymm7h} for i386
40167 @samp{ymm0h} through @samp{ymm15h} for amd64
40170 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
40171 describe a single register, @samp{orig_eax}.
40173 @node MIPS Features
40174 @subsection @acronym{MIPS} Features
40175 @cindex target descriptions, @acronym{MIPS} features
40177 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
40178 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
40179 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
40182 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
40183 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
40184 registers. They may be 32-bit or 64-bit depending on the target.
40186 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
40187 it may be optional in a future version of @value{GDBN}. It should
40188 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
40189 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
40191 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
40192 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
40193 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
40194 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
40196 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
40197 contain a single register, @samp{restart}, which is used by the
40198 Linux kernel to control restartable syscalls.
40200 @node M68K Features
40201 @subsection M68K Features
40202 @cindex target descriptions, M68K features
40205 @item @samp{org.gnu.gdb.m68k.core}
40206 @itemx @samp{org.gnu.gdb.coldfire.core}
40207 @itemx @samp{org.gnu.gdb.fido.core}
40208 One of those features must be always present.
40209 The feature that is present determines which flavor of m68k is
40210 used. The feature that is present should contain registers
40211 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
40212 @samp{sp}, @samp{ps} and @samp{pc}.
40214 @item @samp{org.gnu.gdb.coldfire.fp}
40215 This feature is optional. If present, it should contain registers
40216 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
40220 @node PowerPC Features
40221 @subsection PowerPC Features
40222 @cindex target descriptions, PowerPC features
40224 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
40225 targets. It should contain registers @samp{r0} through @samp{r31},
40226 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
40227 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
40229 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
40230 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
40232 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
40233 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
40236 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
40237 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
40238 will combine these registers with the floating point registers
40239 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
40240 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
40241 through @samp{vs63}, the set of vector registers for POWER7.
40243 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
40244 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
40245 @samp{spefscr}. SPE targets should provide 32-bit registers in
40246 @samp{org.gnu.gdb.power.core} and provide the upper halves in
40247 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
40248 these to present registers @samp{ev0} through @samp{ev31} to the
40251 @node TIC6x Features
40252 @subsection TMS320C6x Features
40253 @cindex target descriptions, TIC6x features
40254 @cindex target descriptions, TMS320C6x features
40255 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
40256 targets. It should contain registers @samp{A0} through @samp{A15},
40257 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
40259 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
40260 contain registers @samp{A16} through @samp{A31} and @samp{B16}
40261 through @samp{B31}.
40263 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
40264 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
40266 @node Operating System Information
40267 @appendix Operating System Information
40268 @cindex operating system information
40274 Users of @value{GDBN} often wish to obtain information about the state of
40275 the operating system running on the target---for example the list of
40276 processes, or the list of open files. This section describes the
40277 mechanism that makes it possible. This mechanism is similar to the
40278 target features mechanism (@pxref{Target Descriptions}), but focuses
40279 on a different aspect of target.
40281 Operating system information is retrived from the target via the
40282 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
40283 read}). The object name in the request should be @samp{osdata}, and
40284 the @var{annex} identifies the data to be fetched.
40287 @appendixsection Process list
40288 @cindex operating system information, process list
40290 When requesting the process list, the @var{annex} field in the
40291 @samp{qXfer} request should be @samp{processes}. The returned data is
40292 an XML document. The formal syntax of this document is defined in
40293 @file{gdb/features/osdata.dtd}.
40295 An example document is:
40298 <?xml version="1.0"?>
40299 <!DOCTYPE target SYSTEM "osdata.dtd">
40300 <osdata type="processes">
40302 <column name="pid">1</column>
40303 <column name="user">root</column>
40304 <column name="command">/sbin/init</column>
40305 <column name="cores">1,2,3</column>
40310 Each item should include a column whose name is @samp{pid}. The value
40311 of that column should identify the process on the target. The
40312 @samp{user} and @samp{command} columns are optional, and will be
40313 displayed by @value{GDBN}. The @samp{cores} column, if present,
40314 should contain a comma-separated list of cores that this process
40315 is running on. Target may provide additional columns,
40316 which @value{GDBN} currently ignores.
40318 @node Trace File Format
40319 @appendix Trace File Format
40320 @cindex trace file format
40322 The trace file comes in three parts: a header, a textual description
40323 section, and a trace frame section with binary data.
40325 The header has the form @code{\x7fTRACE0\n}. The first byte is
40326 @code{0x7f} so as to indicate that the file contains binary data,
40327 while the @code{0} is a version number that may have different values
40330 The description section consists of multiple lines of @sc{ascii} text
40331 separated by newline characters (@code{0xa}). The lines may include a
40332 variety of optional descriptive or context-setting information, such
40333 as tracepoint definitions or register set size. @value{GDBN} will
40334 ignore any line that it does not recognize. An empty line marks the end
40337 @c FIXME add some specific types of data
40339 The trace frame section consists of a number of consecutive frames.
40340 Each frame begins with a two-byte tracepoint number, followed by a
40341 four-byte size giving the amount of data in the frame. The data in
40342 the frame consists of a number of blocks, each introduced by a
40343 character indicating its type (at least register, memory, and trace
40344 state variable). The data in this section is raw binary, not a
40345 hexadecimal or other encoding; its endianness matches the target's
40348 @c FIXME bi-arch may require endianness/arch info in description section
40351 @item R @var{bytes}
40352 Register block. The number and ordering of bytes matches that of a
40353 @code{g} packet in the remote protocol. Note that these are the
40354 actual bytes, in target order and @value{GDBN} register order, not a
40355 hexadecimal encoding.
40357 @item M @var{address} @var{length} @var{bytes}...
40358 Memory block. This is a contiguous block of memory, at the 8-byte
40359 address @var{address}, with a 2-byte length @var{length}, followed by
40360 @var{length} bytes.
40362 @item V @var{number} @var{value}
40363 Trace state variable block. This records the 8-byte signed value
40364 @var{value} of trace state variable numbered @var{number}.
40368 Future enhancements of the trace file format may include additional types
40371 @node Index Section Format
40372 @appendix @code{.gdb_index} section format
40373 @cindex .gdb_index section format
40374 @cindex index section format
40376 This section documents the index section that is created by @code{save
40377 gdb-index} (@pxref{Index Files}). The index section is
40378 DWARF-specific; some knowledge of DWARF is assumed in this
40381 The mapped index file format is designed to be directly
40382 @code{mmap}able on any architecture. In most cases, a datum is
40383 represented using a little-endian 32-bit integer value, called an
40384 @code{offset_type}. Big endian machines must byte-swap the values
40385 before using them. Exceptions to this rule are noted. The data is
40386 laid out such that alignment is always respected.
40388 A mapped index consists of several areas, laid out in order.
40392 The file header. This is a sequence of values, of @code{offset_type}
40393 unless otherwise noted:
40397 The version number, currently 7. Versions 1, 2 and 3 are obsolete.
40398 Version 4 uses a different hashing function from versions 5 and 6.
40399 Version 6 includes symbols for inlined functions, whereas versions 4
40400 and 5 do not. Version 7 adds attributes to the CU indices in the
40401 symbol table. @value{GDBN} will only read version 4, 5, or 6 indices
40402 if the @code{--use-deprecated-index-sections} option is used.
40405 The offset, from the start of the file, of the CU list.
40408 The offset, from the start of the file, of the types CU list. Note
40409 that this area can be empty, in which case this offset will be equal
40410 to the next offset.
40413 The offset, from the start of the file, of the address area.
40416 The offset, from the start of the file, of the symbol table.
40419 The offset, from the start of the file, of the constant pool.
40423 The CU list. This is a sequence of pairs of 64-bit little-endian
40424 values, sorted by the CU offset. The first element in each pair is
40425 the offset of a CU in the @code{.debug_info} section. The second
40426 element in each pair is the length of that CU. References to a CU
40427 elsewhere in the map are done using a CU index, which is just the
40428 0-based index into this table. Note that if there are type CUs, then
40429 conceptually CUs and type CUs form a single list for the purposes of
40433 The types CU list. This is a sequence of triplets of 64-bit
40434 little-endian values. In a triplet, the first value is the CU offset,
40435 the second value is the type offset in the CU, and the third value is
40436 the type signature. The types CU list is not sorted.
40439 The address area. The address area consists of a sequence of address
40440 entries. Each address entry has three elements:
40444 The low address. This is a 64-bit little-endian value.
40447 The high address. This is a 64-bit little-endian value. Like
40448 @code{DW_AT_high_pc}, the value is one byte beyond the end.
40451 The CU index. This is an @code{offset_type} value.
40455 The symbol table. This is an open-addressed hash table. The size of
40456 the hash table is always a power of 2.
40458 Each slot in the hash table consists of a pair of @code{offset_type}
40459 values. The first value is the offset of the symbol's name in the
40460 constant pool. The second value is the offset of the CU vector in the
40463 If both values are 0, then this slot in the hash table is empty. This
40464 is ok because while 0 is a valid constant pool index, it cannot be a
40465 valid index for both a string and a CU vector.
40467 The hash value for a table entry is computed by applying an
40468 iterative hash function to the symbol's name. Starting with an
40469 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
40470 the string is incorporated into the hash using the formula depending on the
40475 The formula is @code{r = r * 67 + c - 113}.
40477 @item Versions 5 to 7
40478 The formula is @code{r = r * 67 + tolower (c) - 113}.
40481 The terminating @samp{\0} is not incorporated into the hash.
40483 The step size used in the hash table is computed via
40484 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
40485 value, and @samp{size} is the size of the hash table. The step size
40486 is used to find the next candidate slot when handling a hash
40489 The names of C@t{++} symbols in the hash table are canonicalized. We
40490 don't currently have a simple description of the canonicalization
40491 algorithm; if you intend to create new index sections, you must read
40495 The constant pool. This is simply a bunch of bytes. It is organized
40496 so that alignment is correct: CU vectors are stored first, followed by
40499 A CU vector in the constant pool is a sequence of @code{offset_type}
40500 values. The first value is the number of CU indices in the vector.
40501 Each subsequent value is the index and symbol attributes of a CU in
40502 the CU list. This element in the hash table is used to indicate which
40503 CUs define the symbol and how the symbol is used.
40504 See below for the format of each CU index+attributes entry.
40506 A string in the constant pool is zero-terminated.
40509 Attributes were added to CU index values in @code{.gdb_index} version 7.
40510 If a symbol has multiple uses within a CU then there is one
40511 CU index+attributes value for each use.
40513 The format of each CU index+attributes entry is as follows
40519 This is the index of the CU in the CU list.
40521 These bits are reserved for future purposes and must be zero.
40523 The kind of the symbol in the CU.
40527 This value is reserved and should not be used.
40528 By reserving zero the full @code{offset_type} value is backwards compatible
40529 with previous versions of the index.
40531 The symbol is a type.
40533 The symbol is a variable or an enum value.
40535 The symbol is a function.
40537 Any other kind of symbol.
40539 These values are reserved.
40543 This bit is zero if the value is global and one if it is static.
40545 The determination of whether a symbol is global or static is complicated.
40546 The authorative reference is the file @file{dwarf2read.c} in
40547 @value{GDBN} sources.
40551 This pseudo-code describes the computation of a symbol's kind and
40552 global/static attributes in the index.
40555 is_external = get_attribute (die, DW_AT_external);
40556 language = get_attribute (cu_die, DW_AT_language);
40559 case DW_TAG_typedef:
40560 case DW_TAG_base_type:
40561 case DW_TAG_subrange_type:
40565 case DW_TAG_enumerator:
40567 is_static = (language != CPLUS && language != JAVA);
40569 case DW_TAG_subprogram:
40571 is_static = ! (is_external || language == ADA);
40573 case DW_TAG_constant:
40575 is_static = ! is_external;
40577 case DW_TAG_variable:
40579 is_static = ! is_external;
40581 case DW_TAG_namespace:
40585 case DW_TAG_class_type:
40586 case DW_TAG_interface_type:
40587 case DW_TAG_structure_type:
40588 case DW_TAG_union_type:
40589 case DW_TAG_enumeration_type:
40591 is_static = (language != CPLUS && language != JAVA);
40600 @node GNU Free Documentation License
40601 @appendix GNU Free Documentation License
40610 % I think something like @@colophon should be in texinfo. In the
40612 \long\def\colophon{\hbox to0pt{}\vfill
40613 \centerline{The body of this manual is set in}
40614 \centerline{\fontname\tenrm,}
40615 \centerline{with headings in {\bf\fontname\tenbf}}
40616 \centerline{and examples in {\tt\fontname\tentt}.}
40617 \centerline{{\it\fontname\tenit\/},}
40618 \centerline{{\bf\fontname\tenbf}, and}
40619 \centerline{{\sl\fontname\tensl\/}}
40620 \centerline{are used for emphasis.}\vfill}
40622 % Blame: doc@@cygnus.com, 1991.